Socialism

Socialism names both a family of political aspirations and a family of projects for transforming economic institutions. Socialist traditions have sought, in different combinations, greater welfare, equality, solidarity, freedom from exploitation and domination, meaningful work, workplace democracy, social control over productive wealth, and ecological sustainability. They have proposed worker ownership, cooperatives, public ownership, social wealth funds, planning, markets under social ownership, municipal institutions, councils, public banks, universal provision, and other arrangements as means to those ends.

The aspirations are not decoration; they are part of what identifies the tradition. Capitalist institutions can arise without participants intending to instantiate capitalism, although capitalism can later become a reflexive political project. Socialism has a more characteristically programmatic genealogy. Communal property, a public enterprise, worker participation, or redistribution does not become socialist merely because socialists might approve of it. Socialist institutions characteristically enter the relevant history through attempts to transform economic power in accordance with recognizably socialist aims. Global histories of socialism accordingly treat the tradition as a set of heterogeneous movements and projects rather than one institutional model diffusing unchanged from Europe (van der Linden 2022).

A useful analysis therefore follows an additional sequence:

aspiration → proposed institution → implemented institution → mechanism → outcome.

Each link can fail.

An attractive aspiration can motivate a poorly designed institution. A coherent proposal can be altered during political implementation. An implemented institution can operate differently from its formal design. Or the institution can function as designed while producing outcomes contrary to the value that motivated it.

A socialist formation can therefore fail by socialist standards while remaining evidence about socialist institutional design. If an attempted democratization of economic power generates administrative concentration, defining the result as not socialism because it is not democratic risks removing precisely the evidence relevant to feasibility.

Conversely, an outcome does not become a socialist success merely because socialists value it. Universal education, public health provision, low inequality, or worker voice occurring in a predominantly capitalist order initially provides evidence for those institutions, not for social ownership generally.

This entry distinguishes four levels.

First, aspirations: what outcomes does socialism seek?

Second, allocation mechanisms: which decisions should be made through markets, planning, bargaining, professions, commons, or other arrangements?

Third, ownership and governance: who receives residual returns, supplies capital, governs productive organizations, allocates investment, bears losses, and possesses residual authority?

Fourth, institutional interaction and system effects: how do these arrangements affect one another, generate feedback, or jointly enable capacities and vulnerabilities?

1. Identifying socialism

1.1 A global and programmatic tradition

Common ownership, redistribution, mutual aid, and communal production long predate socialism. Their presence alone therefore does not identify a socialist formation.

Modern socialism is distinguished partly by a reflexive project of transformation, but that project has taken different forms across Europe, Asia, Africa, Latin America, and the Middle East: early egalitarian traditions, state-centred and anti-statist socialisms, labour movements, anti-colonial projects, and distinct regional traditions across several continents (van der Linden 2022). Industrial class relations have been central in some of these contexts, while anti-colonial sovereignty, agrarian transformation, rapid development, national independence, or resistance to foreign economic domination have been central in others. Socialism should therefore not be identified with one European or Soviet institutional model.

1.2 Why socialization of economic power?

Socialism can be organized conceptually around social ownership, equality, cooperation, ethos, democratic social power, or other ideas; this entry organizes it around economic power.

This entry gives special weight to socialization of economic power, not because power is the only socialist value, but because socialist projects characteristically seek to alter some bundle of the powers described below.

The conception is especially close to Erik Olin Wright’s (2010) account of socialism in terms of social empowerment over economic activity: increasing social power rooted in collective association over the allocation and use of economic resources. The present formulation differs mainly in decomposing economic power more explicitly and in remaining neutral about whether civil society should be the privileged institutional locus of that power.

Legal ownership alone does not tell us who receives residual income, selects investments, governs firms, appoints managers, or can override whom.

Distribution alone does not tell us who exercises productive authority.

Worker control and citizen ownership can diverge.

Public ownership can coexist with either centralized or highly delegated decision-making.

Socialization is therefore understood as a project to transfer some bundle of productive powers away from a distinct class of private capital owners and toward workers, citizens, communities, public bodies, or other collectively constituted institutions.

The relevant powers include:

• residual income;
• appointment and governance rights;
• operational authority;
• investment authority;
• transfer or liquidation rights;
• residual authority where existing rules do not settle a dispute.

Its advantage is not that it matches ordinary usage better, but that it better tracks the distinctions that matter for explanation and comparison.

1.3 Ideal, project, formation, and feasibility

At the ideal level, socialism articulates values.

At the project level, actors propose institutional changes intended to realize them.

At the formation level, those institutions operate under actual informational, motivational, administrative, and political conditions.

This distinction places the present approach within an established socialist feasibility tradition. Nove (1983) asks what a feasible socialist economy would actually require rather than comparing capitalism with an institutionally unspecified ideal; Wright (2010) distinguishes desirable, viable, and achievable alternatives and treats transformation strategy as part of the analysis.

The present framework adds a more explicit causal translation problem. An institution can be viable in isolation yet be transformed during implementation; it can operate as designed while frustrating the aspiration; or its interaction with other institutions can generate an unanticipated equilibrium.

1.4 Social democracy and boundary cases

Social democracy illustrates why genealogy, institution, and formation must remain distinct.

A socialist political movement can govern a predominantly capitalist economy and build social insurance, public services, collective bargaining, or redistribution without thereby socializing most productive ownership. Przeworski’s analysis of social democracy centres precisely on the strategic and institutional problems faced by socialist parties operating within capitalist democracy (Przeworski 1985).

The Swedish Meidner wage-earner-fund proposal was more directly socialist in the institutional sense because it sought gradually to alter claims over corporate capital rather than merely redistribute current private income. The eventual funds were substantially diluted relative to the original proposal, and were later abolished; the episode is therefore especially informative about transition, coalition stability, and the political response of incumbent interests rather than the steady-state performance of a completed social-ownership system (Meidner 1978; Westerberg 2023).

Similar distinctions apply to social wealth funds, public pension funds, worker cooperatives, and state-owned enterprises. Each can instantiate some socialist institutional element without making the surrounding formation socialist.

These cases motivate an explicit layered rule. A socialist project is identified by genealogy: an organized attempt to transfer productive powers toward collectively constituted bodies in the name of recognizably socialist aims. A socialist institution is identified by structure: it actually vests some bundle of the powers listed in §1.2 in such bodies, whatever its sponsors intended. A socialist formation is identified by extent: enough of those powers, over enough of productive life, are socialized that the configuration itself, not merely an element within it, is organized around social control.

The three come apart. A formation with substantially socialized productive power but no socialist self-description (a developmental state holding commanding-heights industries, a corporatist order administering investment through peak associations) categorizes by its structure, with its genealogy recorded separately; its record bears on socialist institutional design exactly to the extent that the socialized powers drove the outcome. Conversely, a socialist movement whose implemented institutions socialize little, the ordinary social-democratic case, remains a socialist project governing a non-socialist formation, and its record is evidence about strategy and transition rather than about social ownership. Evidence eligibility follows the powers, not the label; genealogy determines what a case is evidence about only where intent shaped how the institution operated.

