Capitalism First or Socialism First? China and the Two Roads to the Mixed Economy.

I. The Debate Behind the Debate

I recently watched a Novara Media discussion in which Michael Walker and Aaron Bastani played back parts of Bastani's debate with Camilla Tominey at the Telegraph. The segment began with Tominey's challenge to Bastani over communism, China, worker exploitation and the political record of states that have called themselves communist. But the more interesting disagreement was not ultimately between Bastani and Tominey. It was the argument that developed between Walker and Bastani after the clips had finished doing their work.

Tominey supplied the initial pressure. If China can build roads, houses, electric cars, batteries and solar panels at extraordinary speed, how much of that development depends on low wages, dangerous factories, weak labour rights and the ability of the state to suppress opposition? When Bastani described China as a market economy but not a capitalist economy, she returned to the workers on the factory floor. Whatever category China belongs to, do those workers experience anything recognisably luxurious about the communism being defended?

Bastani's answer was not that Chinese industrialisation had been painless. His answer was historical. Britain also developed through appalling working conditions, and China had achieved an economic transformation that should be compared with other developing countries rather than with a country that had already accumulated several centuries of industrial advantage. The existence of exploitation did not, for him, settle the question of whether the overall process was historically progressive.

Walker then sharpened the problem. If the defence of China is that development has raised living standards so dramatically that the package is worth accepting despite its injustices, is that not also the strongest defence of capitalism? Capitalism has always been defended not because every workplace is fair or every owner virtuous, but because the system generated extraordinary increases in productive capacity, technology and material wealth. Walker's point was not that China and Britain are identical. It was that the moral structure of the defence looked surprisingly similar.

This is the part of the conversation I want to examine. I am not especially interested in declaring Bastani or Walker the winner, and Tominey's criticism should not be dismissed merely because it came from the Right. The exchange is interesting because each participant identifies something real, but the categories begin to slide into one another. Capitalism sometimes means markets, sometimes private ownership, sometimes exploitation and sometimes liberal government. Communism sometimes means a declared political destination, sometimes state ownership, sometimes the rule of a communist party and sometimes the completed abolition of class society. Development is treated as though it were one outcome, while exploitation is treated as though it were one injury.

The central question is therefore not whether China is really capitalist, socialist or communist. It is whether there is a material difference between two historical routes into a mixed economy. Britain developed through capitalism and later constructed democratic, trade-union and social constraints around it. China began from a state-socialist political order and later reintroduced markets, profit, private firms and foreign capital under the direction of a communist-led party-state. Both routes now contain mixtures of public and private ownership, markets and planning, profit and social provision. The important issue is not which label sounds more radical. It is how the mixture is organised, who directs it, who bears its costs and who has the power to correct it when it goes wrong.

That is a better question because it does not allow either capitalism or socialism to function as a moral incantation. It asks what the institutions actually do.

II. The Categories Have to Track Something

A Constraint First Material Ontological (CFMO) approach begins by refusing to treat political categories as explanations in themselves. Calling a system capitalist does not yet tell us how its state is organised, how strong its unions are, how investment is directed or how much of the social surplus is publicly controlled. Calling a system socialist does not tell us whether workers possess power, whether managers are accountable, whether commodities continue to circulate or whether the state has become an insulated hierarchy. A category is useful only where it identifies a materially reproduced relation, rules something out and changes what we should expect to observe.

To compare these systems, I need to separate legal ownership from effective control, and both from the mechanisms through which production is coordinated. Productive assets may formally belong to one institution while decisions over investment, output and closure rest somewhere else; prices, contracts, planning and administrative command may also operate together rather than as mutually exclusive alternatives. Accumulation adds the question of what compels reinvestment and expansion, while the organisation of labour concerns who governs the workplace and the working day, and control of the surplus concerns who decides how the product remaining after immediate reproduction is used.

These relations sit inside a wider structure of power. State institutions can impose collective decisions, but their practical character depends on which counter-powers can resist or correct them; the range of available choices is then narrowed further by trade, finance, technology, military competition and dependence on external resources. Treating these dimensions separately matters because legal ownership, practical control and political power do not necessarily sit in the same hands.

These dimensions overlap, but they are not the same. That distinction prevents several familiar shortcuts.

State ownership does not automatically create worker control. The presence of private firms does not automatically produce laissez-faire capitalism. The abolition of private capital does not prove that domination has disappeared. The state can direct capital without itself functioning as capital in the Marxian sense. A system can be non-capitalist while remaining exploitative, authoritarian and cruel. It can also be capitalist while substantially reducing death, injury and arbitrary power at work.

This is why I do not think the question can be resolved by choosing the correct noun. A society can legally socialise ownership while leaving control concentrated in a party bureaucracy. It can retain private ownership while socialising healthcare, education, infrastructure, labour security and a substantial share of investment. It can use markets without allowing capital to become sovereign, or abolish many markets while allowing the administrators of production to become a new centre of privilege and command.

The useful comparison is not a line from capitalism to socialism, but two interacting questions. How much capacity exists to direct capital towards collective ends, and how much effective power exists among workers and the public to constrain whoever is doing the directing? Nineteenth-century laissez-faire capitalism performs badly on both. Postwar social democracy strengthens counter-power while leaving much investment private. The Soviet model and contemporary China possess much stronger strategic direction, but weak independent worker power. Democratic socialism would have to combine both capacities.

The recurring question of this article is therefore simple:

Who can discipline capital, and who can discipline the institution doing the disciplining?

III. Capitalism Is Not the Same Thing as Laissez-Faire

Capitalism is often used so broadly that it becomes almost impossible to test. If every market, wage, commodity, profit and privately owned workshop is capitalism, then almost every complex economy becomes capitalist by definition. But if capitalism means only an idealised nineteenth-century system of weak regulation and private ownership, then the word cannot describe the heavily regulated, publicly financed and institutionally mixed economies that have existed for most of the last century.