How much socialization makes a formation socialist rather than social-democratic, corporatist, or developmental-statist is a threshold question of the same kind as capitalism’s lower boundary: no single power suffices, and the test is whether socialized control materially structures investment, enterprise governance, and residual claims across a substantial part of productive life, or instead operates as an enclave within a configuration organized otherwise. The categories also genuinely overlap: a formation can satisfy capitalism’s structural test while socializing substantial powers, which is the hybrid space, and many actual formations live in it. The threshold question is therefore not where a sharp line falls but which powers, socialized to what extent, change which mechanisms.

1.5 Constitutive, complementary, background, and recruited institutions

Socialist formations can also recruit institutions not intrinsic to socialism.

State planning, one-party rule, particular bureaucratic hierarchies, nationalism, military organization, policing, education systems, media, or particular labour institutions can become integrated into a socialist formation without being definitionally socialist.

It is useful to distinguish:

• institutions constitutive of a particular project of socialization;
• complementary institutions affecting how socialized powers operate;
• background conditions such as administrative capacity, knowledge, ecology, and inherited productive resources;
• historically recruited institutions incorporated into specific formations.

Recruited institutions can include capitalist ones. A socialist formation can deliberately incorporate markets, private enterprise, or foreign capitalist finance, as the New Economic Policy and reform-era China both did, without the recruitment by itself dissolving the project; the categorization question then falls to the layered rule of §1.4.

This prevents both “everything a socialist state did was socialist” and “anything inconsistent with socialist ideals was external to socialism.”

1.6 Categorization and attribution are different tests

Two distinct questions run through this entry and its companion, Capitalism. Categorization asks what a formation is: which institutions are present, and whether their configuration makes it socialist, capitalist, hybrid, or neither. Attribution asks what a formation’s record shows: which mechanism produced an outcome, and how far the inference travels. A formation can be straightforwardly socialist while its record supports no conclusion about socialism in general, and an outcome can be well identified while the formation that produced it resists categorization.

Six questions order the analysis:

1. What institutions are present?

2. What makes the formation capitalist, socialist, hybrid, or neither?

3. What mechanism is being tested?

4. Is that mechanism constitutive of the project of socialization, complementary, background, or recruited (§1.5)?

5. What outcome does the evidence identify?

6. How far may the inference travel?

The worked inferences in §§6.6–6.7 apply questions 3 through 6 as a five-step evidential test followed by a scope restriction, stated here so that the application uses a rule announced in advance: identify the mechanism; show that it operated; connect it to the institutional structure at the scope claimed; test whether it resisted correction; assess robustness across variants; then restrict the verdict accordingly.

The pair’s definitions are stipulations, answerable to the explanatory work a category can do rather than to ordinary usage. Stipulation also makes room for field relativity: different explanatory projects may draw somewhat different boundaries around socialism and capitalism, so the labour historian’s socialism, a family of movements and projects, and the comparative economist’s socialism, a family of institutional formations, can overlap substantially in extension while remaining distinct categories, each answerable to what its field needs the category to explain. Nothing in what follows requires one of them to be the real one; it requires each claim to say which category, at what scope, it concerns.

The definitions of socialism and capitalism are deliberately different in kind. Capitalism is identified structurally, through recurrent accumulation, and can arise with no one intending it; socialist projects are identified partly through programmatic genealogy, while socialist institutions and formations are categorized by the productive powers actually socialized and their extent (§1.4). The asymmetry is both a historical claim (the traditions did enter history differently) and a stipulative choice, made because unintended socialization rarely organizes a formation the way unintended accumulation can. Nothing evidential hangs on it. Evidence eligibility is governed by questions 3 through 6, which treat favourable and adverse records of both families identically: a socialist formation that fails by socialist standards is admitted as evidence by the same rule that admits a capitalist formation nobody designed.

The layered distinctions of §§1.3–1.5, the functional-conservation test (§6.5), the six questions above, and the attribution rule (§6.8) are this entry’s framework, shared with the companion entry, rather than a summary of settled positions. The survey sections report established literatures.

2. Socialist aspirations

Capitalist, socialist, and hybrid systems should ultimately be evaluated in the same outcome space: productive capacity; welfare; security; equality; liberty and effective exit; workplace voice; meaningful work; freedom from domination; political pluralism; innovation and discovery; ecological sustainability; resilience and corrigibility.

Aspirations are not what distinguishes socialists as people: many defenders of capitalism hold broadly similar hopes for welfare, security, and meaningful work. The difference is architectural. Socialism builds aspirations into its own identification, while capitalism is identified without reference to anyone’s hopes, which is partly why capitalism can be made to look as if it does not care: there is no aspirational clause in its identification to care with. Socialism’s distinctive burden is therefore to explain which institutional transformations are supposed to realize its characteristic aspirations, and why those transformations rather than others.

2.1 Welfare, production, and flourishing

A welfare-based socialist argument must begin with productive capacity as well as distribution.

The relevant pipeline is:

inherited and external capabilities
→ discovery and innovation
→ investment and mobilization
→ production
→ diffusion and access
→ distribution and risk
→ welfare
→ subsequent productive capacity.

Socialist and social-democratic institutions can improve different links, and the modality of that “can” differs by link. Some contributions are demonstrated: social insurance, public services, public investment in weakly appropriable activities, and redistribution have documented welfare effects inside predominantly capitalist economies. Others are design potentials: that planning can coordinate known complementary investments, or that universal provision can improve access at acceptable cost, is established for some cases and proposed for others.

But welfare institutions depend on real productive resources.

A welfare state operating over a productive capitalist economy demonstrates the effects of social insurance, public services, and redistribution before it demonstrates the welfare effect of replacing private productive ownership.

The joint counterfactual matters: what happens when productive ownership, investment, distribution, and provision all change?

The converse is equally important. Capitalist production operates inside systems containing public education, health, infrastructure, scientific research, monetary institutions, and social insurance. Their productive contribution cannot be held fixed while all output is credited to private ownership.

Redistribution and social insurance also change behaviour. Optimal-tax theory treats distribution and behavioural response jointly rather than as a static transfer from one fixed income pool to another (Diamond and Saez 2011). Chetty’s analysis of unemployment insurance similarly distinguishes the welfare value of liquidity from behavioural changes often summarized as moral hazard (Chetty 2008).

2.2 Equality, incentives, and dynamic divergence

Equalizing assets or income is not the end of the causal story.

People differ in skills, health, preferences over work and leisure, willingness to defer consumption, risk tolerance, saving, family obligations, and luck. Some consume more now; others save. Some undertake costly education or risky investments; others prefer leisure, care, or current consumption. An initially equal distribution of productive resources can therefore become unequal without any subsequent act of coercion or formally unequal rule.

This problem sits near two familiar debates in distributive justice. Dworkin’s equality-of-resources framework seeks to distinguish inequalities connected to choices and gambles from those reflecting circumstances beyond agents’ control, while Nozick’s critique of patterned distribution emphasizes that voluntary transactions can disrupt an initially preferred pattern (Dworkin 1981; Nozick 1974).

The present argument does not presuppose either normative theory. The institutional question is dynamic:

Which inequalities produced by choice, luck, differential ability, inheritance, or institutional power are objectionable, and what happens when institutions continually correct them?

Repeated correction can change incentives to work, save, invest, migrate, insure, avoid taxes, or take risks. Failing to correct divergence can allow inequalities initially associated with voluntary choice or luck to become inherited differences in opportunity and bargaining power for people who made none of the original choices.

Socialism therefore does not escape incentive problems by moving redistribution upstream into ownership. It changes the incentive architecture.

2.3 Exploitation

Exploitation is a distinct socialist criticism rather than merely another word for inequality.

Marxian exploitation concerns relations through which capital owners appropriate surplus from productive activity (Marx 1867/1990). Roemer reconstructs exploitation in terms of unequal productive endowments and the feasible alternatives available to differently situated agents, rather than relying on the labour theory of value in its classical form (Roemer 1982).

Vrousalis instead understands capitalist exploitation as enrichment through domination: one party benefits through power over another’s productive activity (Vrousalis 2023).

These accounts point toward different remedies.

If unequal productive assets are the relevant problem, broader ownership or social funds may matter.

If domination is central, worker governance, outside options, or changes in residual authority may matter more.

If nonworker appropriation of surplus is itself the problem, worker ownership becomes more directly relevant.

But productive endowments extend beyond physical assets. People differ in skills and ability to work, and labour itself is a productive endowment. There is also an important asymmetry between labour and financial capital. Labour is attached to persons and ordinarily poorly diversifiable; financial capital can be alienated, accumulated, and diversified. Yet modern output usually depends on complementarities among human activity, accumulated knowledge, infrastructure, and capital goods.

A serious exploitation argument must therefore identify the wrong-making relation rather than assuming that one productive contribution alone creates all entitlement to the surplus.

2.4 Economic democracy and worker-specific claims

Worker governance has a stronger justification than the claim that every provider of an input should possess equal governance rights.

Workers contribute their own time and activity, are directly subject to managerial authority, and ordinarily cannot diversify their labour across hundreds of firms as investors diversify financial holdings. Anderson’s (2017) account of workplace “private government” and Dahl’s (1985) case for economic democracy develop different versions of the concern that workplace authority itself requires normative justification.

These facts can justify special worker voice, but they do not establish worker ownership as the complete socialist solution. Codetermination may address authority without transferring residual income; social wealth funds may distribute capital income without democratizing workplaces; and worker cooperatives may democratize governance internally while remaining sectional organizations relative to consumers, creditors, communities, and future workers.

2.5 Alienation and meaningful work

Marx’s 1844 manuscripts provide the canonical socialist formulation of alienated labour as a problem involving workers’ relation to their activity, its products, and their own capacities (Marx 1844/1992). Socialism has also sought forms of work permitting greater self-realization, autonomy, creativity, and identification with productive activity.

This aspiration should not be inferred directly from ownership. Public or socially owned enterprises can be hierarchical, monotonous, and alienating. Private organizations can sometimes provide autonomy, professional meaning, or creative work.

Job satisfaction, engagement, autonomy, task significance, and worker control bear on parts of the phenomenon, but there is no basis for treating social ownership as a validated empirical proxy for lower alienation.

The institutional question is therefore which changes actually affect work experience: worker governance, task design, autonomy, occupational mobility, working hours, professional norms, or ownership.

Alienation remains a legitimate outcome dimension, but ownership form is an empirical hypothesis about it, not a proxy.

2.6 Solidarity

Socialist institutions are often defended as promoting solidarity or community.

But institutions cannot assume the motivation they are supposed to generate.

Worker owners can pursue sectional interests. Public agencies can bargain for resources. Universal institutions can generate solidarity but also political conflict over access or contribution. Nor would solidarity observed under social ownership identify ownership as the mechanism: shared stakes, repeated interaction, and small scale are plausible alternative explanations, and the comparative question is which institutional features actually generate and sustain it.

Solidarity should therefore be treated as an outcome requiring explanation.

2.7 Ecological sustainability

Ecosocialist arguments should distinguish welfare and consumption broadly understood from damaging material and energy throughput.

The relevant ecological quantities include extraction, emissions, waste, land transformation, and ecosystem disruption, not “consumption” without qualification, and insofar as energy supply decarbonizes and growth shifts toward less materially intensive goods and services, the coupling of welfare to throughput loosens. Hickel, for example, defines degrowth as planned reduction of energy and resource use rather than as a simple policy of lowering GDP or welfare (Hickel 2021). Saito advances an explicitly Marxian case for “degrowth communism” and criticizes productivist socialism as well as capitalism (Saito 2023).

Social ownership does not automatically solve these problems. Public enterprises and historical socialist formations have caused severe environmental damage.

A socialist ecological advantage would require mechanisms such as more stable democratic constraints on throughput, investment less dependent on private appropriability, or reduced political power of actors benefiting from environmental externalization.

These claims should be compared with alternative mechanisms such as environmental pricing, regulation, commons governance, technological substitution, and public investment under other ownership systems.

3. Allocation mechanisms under socialism

3.1 Markets and dispersed knowledge

The modern debate opens with Mises’s calculation argument: without market prices for capital goods, a socialist administration cannot reduce heterogeneous production possibilities to a common cost measure, so rational allocation of producer goods becomes impossible (Mises 1920/1935). Lange’s reply conceded the need for prices while denying that they require private ownership: a planning board could post accounting prices and adjust them by trial and error against observed shortages and surpluses (Lange 1936–1937).

Hayek’s knowledge argument moved the debate from calculation to discovery, and it imposes a substantial burden on socialist systems that suppress price formation across wide domains. Prices can communicate scarcity and local information without requiring centralized collection (Hayek 1945).

The inference is nevertheless narrower than a defence of private ownership. Bardhan and Roemer make the distinction explicit in their proposal for competitive socialism: they seek to preserve competitive markets while altering ownership and monitoring institutions (Bardhan and Roemer 1992).

Market socialism therefore accepts much of the informational case for prices while relocating the disagreement to capital ownership, investment, governance, and residual claims.

The information argument is a constraint on allocation design rather than a refutation of socialism. Coase’s (1937) make-or-buy margin survives socialization for the same reason: whoever owns firms, a boundary between administrative coordination and market exchange must be set somewhere, and it moves with the costs of transacting.

Stiglitz’s information-economics assessment of market socialism supplies a complementary caution: preserving prices does not restore the complete-information assumptions of the Walrasian benchmark, so market-socialist design still confronts problems of imperfect information, incomplete markets, incentives, and governance (Stiglitz 1994).

The non-market side of the debate did not end with Lange’s market simulation. Later proposals argue that modern computation makes direct calculation in labour time or in kind feasible at scale (Cockshott and Cottrell 1993). This entry takes no position on that revival beyond the constraint already stated: an allocation design must say where dispersed information enters and what disciplines producers when it does not.

3.2 Planning, shortages, and scale

Planning problems should be specified mechanistically.