For the present argument, capitalism is best understood as a system in which a substantial part of production is organised through commodity exchange, wage labour, monetary investment and the accumulation of capital, with enterprises significantly disciplined by profitability and competitive survival. The central process is not simply that rich people own things. Capital is value advanced through production and exchange in order to return enlarged. The capitalist class is not what creates this relation by possessing the wrong moral character. The relation reproduces the capitalist class because those who control investment, productive property and the purchase of labour-power occupy the position through which capital expands.

This matters when we later reach Tony Cliff. An elite does not become bourgeois merely because it commands production or extracts a surplus. Feudal lords, slave owners, imperial administrations and bureaucratic states have all controlled labour and appropriated output. Capitalism is a more specific structure than domination by an elite, even though domination is one of its possible effects.

Laissez-faire capitalism is a particular settlement inside that broader category. It is capitalism with relatively weak collective limits on investment, labour conditions, ownership and surplus allocation. Its strongest argument is not usually presented under the label trickle-down economics, which is more often a hostile description. The serious claim is that secure profits, investment freedom and capital accumulation expand production, employment, wages and consumption widely enough that the whole society benefits.

Postwar social democracy did not abolish capitalism. It changed the conditions under which capitalism operated. Organised labour increased workers' capacity to refuse. Full employment reduced the employer's power to discipline through unemployment. The welfare state socialised part of the reproduction of labour. Public ownership moved some essential systems outside ordinary private accumulation. Regulation made certain forms of injury and exploitation illegal. Taxation redirected part of the surplus after production, while collective bargaining altered its initial distribution.

Social democracy can therefore be defined as a capitalist economy in which democratic government, organised labour, public provision and regulation constrain accumulation and partially socialise its risks and returns. It remains capitalist because private accumulation continues to organise much of production and investment. But it is not merely laissez-faire capitalism with a larger benefits office. Its institutions alter power inside and outside the workplace.

Democratic socialism goes further. It does not merely regulate capital more humanely; it progressively democratises ownership, investment, workplace authority and surplus allocation. That does not require every private firm or market to disappear. My own Competitive Socialist Guilds model still allows capitalist initiative, private intellectual property, profit, contracts and market testing. What it attempts to remove is the assumption that initiating production gives capital sovereign command over labour. That model is not communism. It is a possible democratic-socialist mixed economy in which private capital remains, but is made to negotiate with worker institutions rather than isolated workers.

The broader socialist principle is that the reproduction of society should not be subordinate to private accumulation alone. Housing, health, education, care, infrastructure, energy, training, labour security and strategic investment are not peripheral corrections to the economy. They are part of the material conditions through which the economy continues to exist. A socialist system socialises enough of those conditions that private capital no longer possesses final authority over them.

State socialism is one institutional route through which this socialisation can occur. Rather than placing primary control in autonomous cooperatives, workers’ councils, guilds, municipalities or federated producer institutions, it uses the state as the principal mechanism through which productive assets, credit, strategic investment and surplus allocation are brought under collective direction. This does not require every market or private firm to disappear. Markets may be replaced in some areas, constrained in others and used instrumentally where they remain useful. What makes the system state socialist is that socialisation is mediated principally through state ownership and authority.

State socialism is not necessarily authoritarian, but it becomes authoritarian where the concentration of economic direction is matched by concentrated political power and workers lack effective institutions through which they can contest managers, alter priorities or constrain the state acting in their name. Such a system may achieve extensive legal and allocative socialisation while remaining weakly democratically socialised. Productive property may formally belong to society and its surplus may be directed towards public objectives, yet actual control remains concentrated in a political-administrative elite.

Communism introduces a further ambiguity because it can name both a process and an achieved condition. A communist movement or state may orient itself towards a future classless society while operating through institutions that are plainly not communist in the completed sense. A communist society, by contrast, is the proposed endpoint in which class domination has disappeared, wage and commodity dependence have substantially receded, productive property no longer gives one group command over another, and coercive state authority has withered or been radically transformed.

China can therefore remain communist as a political project while its present institutional form is better described as an authoritarian state-socialist party-state governing a state-directed mixed economy. These descriptions answer different questions. “Communist” identifies the declared and possibly real direction of travel; “authoritarian state socialist” identifies the mechanism through which strategic ownership and economic direction are presently organised; and “mixed economy” recognises that public ownership, state-capitalist arrangements and ordinary private capitalist production coexist within that structure. The contradiction only appears if political project, present institutional form and proposed endpoint are collapsed into the same word.

The real question is whether the present arrangement is actually producing institutions that make further movement possible. A destination written into a party constitution is evidence of declared intent, but it cannot prove material direction. The Communist Party of China still formally describes the realisation of communism as its highest ideal and ultimate goal. The CFMO question is whether the structure of the economy is progressively transferring power over production, surplus and administration towards workers and society, or whether the party-state, managerial hierarchy and capitalist sectors are becoming a permanent settlement. The declaration matters, but the direction matters more.

IV. Three Ways Workers Can Lose

The word exploitation caused much of the confusion in the Novara exchange because at least three different relations were being discussed.

The first is Marxian exploitation. Workers produce a surplus beyond what is required for the reproduction of their labour-power, while another class or institution controls that surplus. This is a structural concept. A worker can be well paid, physically safe, respected by management and protected by law while remaining exploited in this technical sense. The question is not whether the employer is unusually cruel, but who controls what collective labour produces beyond its immediate reproduction.

The recipient need not be an individual capitalist. A state enterprise can extract and allocate a surplus. A public administration can direct more labour towards accumulation than workers would choose themselves. Whether this remains exploitation depends partly on whether workers possess meaningful control over the institutions making that allocation. Formal public ownership does not settle the question because the public is not a person capable of exercising control without institutions.

The second form is abusive or gross exploitation. This is what Tominey was mainly pointing towards when she referred to sweatshops and abused workers. Workers' vulnerability is used to impose avoidable danger, excessive hours, wage theft, exhaustion, insecurity or preventable injury for the sake of profit margins, output targets or administrative convenience. Two workplaces can both produce a surplus while differing radically in this respect. One may operate under negotiated hours, strong safety law and meaningful compensation. The other may consume workers' health because replacing them is cheaper than changing the production process.