Administered prices, quantity targets, incentives to misreport, weak enterprise autonomy, or extensive upward referral can produce persistent mismatches between supply and demand. Classical analyses of socialist shortage economies and soft budget constraints identify several such mechanisms rather than treating scarcity as a semantic property of socialism (Kornai 1980; Kornai, Maskin, and Roland 2003).

Planning infrastructure or coordinating a known public investment problem is different from detailed economy-wide administrative allocation. The informational problem depends on scale, uncertainty, local knowledge, and the ability of lower units to act autonomously.

3.3 Market failures remain

Market socialism retains many market failures. Worker-owned firms can pollute; socially owned insurers can face adverse selection; market-socialist industries can concentrate; consumers can remain boundedly rational; and aggregate-demand coordination problems can persist. Changing ownership therefore does not automatically correct a market failure: whoever owns a firm, affected outsiders do not thereby join the ownership constituency, the constituency-partiality point that §4.1 develops.

3.4 What planning arguments establish

Generic planning failures become arguments against socialism tout court only if meaningful socialization cannot coexist with adequate price formation, decentralized information, and enterprise autonomy.

That stronger claim remains contested. Competitive market-socialist proposals are designed precisely to separate socialized ownership from comprehensive administrative allocation (Bardhan and Roemer 1992, 1994).

The more distinctively socialist questions therefore concern ownership, investment, residual authority, and the institutional consequences of socialization, as the third of the introduction’s four levels anticipated.

4. Ownership, governance, and socialization

4.1 The ownership-structure challenge

Hansmann’s (1996) theory of enterprise ownership makes a basic structural point: investor-owned corporations, worker cooperatives, consumer cooperatives, and producer cooperatives assign residual claims and governance rights to different constituencies. Heath (2025) presses the point normatively in comparing cooperatives with corporations.

A worker cooperative therefore does not automatically convert a partial organization into one representing society. It changes which constituency occupies the residual position.

Workers can still impose costs on consumers, creditors, nonmember employees, communities, or future workers.

Cooperative ownership is therefore not automatically society-wide social ownership or general democratic control.

4.2 The worker-specific reply

The structural analogy does not erase the special position of workers established in §2.4: personal exposure to managerial authority, and labour that cannot be diversified in the way a financial portfolio can. That position grounds a stronger claim to workplace governance than capital providers possess merely as suppliers of an alienable input.

The conclusion is nonetheless specific: worker voice is strongly supported before comprehensive worker ownership or socialism is.

This suggests dividing institutional tasks rather than asking worker cooperatives to solve everything:

• worker governance for workplace authority;
• broad citizen ownership for capital-income distribution;
• plural investment institutions for project selection;
• public regulation and stakeholder institutions for effects on outsiders.

4.3 Capital provision and functional conservation

Removing shareholders does not remove the need for capital.

Resources must still be committed before returns exist. Risks must be borne. Managers must be monitored. Projects must fail or expand.

Worker cooperatives can use retained earnings or debt. Socialism can use public banks, social funds, cooperative finance, or other arrangements.

Hansmann’s (1996) analysis matters here because investor ownership can be understood partly as an organizational response to the costs of contracting with different constituencies and to the problem of supplying residual risk capital, a reading Heath (2025) adopts.

The functional-conservation question is:

Where do capital provision, risk-bearing, monitoring, failure, and scaling go after investor ownership is transformed?

Any plausible socialist design must answer it.

4.4 Social investment, plurality, and entry

Several social investment funds do not automatically constitute epistemic decentralization.

The harder question is whether dissenting judgment can create a new allocator.

Private capital markets contain one possible mechanism: an actor controlling resources can sometimes fund an unorthodox project or establish a new intermediary without persuading existing allocators. This mechanism is constrained by unequal wealth, herd behaviour, regulation, and incumbent power, but it identifies an important function.

The difficulty is structural, not merely attitudinal. Where a single social owner stands behind every fund, competition among the funds is partly the owner competing with itself: losses anywhere land on the same balance sheet, so the owner’s incentive to let one of its funds beat another weakens, and under tight resources the temptation is to coordinate rather than contest. Plurality of funds therefore delivers epistemic decentralization only if their charters bind the owner to tolerate, and keep financing, dissent against its own current view.

Market-socialist proposals contain partial attempts to preserve investment plurality. Roemer’s (1994) A Future for Socialism develops a market-socialist architecture in which public ownership is combined with market valuation rather than a single comprehensive investment bureaucracy; Bardhan and Roemer (1992) likewise propose competitive socialist arrangements intended to preserve monitoring and market discipline under altered ownership.

A socialist investment architecture therefore needs not just plural funds but contestable investment authority.

Who creates fund seven when funds one through six share the same mistaken view?

Possible mechanisms include automatic chartering rules, citizen allocation rights, regional autonomy, cooperative financial entry, or institutionally protected minority investment funds. Each raises further governance questions.

The key variable is the possibility of funded dissent.

4.5 Soft budgets

A soft budget constraint arises when an organization expects support from another actor in financial trouble (subsidy, refinancing, tolerated arrears, guarantees, price support, or liabilities eroded by inflation) rather than facing contraction, restructuring, or exit, and the expectation shapes its behaviour before any losses become persistent. Kornai made the phenomenon central to socialist economic analysis; Kornai, Maskin, and Roland later generalized the concept explicitly across socialist, transitional, and market economies (Kornai, Maskin, and Roland 2003).

The problem can therefore survive market pricing and is not uniquely socialist.

A socially owned enterprise may nevertheless face a characteristic rescue mechanism because closure can impose unemployment, regional costs, service disruption, or political embarrassment on institutions that are simultaneously its owners and politically accountable for its consequences.

One possible sequence is:

social responsibility for enterprise outcomes
→ high political cost of failure
→ expectation of rescue
→ weakened disciplinary effect of losses
→ distorted investment or resource demands.

Every link is contingent. Independent ownership bodies, explicit subsidy rules, credible bankruptcy, or strong enterprise autonomy can interrupt the sequence.

The socialist-specific question is whether social ownership makes rescue obligations sufficiently systematic that credible failure becomes difficult.

4.6 Public choice and administrative capacity

Social ownership does not turn public officials into a social-welfare function.

Politicians, administrators, managers, unions, organized beneficiaries, and voters possess their own interests, information, and cognitive limitations. Stiglitz and Krueger provide influential economic accounts of how public institutions can be diverted by incentives, institutional structure, and political behaviour rather than implementing a notional social optimum (Krueger 1990; Stiglitz 1998).

Public investment can be influenced by patronage; universal provision can create administrative burdens; redistribution can generate avoidance and evasion; and political constituencies can support expenditures whose costs are spread widely. Social insurance changes search and other behaviour too, though the welfare consequences depend on the mechanism; Chetty’s (2008) unemployment-insurance analysis, for example, distinguishes liquidity effects from conventional moral hazard.

These are not reasons to presume private provision superior. Private systems have agency, transaction, monopoly, information, and capture problems of their own.

The comparison must treat both sets of actors symmetrically.