The third is labour domination or the deprivation of worker autonomy. Workers may be safe and reasonably paid while possessing almost no meaningful authority over their work. They may be unable to choose representatives, challenge managers, influence hours, refuse unreasonable instructions, contest production targets or participate in deciding how the surplus is used. This can happen in a private company, a public corporation, a cooperative captured by its managers or an authoritarian socialist state.

This third dimension is especially important because it prevents socialism from collapsing into benevolent administration. A state can provide housing, healthcare, stable wages and safe factories while workers remain objects of production rather than participants in its governance. That is better than a sweatshop, but it is not the same thing as worker emancipation.

The three dimensions should therefore remain separate. Social democracy can substantially reduce abusive exploitation and partially reduce labour domination while leaving Marxian exploitation intact. Authoritarian state socialism can eliminate private capitalist ownership, perhaps reduce some forms of abuse, and still leave workers dominated by managers and the state. A democratic socialist system would have to address all three, though it might not abolish every technical form of surplus extraction where private firms remain.

This also clarifies why pointing to British exploitation does not answer Tominey. It exposes a possible hypocrisy, but the existence of exploitation in Britain does not make Chinese exploitation unreal. Nor does China's extraordinary reduction of poverty show that the terms on which workers produced that development were the only possible ones. Absolute improvement and domination can coexist.

The sharper criticism of China is therefore not simply that development treated workers badly. It is that present workers had limited power to determine the sacrifices imposed in their name. The party-state did not merely administer production. It also decided which sacrifices were necessary, how long they should continue, which resistance was legitimate and how much present welfare could be deferred for future development.

That problem does not disappear when the promised future is sincere.

V. Development and the Soak Hose

Development is often treated as though it were one variable. A country either develops or does not, and once the answer is yes the argument appears to be over. But development includes several capacities that can move in different directions: productive capacity, infrastructure, material welfare, public health, labour reproduction, institutional competence, worker autonomy and ecological durability. A country can build excellent trains while exhausting workers, raise household consumption while degrading the environment, or create enormous technological capacity while concentrating political authority.

China has plainly developed. The relevant argument is not whether its industrial, infrastructural and material transformation occurred, but what kind of development it produced, who bore the costs and who gained control over the resulting capacity. Bastani is right to reject any account in which Chinese growth becomes morally invisible because the system is not liberal. Tominey is right that aggregate development does not answer what happened to particular workers inside the process. Walker is right that the same difficulty appears in the historical defence of capitalism.

The difference between the models can be illustrated through a garden hose.

Laissez-faire capitalism resembles an ordinary hose carrying water towards the destination selected by its owner. Some water may leak through imperfect fittings and feed plants along the route. Firms employ workers, buy from suppliers, spread skills, pay taxes and sometimes make goods cheaper. Those benefits can be substantial, but they are not the immediate organising purpose of the pipe. From the owner's point of view, leakage may be a cost. Tightening the fittings increases the quantity reaching the intended destination: private accumulation.

This does not mean capitalists want workers to starve or society to collapse. Capitalism as a whole requires healthy workers, infrastructure, education, law and social stability. But individual firms can benefit by avoiding their share of those costs. They can pay less, train less, externalise illness, rely on public infrastructure and leave families or the state to absorb the reproduction of labour. The system needs the plants watered, while each owner has an incentive to tighten their own section of hose.

Social democracy leaves much of the hose privately owned but attaches taps, regulators and redistribution mechanisms. Unions prevent all labour costs being squeezed out of the flow. Taxation redirects part of the water. Public services ensure that some plants receive what they need without relying on the owner's discretion. Regulation limits the pressure that can be imposed on workers and the environment. The system remains capitalist, but the leakage is no longer merely accidental. Outside institutions force part of the flow towards social reproduction.

China's developmental claim is different. It resembles a soak hose deliberately perforated throughout. Accumulation still supplies the pressure, and private firms, markets, profit and wage labour remain inside the system, but the party-state attempts to design where the benefits emerge. State finance, infrastructure, industrial policy, public enterprises, education, transport and regional development are intended to make productive growth release material gains across the economy rather than leaving diffusion to the private decisions of capital owners.

This is not ordinary trickle-down economics. China attempts to design the trickle.

The metaphor is useful precisely because it does not settle whether the result is socialist. A soak hose can be badly designed. Some sections can receive far more water than others. Pressure can be retained on the claim that future development requires it. Strategic industries, property owners, urban regions, party institutions or politically connected firms may receive priority while migrant workers and poorer households carry a disproportionate burden.

More importantly, the plants remain passive in the metaphor. They receive water but do not control the irrigation system. That is the difference between developmental provision and democratic socialism. A state can deliberately raise living standards while leaving workers with little power over the rate, destination and cost of the flow.

The decisive questions are therefore who controls the holes, who sets the pressure and who verifies the claim that more water must be retained for future use. China may distribute the gains of accumulation more intentionally than laissez-faire capitalism, but workers remain largely recipients of the flow rather than co-governors of the system.

This leads directly to the problem of sacrifice. Every economy must divide output between present consumption, maintenance, defence, infrastructure and future productive capacity. Some sacrifice is unavoidable because a society that consumes everything now will not reproduce the means through which future consumption becomes possible. The problem is not that a developmental state asks people to defer anything. It is that the institution demanding the sacrifice may also possess exclusive authority to decide its necessity, scale, distribution and duration.

The promised communist future intensifies this problem. If communism is difficult, distant or perhaps unattainable, temporary sacrifice has no natural stopping point. Failure can always be explained by insufficient development, hostile international conditions or inadequate discipline. The future ideal then protects the present institution from falsification. Existing workers are asked to suffer for an abstract future worker whose interests are defined by the state.

A credible transition requires a stopping rule. Rising productivity should eventually appear as safer work, shorter hours, greater material security, more control over production and a reduced need for coercive administration. Sacrifice is only meaningfully transitional where the development it enables progressively enlarges workers' power to decide whether further sacrifice is necessary. Otherwise the road becomes the destination.