4.7 Residual authority

Social ownership creates an unavoidable constitutional problem.

If workers, consumers, communities, social funds, and national political institutions disagree about the use of a socially owned asset, who finally decides?

If the answer is always a central owner, decentralization can become fragile; if lower units possess rights the social owner cannot override, society-wide ownership has less operational content.

Hodgson’s criticism of guild socialism identifies one possible mechanism: insufficiently autonomous lower-level bodies can refer disputes upward, progressively increasing the information burden and authority of the centre (Hodgson 2023).

A democratic socialism therefore requires constitutional rules distributing authority over socially owned resources rather than the phrase democratic control alone.

4.8 Firms and society: structural dependence remains

Changing firms into worker cooperatives does not make firm interests identical to social interests.

Governments still depend on productive organizations for employment, output, investment, tax revenue, and technological capability. Lindblom’s structural-dependence argument was developed for capitalist business, but its underlying mechanism suggests a broader question about dependence on autonomous producer organizations (Lindblom 1977).

A worker-owned firm can therefore possess structural importance in much the way a privately owned firm can, although changing ownership changes the interests and mobility of the controlling constituency.

Worker-members may prioritize jobs, current income, working conditions, or preservation of the enterprise. Consumers may prefer lower prices. Society may prefer environmental restrictions, regional reallocation, automation, or closure.

Socialism therefore does not abolish structural dependence by changing the owner. It changes which organized constituency controls productive decisions.

This is why worker control is not equivalent to social control.

Markets, competition, environmental rules, taxation, consumer representation, social investment, and credible failure can remain necessary even where workers govern firms.

5. Institutional interaction and system effects

5.1 Complementarity

Social insurance can improve worker exit and therefore change the effects of workplace governance.

Markets can alter how socially owned enterprises respond to demand.

External finance can change the behaviour of worker-controlled firms.

These are complementarities, not necessarily socialist system properties.

5.2 Sequential feedback

Some important socialist hypotheses are causal sequences.

One soft-budget sequence was described above.

A centralization sequence might be:

weak local legal autonomy
→ disputes move upward
→ central information and adjudication burdens expand
→ central authority becomes more consequential.

A political-concentration sequence might be:

central social ownership
→ political control over employment and investment
→ greater dependence of organizations on incumbents
→ weaker independent material bases for opposition
→ greater political concentration.

Each link needs evidence. The point of the sequence is precisely to keep authoritarian outcomes inside the evidential comparison without attributing them to “socialism” merely by co-occurrence.

5.3 Joint capacities

Several socialist institutions might jointly enable combinations unavailable from any one alone.

Broad citizen ownership could distribute capital income; plural social funds could preserve investment diversity; market pricing could preserve decentralized information; worker governance could reduce workplace domination; and universal provision could strengthen outside options. A package that achieved all of these without creating a dominant residual authority would possess a genuine joint capacity.

At present this is more a design hypothesis than a historically demonstrated socialist system property. That asymmetry with some well-developed capitalist financial architectures should be acknowledged rather than filled by aspiration.

5.4 Voice and exit

Socialist arguments characteristically emphasize voice: people should possess meaningful influence over organizations exercising consequential power over them. Liberal and market arguments frequently emphasize exit: people should be able to leave an objectionable organization and choose an alternative.

Hirschman’s distinction between exit and voice is useful precisely because the two mechanisms can substitute for or complement one another rather than belonging to opposing ideologies (Hirschman 1970).

Worker governance can constrain managers internally while alternative employment constrains the organization externally. Social ownership can coexist with multiple investment bodies rather than a single allocator. Universal services can make exit from an employer less costly.

A pluralist socialism therefore has reason to preserve both voice and exit where feasible.

5.5 Broad ownership versus dispersed authority

Millions of citizens can be beneficial owners through a social wealth fund while investment authority remains highly concentrated.

Likewise, millions of private pension beneficiaries can own claims while governance rights are exercised through a few intermediaries.

The relevant variables are separate:

• breadth of beneficial ownership;
• distribution of income;
• governance rights;
• investment authority;
• residual political-economic power.

Socialism’s democratic aspirations concern the latter dimensions at least as much as nominal breadth of ownership.

6. Evidence, cases, and comparison

6.1 Economic-system transitions

Both capitalist and socialist transitions have occurred under war, political rupture, low initial income, agrarian conditions, colonial or postcolonial state-building, geopolitical pressure, and rapid catch-up.

Adverse starting conditions are not a peculiarly socialist confounder.

Socialist transitions are often easier to identify because the project is programmatic. Capitalist change can be gradual and retrospective, although deliberate liberalization, privatization, reconstruction, and capitalist development strategies create similar identification problems.

Przeworski’s analysis of social-democratic strategy is useful here because it treats transition as an endogenous political problem rather than assuming that a normatively preferred endpoint can simply be implemented (Przeworski 1985).

6.2 Historical inheritance and external complements

Socialist formations inherited technologies, knowledge, infrastructure, and organizational possibilities developed elsewhere. Capitalist formations likewise inherited enormous capabilities from earlier noncapitalist societies.

National systems also depend on contemporary external complements.

Nordic social democracies operate inside wider European and transatlantic security, technological, financial, and trading systems; China’s development has depended partly on foreign markets, technologies, and capital (Naughton 2018); the Soviet economy imported technologies and interacted strategically with capitalist rivals; and the United States itself draws on global science, migration, supply chains, finance, and allied institutions. No national arrangement should automatically receive credit for capabilities supplied externally.

The causal model is therefore:

initial conditions + inherited capabilities + external complements
→ institutional arrangement
→ behavioural and organizational responses
→ intermediate outcomes
→ political and economic feedback
→ welfare and other outcomes.

6.3 Interacting systems

Capitalist and socialist systems also alter one another.

The existence of socialist movements or geopolitical rivals may change welfare-state institutions and labour bargaining inside capitalist democracies.

Capitalist technological systems change socialist catch-up opportunities.

Sanctions, military competition, migration, ideological competition, and trade change both.

Historical systems are therefore not independent treatment units.

6.4 Worked cases

Yugoslavia is useful because it deliberately combined social ownership, worker self-management, enterprise decentralization, and increasing reliance on markets. Uvalić’s study of Yugoslav investment and property rights treats it explicitly as a self-managed market-socialist experiment and shows why investment behaviour remained a key unresolved institutional problem (Uvalić 1992).

The Meidner wage-earner-fund proposal tests incremental socialization inside parliamentary capitalism. Its dilution and reversal bear especially on transition, coalition stability, and the political response of incumbent capital rather than on the steady-state economics of a completed socialist system (Westerberg 2023; Przeworski 1985).

Uruguay’s worker-managed firms test self-management under competitive product markets, with no financial accommodation documented, using administrative registry data that follow cooperative and conventional firms alike through shocks and dissolution (Burdín and Dean 2009; Burdín 2014). Section 6.7 develops the case.

China tests component-level inference because market allocation, private ownership, state ownership, investment institutions, and political authority have changed at different rates (Naughton 2018). The question is not which single label China deserves, but which outcomes changed as particular mechanisms changed.