A democratic socialist system would therefore not choose between the social-democratic network of hoses, connectors and taps and the developmental soak hose. It would incorporate soak hoses into that wider network, using deliberate public direction where coordinated development is required while preserving multiple channels of provision, local adaptation and institutions through which workers and communities can alter the flow. The precise configuration could not be universal: geography, settlement patterns, productive structure, social institutions and available resources would determine how much should be organised through national planning, municipalities, guilds, cooperatives, public enterprises or regulated private firms. Democratic socialism is not one ideal plumbing diagram. It is a system in which the network is deliberately designed for social reproduction and remains answerable to the people whose lives depend on it.

VI. Two Roads Into the Mixed Economy

This brings the Walker-Bastani disagreement into focus. Britain and China did not merely arrive at similar mixtures through different rhetoric. They followed opposite historical sequences, and those sequences left them with different concentrations of power.

Britain followed the route of capitalism first and social constraint later. Private capital developed substantial autonomy through industrialisation. The working class then organised against the conditions created by that development, while democratic politics, public provision and state ownership imposed additional constraints. The postwar settlement did not abolish capitalism, but it changed the relation between labour, capital and the state.

This is why the postwar comparison matters more than a simple contrast between nineteenth-century Britain and contemporary China. Nineteenth-century Britain demonstrates that industrial development can occur through brutal labour discipline. Postwar Britain demonstrates that a capitalist economy can be made materially different when workers possess organisations capable of imposing costs on employers and governments. Full employment, unions, public healthcare, social housing, pensions and regulation did not end the capital relation, but they reduced workers' direct dependence on the private employer and constrained gross exploitation.

That settlement was also historically reversible. Capital retained control over much investment, while labour's counter-power depended on institutions that could be weakened. Thatcherism did not need to abolish British democracy to change the balance. It broke organised labour, privatised public assets, liberalised finance and reconstructed the economy around a different relationship between state, capital and work. The formal constitution survived while the material distribution of power changed.

Bastani argues that the geopolitical pressure of the Soviet Union helped force Western capitalism to share productivity gains and maintain legitimacy. That mechanism is plausible, but it is not sufficient on its own. Domestic trade union strength, memories of interwar unemployment, mass democracy, wartime mobilisation, capital controls, high productivity growth and organised socialist politics also mattered. The Soviet challenge may have altered the external environment, but it did not build the NHS or bargain over wages by itself. National institutions transmitted and transformed the pressure.

China followed the opposite sequence. Political power and state ownership came first. Markets, foreign investment, private firms, profit incentives and substantial private accumulation were later reintroduced under the continuing supremacy of the Communist Party. Rather than allowing private capital to become politically autonomous and then building constraints around it, the party-state attempted to retain the capacity to define the strategic boundaries within which capital operates.

This is a real difference. Chinese private capital is not politically situated in the same way as British capital. The party-state can redirect credit, restructure sectors, prioritise infrastructure and discipline individual billionaires in ways that would be far harder for a liberal state whose political economy depends on autonomous private investment. Capital is powerful in China, but its power is mediated through a party that claims authority over the direction of national development.

The corresponding weakness is that labour does not possess an equally autonomous institutional position. In the British social-democratic route, capital, the state and organised labour were differentiated enough to constrain one another, however unequally. In China, the party-state directs capital, authorises labour organisations, defines legitimate political opposition and regulates the disputes produced by development. The same hierarchy that claims to act for workers also controls the forms through which workers can challenge it.

Britain therefore developed stronger traditions of constraining authority from below but weaker capacities for directing capital from above. China developed much stronger capacities for directing capital from above but weaker institutions for constraining the state and management from below.

Neither route has an inherent guarantee. Britain can continue moving towards a rentier capitalism in which democratic forms survive while effective power over investment, housing and work becomes increasingly concentrated. China can deepen social ownership and widen worker power, or it can consolidate a party-managerial-capitalist bloc that directs development without ever surrendering authority to the workers in whose name it rules.

The two routes solve different halves of the problem.

VII. State Capitalism: One Phrase, Several Theories

The phrase "state capitalism" appears to offer a solution because it combines the two visible features of China: a powerful state and extensive capitalist activity. But the phrase carries several incompatible theories, and using it without qualification can make the terminology decide the argument in advance.

The most useful meaning for the present discussion comes from Lenin's treatment of the New Economic Policy (NEP). Lenin used state capitalism for identifiable capitalist arrangements operating inside a state-directed transition: foreign concessions, mixed companies, controlled trade and private profit under conditions set by the Soviet state. The point was not that the whole Soviet economy had become capitalist. Capitalist forms could exist as bounded components inside a wider socialist project.

This produces an important distinction:

An economy can contain state-capitalist elements without the whole social order being state capitalist.

On Lenin's definition, a private or foreign capitalist may invest, employ labour and make a profit, while the state determines the terms, retains strategic ownership or limits where the arrangement can operate. State capitalism is therefore a relationship between capitalist activity and state control, not necessarily a diagnosis of the entire mode of production.

This is close to the strongest defence of Chinese reform. Capitalist mechanisms can be used instrumentally to attract technology, organise production, test demand, expand trade and accelerate capital formation, while the state retains command over strategic sectors and long-term development. The existence of profit does not automatically make the project anti-communist. It raises the further question of whether capitalist forms remain bounded instruments or become self-reproducing centres of power.

Unlike some later Trotskyists, Trotsky retained this more limited meaning of state capitalism and rejected its application to the Soviet economy as a whole. His classification was the degenerated workers' state. Capitalist private property had been abolished, the major productive assets remained nationalised, and the bureaucracy did not own the economic system as ordinary private, alienable capital. Yet the bureaucracy had politically expropriated the working class, strangled the soviets, unions and party democracy, and developed extensive privilege.

I think Trotsky is partly right on the narrow classificatory question. The Soviet Union was not capitalist in the ordinary Marxian sense merely because an elite controlled production. It was not principally organised through independently owned capitals advancing money in pursuit of private monetary expansion. The bureaucracy's access to the social surplus depended heavily on office, hierarchy and administrative allocation rather than ordinary capitalist title.