Classical Soviet-type systems remain especially relevant for hypotheses about comprehensive planning, administered prices, soft budgets, and concentration of political-economic authority (Kornai, Maskin, and Roland 2003).

These cases should test mechanisms rather than contribute points to a socialism scorecard.

6.5 Two tests before verdicts

Every reform in the cases above faces two tests. Functional conservation asks where the transformed institution’s functions go, the question §4.3 put to capital provision. Dynamic response asks what agents do after the reform and what distribution their behaviour generates: an egalitarian asset allocation can reconcentrate, a social fund can become politically or professionally concentrated, a worker cooperative can hire nonmember workers, a welfare guarantee can alter saving or labour supply, and a private capital market can concentrate and then change political rules. Institutional evaluation should track not only the immediate allocation an arrangement creates but the behavioural responses, subsequent distributions, external dependencies, and political feedback it generates over time; the relevant object is the institutional trajectory rather than the initial constitutional diagram.

The same discipline covers qualitative aspirations. Ownership form should not stand in for meaningful work, autonomy, or worker satisfaction, and comparative evidence bears on them only through identified governance and labour mechanisms: low reported satisfaction under a nominally socialist formation would not refute socialism generally, but it would constrain any claim that social ownership by itself removes alienation.

6.6 A worked inference: Yugoslav soft budget constraints

The claim under test is that socialism causes soft budget constraints, cut down to a form evidence can reach: did Yugoslavia’s socialist architecture cause the softness observed in its enterprises, and how far does that finding travel? The example is adverse by design: a framework for a programmatic category must be able to treat an implementation as evidence even when the outcome violates the motivating ideal, or failure can always be defined away. The rule of §1.6 privileges neither valence: the companion entry’s §6.7 runs the same schema to a favourable verdict about a capitalist financial architecture, and §6.7 below runs it to a favourable verdict about a socialist institutional element.

Yugoslavia is a demanding test because its market-socialist formation combined social ownership and worker self-management with substantial decentralization and reliance on markets, rather than simply reproducing Soviet central planning (Estrin 1991; Uvalić 1992). The market reliance varied by period: the 1974–1988 reforms were a deliberate move back from allocation by markets toward bureaucracy and bargaining, so the 1986 evidence below comes from a decentralized but only partly market-allocated phase (Estrin 1991).

At least by the 1980s, the outcome is unusually well identified at the enterprise level. Using firm-level data for Yugoslavia’s manufacturing sector in 1986, Kraft and Vodopivec find massive and pervasive redistribution through a soft budget constraint despite the virtual absence of direct government subsidies to firms and the existence of open unemployment. The important channels were financial rather than ordinary budgetary subsidies: assets and liabilities were generally not indexed in a highly inflationary environment, and some formally repayable investments were returned at face value or low nominal interest rates, implying substantially negative real rates. The resulting redistribution sharply compressed differences in income per worker across enterprises. Yugoslavia’s channels differed significantly from those documented in other socialist economies, but Kraft and Vodopivec identify the same underlying objective: the pursuit of job and wage security (Kraft and Vodopivec 1992).

Inflation therefore supplied an important vehicle for the 1986 redistribution, but it is not by itself the institutional explanation. The relevant questions are why non-indexed or negative-real-rate finance remained available, who ultimately absorbed the losses, and why the financing relationship was not hardened. Uvalić’s broader institutional analysis bears on those questions. Yugoslavia had deliberately decentralized decision-making and, relative to Soviet-type planning, increased reliance on market forces, yet social property persisted alongside a paternalistic relationship in which firms continued to receive protection and support from political institutions before and after market-oriented reforms (Uvalić 1992).

The pervasiveness matters for attribution. Kornai, Maskin, and Roland show that soft budget constraints can occur in market economies too; rescue is therefore not a semantic property of socialism (Kornai, Maskin, and Roland 2003). Kraft and Vodopivec, however, are not documenting an exceptional bailout of one salient enterprise. They find a broad redistributive architecture across ordinary manufacturing firms that systematically narrowed enterprise-income differences. Their additional finding that Yugoslavia used markedly different channels from other socialist economies while pursuing the same job-and-wage-security objective suggests that the underlying objective can work through different institutional vehicles.

The attribution test can therefore be run visibly. First, a mechanism is identified, and its parts have distinct owners. Kornai, Maskin, and Roland supply the theory: budget constraints soften when commitments to withhold rescue lack credibility, because the supporting organizations have done nothing to alter their vested interest in providing support (Kornai, Maskin, and Roland 2003). Kraft and Vodopivec supply the driving force: Yugoslavia’s redistributive channels differed from those of other socialist economies but shared “the pursuit of job and wage security” (Kraft and Vodopivec 1992). Second, ex post accommodation is observed: firm-level redistribution in 1986 was pervasive, large enough to compress enterprise incomes sharply, and channelled substantially through softened finance. The expectations that constitute softness in the Kornai–Maskin–Roland sense are not directly observed in the 1986 data (expectations rarely are), so the claim that termination lacked credibility is an inference from observed accommodation together with the persistence of the support relationship, not an independent measurement. Third, it was connected to the institutional structure rather than merely to product-market suppression: social property, politically mediated wage objectives, finance, and state–enterprise paternalism persisted in a system that had already decentralized substantially, though one no longer relying primarily on market allocation. Fourth, the mechanism resisted partial correction: Uvalić finds the protective state–enterprise relationship continuing through reforms intended to increase market discipline. Fifth, the evidence supports a limited form of robustness at the level of the objective rather than the mechanism: the redistributive channels differed from those elsewhere in Eastern European socialism while the job-and-wage-security objective recurred, which shows that the objective can recruit different institutional vehicles, not that one mechanism operated identically everywhere. That supports projection to architectures combining social responsibility for enterprise survival with finance that cannot credibly be hardened, not to socialism in general.

The framework therefore warrants a bounded verdict:

At least by the 1980s, Yugoslavia’s self-managed market-socialist formation had materially soft enterprise budget constraints, and its combination of social property with politically and socially mediated responsibility for job and wage security, enterprise finance, and survival was a causal contributor to that softness.

The scope restriction is equally substantive. The evidence does not warrant worker self-management causes soft budgets, social ownership causes soft budgets, or socialism causes soft budgets. Kornai, Maskin, and Roland’s cross-system analysis itself rules out that inference. Nor does the Yugoslav case directly test market-socialist designs in which investment institutions are legally independent, failed firms face credible insolvency, asset rights are clearer, finance is indexed or otherwise insulated from inflationary redistribution, and political bodies cannot readily refinance individual enterprises. It establishes a vulnerability of socialist architectures that combine social ownership with weakly credible exit and residual political responsibility for firm survival. Whether alternative forms of social ownership can interrupt that mechanism remains an open comparative question.