But that conclusion tells us only what the USSR was not. It does not make the system humane, democratic or substantively controlled by workers. Trotsky continued to call the proletariat the ruling class after acknowledging that it had been politically expropriated. The property structure carries too much weight in that classification. If workers cannot independently organise, replace decision-makers, control the surplus or contest the sacrifices imposed upon them, the mechanism through which they remain the ruling class becomes increasingly unclear.

Trotsky did not simply ignore democracy. He defended a restoration of Soviet democracy and a political revolution against the bureaucracy. But he treated democracy primarily as the political form through which an already proletarian property structure should properly express itself, while international revolution remained the ultimate condition for reversing bureaucratic degeneration. My disagreement is that effective worker power cannot be treated as a desirable political correction to an otherwise proletarian system. It is part of what makes the property social in the first place.

Stalin's later political economy extends parts of Lenin's argument. In Economic Problems of Socialism in the USSR, he insists that economic laws remain objective under socialism, that commodity production can persist and that the law of value continues to operate where commodities exist. There is little in that narrow philosophical position that I find objectionable. A change in legal ownership does not abolish scarcity, information, labour allocation, accounting or the need to generate and distribute a surplus.

The political contradiction is more severe. Stalin recognises that economic laws cannot simply be abolished by will, yet governs through a system with exceptionally weak mechanisms for correcting the will of the governing apparatus. If information is incomplete and objective constraints persist, concentrating political and economic authority inside an institution that largely regulates itself is especially dangerous.

Stalin also did not claim that complete communism had already been achieved, as some people state. He claimed that socialism had been established and treated complete communist society as a future stage. A CFMO classification can therefore say that the USSR was an authoritarian state-socialist system presented as an achieved socialist stage inside a continuing communist project. Whether it was materially moving towards communism is a different question from whether its leadership declared that destination.

Tony Cliff rejects Trotsky's category. His strongest initial move is that workers cannot collectively own production through a state they do not politically control. That is correct. State ownership is not worker ownership where actual workers possess no means of directing or contesting the state.

But the negative conclusion does not establish the positive one. From the proposition that the USSR was not substantively a workers' state, it does not follow that it was capitalist. Capitalism and worker-controlled socialism do not exhaust every possible form. The alternatives include authoritarian state socialism, bureaucratic collectivism, a non-capitalist command economy or a novel exploitative system.

Cliff understands that he needs more than the absence of worker power. His positive theory argues that Soviet workers were separated from production, the bureaucracy extracted a surplus, accumulation dominated consumption, and international military and economic competition compelled the Soviet state to behave as a collective capitalist. The USSR functioned as one enormous national capital within a capitalist world system.

This is a serious argument, but its load-bearing step remains doubtful. Geopolitical competition can force any state to raise productivity, build heavy industry, extract an investible surplus and prioritise defence. Those pressures make Soviet accumulation comparable to capitalist accumulation in some respects. They do not by themselves prove that production is governed by capital as self-expanding value.

A state that accumulates steel, machinery, infrastructure and military capacity may be expanding use-values and national power under external pressure rather than pursuing the monetary valorisation expressed in the movement M-C-M'. Forced accumulation, elite privilege and worker subordination are broader categories than capitalism. Cliff is convincing when he says the Soviet Union was not ruled by workers. He is less convincing when he treats that failure as evidence for capitalism rather than the beginning of a separate classification problem.

The clearer provisional category is authoritarian state socialism: a system with extensive legal and allocative socialisation, but weak democratic socialisation. The state formally owns production and directs the surplus towards collective or state-defined purposes, but workers and citizens do not possess effective control over the institutions making those decisions.

This distinction prevents the removal of capitalism from becoming a moral absolution. The USSR abolished private capitalist ownership without abolishing domination, elite privilege, dangerous accumulation or the crushing of workers. A non-capitalist society can be despotic; it does not need to be re-described as capitalist before it can be condemned.

Marxism becomes religified when capitalism is treated as original sin, revolution as redemption, the party as guardian of historical truth, sacrifice as purification and communism as deferred salvation. The materialist question is instead what made capitalism objectionable and whether those mechanisms were overcome. Replacing private command with administrative command, private appropriation with bureaucratic privilege and market insecurity with political subordination changes the mechanism without necessarily delivering emancipation.

State capitalism must therefore be specified by level. A state-capitalist element is a capitalist arrangement operating under state terms; a state-capitalist sector is a substantial group of them; state-directed accumulation describes strategic government direction of public and private capital; and Cliffite bureaucratic state capitalism is the stronger claim that the state itself performs the systemic function of capital.

China clearly contains the first three. The fourth remains an argument to be demonstrated, not a label to be assumed.

VIII. The Petty Bourgeoisie Is Not the Main Enemy

Lenin's state capitalism is clearer than Cliff's because it describes a real economic arrangement rather than converting every form of accumulation into capitalism. But I do not share Lenin's hostility towards the petty bourgeoisie as a social layer that must eventually be overcome or destroyed.

Small private producers can perform useful functions inside a highly socialised economy. They can identify unmet local demand, provide specialist services, experiment at low social cost, respond more quickly than large administrations and patch gaps left by public, cooperative or larger private institutions. Their profit may be the reward for noticing something that the wider system failed to see.

The presence of such producers is not evidence that socialism has failed. It may be part of how a complex system adapts.

A sufficiently good socialised economy should not need to eliminate small private employers merely to retain workers. If public enterprises, cooperatives or worker-governed institutions offer good wages, strong conditions, stability, autonomy and meaningful work, private employers must compete against that standard. They may attract workers by paying more, organising work better, offering a different kind of activity or taking a risk that the public sector would not take. If the socialised sector consistently loses workers because its conditions are worse, the first question should not be how to close the escape route. It should be whether the institution deserves to survive in its existing form.

This does not make private capital harmless. A small firm can grow into a monopoly, become central to social reproduction, dominate suppliers, acquire control over essential infrastructure or translate economic size into political sovereignty. But the answer need not be pre-emptive destruction. Once a private enterprise becomes a natural monopoly, systemically important or indispensable to social reproduction, the state can compulsorily purchase it on fair or generous terms and integrate it into the socialised system.

The people who built it can receive a very comfortable retirement. Their achievement can be rewarded without granting them permanent hereditary sovereignty over something the wider society can no longer safely leave under private command.

The principle is straightforward:

Reward the creation of socially useful institutions, but do not confuse that reward with an unlimited right to govern them forever.

This is consistent with the wider argument developed in my Competitive Socialist Guilds model. Markets and private initiative can remain as adaptive and informational mechanisms while labour reproduction, essential infrastructure and monopoly power are socialised. The issue is not the existence of private capital. It is whether capital becomes sovereign.

IX. Worker Power Cannot Be Replaced by a Conception of the Worker

The Soviet dispute becomes morally serious at the point where the party's conception of workers' interests replaces the workers themselves.

Trotsky writes that the dictatorship of a class does not mean that the whole mass of that class always participates in managing the state. That is obviously true. No complex society can require every worker to administer every railway, hospital, factory, budget and planning office personally. The alternative to bureaucracy cannot be universal permanent attendance at one enormous meeting.

But that does not answer the real objection. The relevant distinction is between delegated administration that remains answerable to workers and administration over workers justified by claiming to know their objective historical interests.

The process of representational substitution is easy to see. The party represents the working class. The central leadership represents the party. The leadership then defines the long-run interests of the working class. Actual workers can be overruled because their present preferences are said to conflict with industrialisation, the survival of the revolution, the international proletariat or the future communist society.

The leadership may sincerely believe this. The problem does not require personal corruption or secret capitalist intent. A small group can govern despotically while possessing a moral conception of the worker and believing that its own power is necessary for workers' eventual emancipation.

Anton Pannekoek provides a stronger criterion. Social ownership is not achieved merely because a state or party governs production on behalf of workers. Workers require institutions through which they exercise power over production themselves. Pannekoek's own preferred form was the workers' council, but the important point for this article is functional rather than prescriptive. There must be real mechanisms through which workers can influence, contest and constrain decisions, and those mechanisms must be capable of imposing material costs on managers or political authorities that disregard them.

I have my own institutional preference in Competitive Socialist Guilds, but that is not the subject here. Worker power could be organised through councils, unions, guilds, representative bodies, public boards or forms not yet developed. The name matters less than the capacity. Can workers obtain independent information? Can they alter decisions rather than merely complain? Can they organise without the permission of the institution they may need to challenge? Can they remove, bypass or constrain decision-makers? Can they demonstrate that a claimed necessity is false and force policy to change?

If not, worker rule has become a statement about historical purpose rather than a description of material power.

This is where Trotsky's treatment of democracy becomes unstable. He is right that liberal parliamentary democracy is not class-neutral and can protect capitalist domination. But democracy is not exhausted by Western representative government. It also names the wider family of mechanisms through which people subjected to collective power can influence, contest, constrain or replace those exercising it.

Workers' councils, independent unions, recallable delegates, legal rights, workplace representation and competing sources of information are not suprasocial abstractions. They are material feedback mechanisms. They allow the governed to reveal information that administrators would prefer not to hear, impose costs on insulated decision-makers and correct policies that have been rationalised as historically necessary.

The absence of bourgeois democracy does not prove the presence of proletarian power. It may only prove that power has fewer checks.

This is also why the suppression of autonomous worker power cannot be treated as a problem invented entirely by Stalin. Pannekoek's criticism of Bolshevism was that party leadership displaced workers' self-government and converted councils from institutions of worker power into instruments subordinated to state and party command. Stalin transformed that concentration into a far more murderous and despotic system, but the institutional conflict between party administration and autonomous worker control did not begin with him.

The point is not to claim that Lenin secretly intended everything Stalin later did. It is that a system which weakens the independent institutions capable of checking the party becomes increasingly dependent on the party's own moral quality. That is an intolerable constitutional gamble, especially where the same institution controls production, political legitimacy, policing, information and the definition of the future good.

A small elite can then accumulate privileged access to housing, goods, services, appointments and authority without owning the economy as private capital. The surplus becomes politically accumulated rather than privately alienated. The society may no longer be capitalist, but the workers can still be ruled by an unchecked elite whose values and priorities become the values and priorities of the whole system.

This is the deeper reason I resist making capitalism itself the ultimate evil. Capitalism is objectionable because of what it does: it subordinates labour to accumulation, gives owners command over socially necessary assets, converts economic power into political power, externalises the reproduction of labour and allows competition to resolve itself through insecurity and harm. If a non-capitalist state reproduces domination, privilege, forced accumulation and worker powerlessness through administrative mechanisms, the moral problem has not disappeared. It has changed form.

X. From the Critique of Capital to Anti-Capitalism

Cliff’s theory matters beyond the classification of the Soviet Union. By describing the USSR as capitalist because workers had lost control and the state was compelled to accumulate through international competition, he widened capitalism from a specific relation of production into something closer to a general category of modern domination. Cliff did not make this move through moral assertion alone: he attempted to ground it in the law of value, world competition and the bureaucracy’s function as collective capital. But the conceptual result was to preserve an exhaustive binary. If workers did not consciously govern production, then capital must still rule, even where private capital had largely disappeared.

This creates one route through which Marxism can become a political theology of anti-capitalism. Capitalism ceases to name a historically specific system with both productive capacities and destructive relations. It becomes the name of the evil itself. Anti-capitalism then becomes evidence of virtue, even where the force opposing capitalism offers no worker control, no institutional accountability and no credible method of coordinating production. The polar opposite of capitalism is presumed to be socialist before anyone has shown what the alternative materially is.

This should not be placed entirely at Trotsky’s door, nor does it reduce Cliff’s own work to a moral slogan. Trotsky could analyse capitalist and socialist tendencies coexisting within the NEP and ask which was gaining material weight. Cliff attempted a political economy of Stalinism and insisted that socialism required worker self-emancipation. The criticism concerns what the resulting framework made politically easy: once capitalism became the universal explanation for modern exploitation, “anti-capitalist” could function as a political identity even after its positive institutional content had become uncertain.