This is the kind of conclusion the attribution rule is intended to permit: stronger than historical co-occurrence, narrower than a verdict on socialism tout court, and informative about which institutional change would have to differ for the inference not to travel. Its design implication is the mirror of its scope restriction: architectures that want social ownership without the softness need legally independent investment institutions, credible insolvency, clearer asset rights, insulated finance, and limits on political refinancing, the features whose absence the verdict identifies.

6.7 A worked inference: worker-managed firms under market discipline

The claim under test is the converse of §6.6’s: does worker self-management itself entail soft budgets and fragility, or can it operate under market discipline? The example is favourable by design, and it pairs with §6.6 deliberately: the same objective, employment security, appears in both, under different financing architectures. It also differs in categorization. Uruguay’s worker-managed firms are a socialist institutional element operating inside a predominantly capitalist formation (§1.4), so the evidence bears on socialist institutional design, worker self-management in particular, rather than on socialism as a formation (§1.6).

The evidence is administrative rather than anecdotal. Monthly social-security records covering April 1996 to December 2005 allow worker-managed firms and conventional firms to be followed through entry, shocks, and dissolution within one legal and macroeconomic environment (Burdín and Dean 2009; Burdín 2014). Two findings matter here. Employment in worker-managed firms is inelastic to output-price shocks while pay adjusts, the mirror image of conventional firms; during the 2002 crisis, both types cut pay to a similar extent while worker-managed firms cut employment significantly less (Burdín and Dean 2009). And worker-managed firms were not more likely to fail: their hazard of dissolution is about 29 per cent lower than that of comparable conventional firms once microenterprises are excluded and each firm’s effective tax burden is controlled for, with about a third of the gap statistically mediated by employment stability (Burdín 2014). The tax control matters because the documented state support for cooperatives took the form of statutory tax exemptions; the survival difference survives netting it out. Neither study documents subsidy, rescue, or refinancing channels of the kind §6.6 turns on.

The attribution test runs as before. First, the mechanism: worker-managed governance internalizes employment security as an objective and holds the authority to trade pay against jobs, so where no external actor accommodates losses, adjustment must come from inside the firm, on the pay margin. Second, it operated: the adjustment asymmetry is exactly what the panel shows, within firms over time. Third, the predicted difference appears within one national legal and macroeconomic environment and survives the studies’ reported controls, which makes governance a plausible explanation of the differing adjustment margins; but the firms are not randomly assigned to governance forms, and Burdín states self-selection of firms into industries and of workers into organizational forms as an identification threat, so the evidence establishes a robust organizational association, and compatibility with market discipline, more securely than an isolated causal effect. Fourth, the arrangement sustained itself under market discipline rather than requiring correction: dissolution was a live outcome, and worker-managed firms experienced less of it. Fifth, robustness is limited and stated: the panel spans a severe crisis, but the survival difference is concentrated in services and not statistically significant in manufacturing or transport, and everything rests on one country’s registry.

The framework therefore warrants a bounded verdict:

In Uruguay over the periods studied, worker-managed firms showed adjustment consistent with protecting employment through pay flexibility, and lower dissolution hazards than comparable conventional firms after the reported controls. The studies document differential tax treatment, but nothing like the broad accommodation of enterprise losses observed in Yugoslavia.

The scope restriction mirrors §6.6’s. The evidence does not warrant self-management outperforms conventional ownership, cooperatives scale economy-wide, or anything about socialized investment: these firms competed for members, inputs, and customers inside capitalist finance and product markets, and the sectoral concentration of the survival result is a real limit. What the pairing licenses is precise. In Yugoslavia, social responsibility for jobs and wages met finance that could not credibly be hardened, and budget constraints softened; in Uruguay, the same objective met an environment where no accommodation is documented, and it was pursued on the pay margin without a survival penalty. Jointly the two cases reject the claim that worker self-management entails soft budgets, and they support the hypothesis that the financing and rescue architecture, rather than governance form alone, determines whether employment-security objectives generate softness: the Kornai–Maskin–Roland point from §4.5 in its positive form. The pairing therefore suggests a design hypothesis: where worker governance permits adjustment through pay while finance remains external and not broadly accommodative, employment-security objectives need not generate soft budget constraints.

6.8 The common institutional frontier

Capitalist, socialist, and hybrid systems should be evaluated along the same dimensions:

• productive capacity;
• welfare;
• security;
• equality;
• liberty and effective exit;
• workplace voice;
• meaningful work;
• freedom from domination;
• political pluralism;
• innovation and discovery;
• ecological sustainability;
• resilience and corrigibility.

Disagreement can concern empirical consequences, feasible and stable institutional packages, or normative weights.

A socialist aspiration does not determine its institutional solution. A capitalist mechanism does not inherit a favourable normative weighting.

The useful question is:

Which feasible institutional packages lie on the relevant frontier once production, distribution, behavioural response, political feedback, external complements, and long-run stability are included?

An outcome provides evidence about an economic system as such only to the extent that an identified mechanism is sufficiently central to the relevant institutional structure, depends on recurrent interactions among its institutions, is reproduced or rendered difficult to correct by endogenous feedback, or remains robust across the range of system variants to which the claim is intended to apply.

The families should be compared as design spaces rather than as bundles of traits; within each family the variants differ on every row. What carries across is the questions, not family-wise answers:

7. Conclusion

Socialism is a family of projects to transform economic power in pursuit of values such as welfare, equality, democracy, non-domination, solidarity, meaningful work, and ecological sustainability.

Its aspirational character is genuine but evidentially demanding. Desirable ends do not establish that institutions proposed in their name will realize them.

A serious analysis therefore follows the sequence from aspiration to institutional proposal, implementation, mechanism, behavioural response, system interaction, and outcome.

Doing so separates ideas that socialist discourse often combines: markets are not capitalism, and planning is not socialism; worker ownership is not society-wide ownership, worker control is not social control, and social ownership is not democratic control; universal provision is not socialism; equal starting assets do not determine the long-run distribution; and changing ownership does not eliminate the economic functions that the previous ownership structure performed.

The central questions concern residual authority and institutional dynamics. Who owns productive claims? Who receives returns? Who supplies capital? Who selects uncertain projects? Who governs firms? Who bears failure? Who can enter with a dissenting investment judgment? Who can override whom? What happens after people respond to the rules?

Worker cooperatives illustrate the general problem. Workers have unusually strong claims to governance because labour is personally exposing and poorly diversifiable. But a worker-controlled firm remains a producer organization whose goals need not coincide with those of consumers, communities, or society at large.

Social investment presents the parallel problem at a larger scale. Social ownership can broaden claims on productive wealth, but plural funds do not produce genuine decentralization unless dissenting investment institutions can enter and unsuccessful ones can exit.

The conclusion this entry draws, a proposal of its own rather than a reported position, is that the most plausible socialist arrangements may require not a single social owner but a constitutional division of socially constituted power: markets in some domains, worker governance in others, plural investment bodies, universal services, public or common ownership where appropriate, and institutional constraints on both producers and political authorities.

Whether such arrangements outperform the best capitalist and hybrid alternatives is an empirical, institutional, and normative question. Socialism’s aspirations make that comparison worth undertaking; they do not determine its result.