CFMO reverses this movement. It does not ask whether a system stands on the opposite side of capitalism. It asks how labour is organised, who controls the surplus, how production is coordinated, which powers remain contestable and whether the system reproduces social life without crushing those who sustain it. A private enterprise may perform a useful social function, while a non-capitalist state may become despotic. The question is not which side of a sacred divide an institution occupies, but which mechanisms it contains and what consequences those mechanisms produce.

XI. Communism as a Process and Authoritarian State Socialism as a Present Form

The distinction between communism as a project and communism as an achieved society is necessary if China is to be discussed without semantic games.

A state can be communist in political orientation where it treats its present institutions as transitional and continues to pursue a society beyond capitalist class rule. It can simultaneously be authoritarian state socialist in its actual form, because strategic production is directed by a party-state while workers possess weak autonomous power over that direction.

These descriptions do not contradict one another. One describes the claimed trajectory. The other describes the present arrangement.

China's Communist Party continues formally to declare communism as its ultimate goal. That makes China a communist project at the level of official political purpose. But CFMO cannot infer a trajectory solely from a declaration, because the real intention of a large institution is difficult to observe and because transitional language can survive long after a transition has stopped.

The material test is whether authoritarian state socialism is creating the conditions for its own displacement. Does development progressively increase worker power, reduce elite insulation, shorten working time, socialise strategic investment and make coercive administration less necessary? Or does it reproduce party monopoly, managerial hierarchy, private accumulation and worker subordination as a stable equilibrium?

If the former, authoritarian state socialism may be a compromised stage in a continuing communist process. If the latter, communism has become a historical identity and legitimating language rather than a material direction.

This distinction also avoids the claim that a system ceases to be communist merely because the endpoint may be impossible. A movement can pursue an unattainable goal. The classification concerns what it is trying to do, not whether history guarantees success. But an unreachable goal creates an even stronger obligation to ensure that present workers are not sacrificed indefinitely for a destination that never arrives.

Authoritarian state socialism can therefore be defended, at most, as a constrained and imperfect transition. If its concentrated authority is progressively displaced by worker and social control, the authoritarian features may genuinely be temporary. If the party-state, managerial hierarchy and political subordination of labour become the permanent settlement, it is no longer a road to communism. It is simply authoritarian state socialism - or, to use its unfortunately accurate abbreviation once, ASS forever.

XII. Internationalism Cannot Coordinate What Has Not Been Built

Trotsky's strongest argument remains that no socialist state exists outside the world economy. Trade, finance, technology, resources, military competition and hostile political powers constrain what can be achieved inside one country. An isolated revolutionary state may face scarcity, invasion, sanctions, technological weakness and pressure to subordinate consumption to defence and accumulation.

The error lies in converting that constraint into a binary law of motion. Trotsky argues that if revolution does not expand internationally along a proletarian spiral, it must contract bureaucratically within the national framework. He identifies a genuine danger, but the word must does more work than the mechanism establishes.

A more defensible claim is that isolation, scarcity, external hostility and weak worker institutions increase the risk of bureaucratic degeneration. They do not prove that the only alternatives are worldwide revolution or internal collapse. Between those poles lie mixed economies, partial socialisation, regional cooperation, temporary retreats, market reforms, democratic corrections and durable but incomplete settlements.

Marx's discussion of national differences in wages adds a useful dog-ear to any theory that gives automatic causal supremacy to the global level. Marx describes different national averages of labour intensity forming a scale measured against universal labour. The universal relation does not abolish the national differences. It is constituted through their comparison.

National economies reproduce different levels of skill, health, education, housing, infrastructure, customary consumption, labour law and collective power. Those differences affect the cost of reproducing labour-power, the productivity of labour and the distribution of necessary and surplus labour. The world market compares national labour regimes; it does not make them identical.

The global therefore does not simply trump the national. National institutions produce the labour forces, infrastructures, currencies, legal orders and productive capacities through which the global economy exists, while international pressure acts back upon them. The relation is recursive rather than hierarchical.

Nation and state should also be kept separate. The state is the administrative, legal and coercive apparatus governing a territory. A nation is a historically reproduced cultural and institutional system through which large populations sustain shared expectations and coordination. Neither requires ethnic homogeneity, and recognising national institutions does not imply hostility to migration. The point is that many mechanisms of labour reproduction, welfare, law and political legitimacy remain organised at this scale.

This does not prove that communism can be completed in one country. It means that states can produce materially different settlements inside the same world economy. Postwar and Thatcherite Britain inhabited broadly the same international order, yet organised labour, welfare and investment very differently.

The strategic consequence is equally important. Even if international coordination is required for the long-run stability of socialism, it does not follow that international revolution has immediate causal priority. An isolated state that aggressively attempts to export revolution may place its head above the parapet before it possesses the capacity to survive the response. It may provoke containment, sanctions, military pressure, overextension and internal securitisation. The attempt to escape isolation can intensify the conditions that produce authoritarianism.

More fundamentally, international socialism cannot be created by proclamation. You cannot coordinate institutions that do not yet exist.

A large meeting of Trotskyist organisations may produce declarations and historical purpose, but it does not produce an international socialist economy. A federation of weak organisations remains weak because coordination can amplify capacity, not manufacture a productive and administrative substrate from nothing. Durable internationalism requires functioning public enterprises, unions, cooperative institutions, legitimacy and productive capacity. The likely sequence runs from viable institutions, to demonstrated success, to replication, cooperation and deeper federation.

This is not an argument for ignoring the world. National consolidation can become an excuse for permanent emergency power and elite entrenchment. It is an argument about sequence: the international may be the eventual scale of stability, but it cannot substitute for constructing the first viable components.

China's post-Mao strategy can be read as a pragmatic answer to Trotsky's binary. It retained party control, opened to world trade, admitted foreign and private capital, and used state direction to build national productive capacity. That strategy has been far more successful as a method of development and accommodation to the world market than an attempt to export immediate revolution would likely have been.

It does not prove that China is reaching communism. It proves that compromise with the world market can produce a durable and transformative settlement. The remaining question is whether the compromise is still a route, or whether it has become the destination.

XIII. Returning to Walker and Bastani

Walker's challenge to Bastani was therefore correct but incomplete. The strongest defence of Chinese development does resemble the strongest defence of capitalism because both can ask workers to tolerate injustice as the price of future abundance. But the institutional routes are not the same, and that difference affects what each system can correct.

Tominey’s challenge keeps the experience of actual workers in view: rapid development cannot excuse sweatshop conditions, preventable harm or political repression. But the force of that criticism is lost if contemporary British institutions are treated as a model China could simply have adopted at the beginning of industrialisation, or if the brutal history of Western capitalist development is forgotten. Bastani restores that developmental context by showing why state finance matters and why the presence of markets does not, by itself, determine the political character of a society. The unresolved question is the transition itself. Productive growth does not become socialist merely because a Communist Party directs it; that claim only gains substance where workers acquire increasing power over production, surplus and the sacrifices imposed in their name.

Walker is right that the package argument cannot be reserved for China. Capitalism also asks to be judged by its productive achievements rather than by every injustice generated along the way. But the comparison should not flatten the difference between benefits that emerge incidentally from private accumulation and benefits deliberately distributed through a developmental state.

Britain and China each possess part of a possible socialist solution and part of a serious failure.

Britain retains traditions of autonomous labour organisation, plural institutions, legal challenge, public accountability and democratic correction. Those traditions have been badly weakened, and formal democracy can coexist with a hollowed-out labour movement, private control of investment and an increasingly rentier economy. Britain could move towards democratic socialism by rebuilding worker power, public ownership and strategic direction. It could just as easily continue rightwards into a weakly democratic system where capital, property and finance dominate the decisions that matter.

China possesses strong state capacity, control over strategic finance, extensive public ownership, long-term planning and the ability to discipline individual capitalists. Those capacities could support a deeper socialisation of production and investment. They could also stabilise an authoritarian order in which party officials, managers and billionaires become an insulated bloc while workers remain politically subordinate.

Neither route contains a hidden guarantee. Britain is not destined to become more democratic because it holds elections. China is not destined to become communist because it is governed by a communist party. Both systems can move left or right, though the mechanisms and language will differ.

The clearest test will come not from the distribution of gains during expansion, but from the allocation of losses when growth slows. Rapid development allows wages, profits, public revenue and household expectations to rise together. Slower growth forces harder choices. Who absorbs property losses? Who pays for debt and pensions? Are wages restrained to protect accumulation? Are elite assets protected while workers carry unemployment and declining services? Does labour gain more power over the adjustment, or is resistance suppressed in the name of stability?

China may be approximately as socialist as its leadership believed compatible with development and survival inside the world market. That is the strongest sympathetic interpretation. But constraints do not determine one unique policy, and the party-state cannot be the sole judge of whether every sacrifice was necessary. Without independent worker power, necessary compromise and politically convenient exploitation become difficult to distinguish.

The same warning applies to Britain from the opposite direction. Democratic forms do not prevent capital from gradually escaping social control. A state can preserve elections, courts and rights while losing the productive assets and organised labour required to make those institutions materially consequential. Counter-power without strategic capacity becomes defensive. Strategic capacity without counter-power becomes authoritarian.

A viable democratic socialism would need both. It would need enough public and social power to direct investment, infrastructure and essential production. It would need enough adaptive plurality to permit markets, small enterprise, experimentation and institutional failure without turning every mistake into a national crisis. And it would need real worker and public power capable of constraining both private capital and the institutions acting in society's name.

The aim is neither capital without social direction nor social direction without effective worker power.

IX. Conclusion

I found Walker and Bastani's conversation really intriguing. They really teased apart several different ideas and views on the topics discussed. This mane me think more carefully about what their disagreement—and the language used to conduct it—actually turns on. Once markets, party-state authority, public ownership and private capital coexist within the same system, calling China capitalist, socialist or communist does not by itself tell us how investment is directed, how development is distributed or what power workers possess over either process. Britain poses the same problem from the opposite direction: liberal-democratic institutions and public services coexist with extensive private control over investment, production and employment. What stayed with me was not a simple verdict on China, but the need to separate these relationships before deciding what any of the familiar labels really explain.

The important distinction is between two roads into a mixed economy. Britain allowed capitalism to develop substantial autonomy and later built institutions to constrain it. China retained political control and later reintroduced capitalist mechanisms as instruments of development. The first route created stronger traditions of counter-power but allowed capital to remain structurally sovereign. The second created stronger strategic direction but left workers weak against the directing state.

Lenin's limited state capitalism helps explain how capitalist activity can operate inside a communist project. Trotsky is right that the Soviet Union did not become capitalist merely because its bureaucracy dominated workers. Cliff is right that state ownership without worker control is not worker ownership. Pannekoek is right that workers cannot be replaced by a party's conception of their historical interests. Stalin is right, at the narrow economic level, that commodities, wages and objective constraints do not disappear by decree. None of these partial truths vindicates the political system that crushed the people it claimed to emancipate.

China's soak hose remains a better metaphor than ordinary trickle-down. The state attempts to make accumulation release development across the system rather than relying on accidental leakage. But a pipe designed from above is not yet socialism in the democratic sense. The question is still who controls the pressure, who chooses where the water goes and whether those carrying the cost can force the design to change.

Britain and China could each move towards a more socialised and humane settlement, or in the opposite direction. Britain can lose the substance of democratic counter-power while retaining its constitutional shell. China can retain public direction while an authoritarian party-managerial-capitalist bloc becomes permanent. The decisive question is whether workers gain or lose the capacity to govern the institutions on which their lives depend.

China's greatest socialist advantage is that the state can discipline capital. Its greatest socialist weakness is that workers cannot independently discipline the state.

Britain's corresponding problem is almost the reverse. It retains more routes through which power can be challenged, but has progressively surrendered the assets, institutions and labour strength required to direct capital towards collective ends.

A serious socialism has to solve both problems at once.

Sources and recordings for verification