The method therefore does not make system-level inference impossible. It usually returns a smaller grammatical subject than ideological argument prefers.

Bibliography

Anderson, Elizabeth. 2017. Private Government: How Employers Rule Our Lives (and Why We Don’t Talk about It). Princeton: Princeton University Press.

Bardhan, Pranab, and John E. Roemer. 1992. “Market Socialism: A Case for Rejuvenation.” Journal of Economic Perspectives 6(3): 101–116. https://doi.org/10.1257/jep.6.3.101.

Bardhan, Pranab, and John E. Roemer. 1994. “On the Workability of Market Socialism.” Journal of Economic Perspectives 8(2): 177–181. https://doi.org/10.1257/jep.8.2.177.

Burdín, Gabriel. 2014. “Are Worker-Managed Firms More Likely to Fail Than Conventional Enterprises? Evidence from Uruguay.” ILR Review 67(1): 202–238. https://doi.org/10.1177/001979391406700108.

Burdín, Gabriel, and Andrés Dean. 2009. “New Evidence on Wages and Employment in Worker Cooperatives Compared with Capitalist Firms.” Journal of Comparative Economics 37(4): 517–533. https://doi.org/10.1016/j.jce.2009.08.001.

Chetty, Raj. 2008. “Moral Hazard versus Liquidity and Optimal Unemployment Insurance.” Journal of Political Economy 116(2): 173–234. https://doi.org/10.1086/588585.

Coase, R. H. 1937. “The Nature of the Firm.” Economica 4(16): 386–405.

Cockshott, W. Paul, and Allin F. Cottrell. 1993. Towards a New Socialism. Nottingham: Spokesman.

Dahl, Robert A. 1985. A Preface to Economic Democracy. Berkeley: University of California Press. https://doi.org/10.1525/9780520341166.

Diamond, Peter, and Emmanuel Saez. 2011. “The Case for a Progressive Tax: From Basic Research to Policy Recommendations.” Journal of Economic Perspectives 25(4): 165–190. https://doi.org/10.1257/jep.25.4.165.

Dworkin, Ronald. 1981. “What Is Equality? Part 2: Equality of Resources.” Philosophy & Public Affairs 10(4): 283–345.

Estrin, Saul. 1991. “Yugoslavia: The Case of Self-Managing Market Socialism.” Journal of Economic Perspectives 5(4): 187–194. https://doi.org/10.1257/jep.5.4.187.

Hansmann, Henry. 1996. The Ownership of Enterprise. Cambridge, MA: Belknap Press of Harvard University Press.

Hayek, F. A. 1945. “The Use of Knowledge in Society.” American Economic Review 35(4): 519–530.

Heath, Joseph. 2025. “Are Cooperatives More Virtuous than Corporations?” Politics, Philosophy & Economics. OnlineFirst. https://doi.org/10.1177/1470594X251387579. https://doi.org/10.1177/1470594x251387579.

Hickel, Jason. 2021. “What Does Degrowth Mean? A Few Points of Clarification.” Globalizations 18(7): 1105–1111. https://doi.org/10.1080/14747731.2020.1812222.

Hirschman, Albert O. 1970. Exit, Voice, and Loyalty: Responses to Decline in Firms, Organizations, and States. Cambridge, MA: Harvard University Press.

Hodgson, Geoffrey M. 2023. “The Institutional Impossibility of Guild Socialism.” Cambridge Journal of Economics 47(1): 21–43. https://doi.org/10.1093/cje/beac057.

Kornai, János. 1980. Economics of Shortage. 2 vols. Amsterdam: North-Holland.

Kornai, János, Eric Maskin, and Gérard Roland. 2003. “Understanding the Soft Budget Constraint.” Journal of Economic Literature 41(4): 1095–1136. https://doi.org/10.1257/jel.41.4.1095.

Kraft, Evan, and Milan Vodopivec. 1992. “How Soft Is the Budget Constraint for Yugoslav Firms?” Journal of Comparative Economics 16(3): 432–455. https://doi.org/10.1016/0147-5967(92)90166-5.

Krueger, Anne O. 1990. “Government Failures in Development.” Journal of Economic Perspectives 4(3): 9–23. https://doi.org/10.1257/jep.4.3.9.

Lange, Oskar. 1936–1937. “On the Economic Theory of Socialism.” Parts 1 and 2. Review of Economic Studies 4(1): 53–71; 4(2): 123–142.

Lindblom, Charles E. 1977. Politics and Markets: The World’s Political-Economic Systems. New York: Basic Books.

Marx, Karl. 1844/1992. “Economic and Philosophical Manuscripts (1844).” In Early Writings, translated by Rodney Livingstone and Gregor Benton. London: Penguin Books in association with New Left Review.

Marx, Karl. 1867/1990. Capital: A Critique of Political Economy, vol. 1. Translated by Ben Fowkes. London: Penguin Books in association with New Left Review.

Meidner, Rudolf, in cooperation with Anna Hedborg and Gunnar Fond. 1978. Employee Investment Funds: An Approach to Collective Capital Formation. London: George Allen & Unwin.

Mises, Ludwig von. 1920/1935. “Economic Calculation in the Socialist Commonwealth.” In Collectivist Economic Planning, edited by F. A. Hayek, translated by S. Adler. London: George Routledge & Sons.

Naughton, Barry J. 2018. The Chinese Economy: Adaptation and Growth. 2nd ed. Cambridge, MA: MIT Press.

Nove, Alec. 1983. The Economics of Feasible Socialism. London: George Allen & Unwin.

Nozick, Robert. 1974. Anarchy, State, and Utopia. New York: Basic Books.

Przeworski, Adam. 1985. Capitalism and Social Democracy. Cambridge: Cambridge University Press. https://doi.org/10.1017/cbo9781139171830.

Roemer, John E. 1982. A General Theory of Exploitation and Class. Cambridge, MA: Harvard University Press. https://doi.org/10.4159/harvard.9780674435865.

Roemer, John E. 1994. A Future for Socialism. Cambridge, MA: Harvard University Press.

Saito, Kohei. 2023. Marx in the Anthropocene: Towards the Idea of Degrowth Communism. Cambridge: Cambridge University Press.

Stiglitz, Joseph E. 1994. Whither Socialism? Cambridge, MA: MIT Press.

Stiglitz, Joseph E. 1998. “Distinguished Lecture on Economics in Government: The Private Uses of Public Interests: Incentives and Institutions.” Journal of Economic Perspectives 12(2): 3–22. https://doi.org/10.1257/jep.12.2.3.

Uvalić, Milica. 1992. Investment and Property Rights in Yugoslavia: The Long Transition to a Market Economy. Cambridge: Cambridge University Press.

van der Linden, Marcel, ed. 2022. The Cambridge History of Socialism. 2 vols. Cambridge: Cambridge University Press. https://doi.org/10.1017/9781108611022.

Vrousalis, Nicholas. 2023. Exploitation as Domination: What Makes Capitalism Unjust. Oxford: Oxford University Press.

Westerberg, Rikard. 2023. “From Marxist Venture to Venture Capitalists: The Swedish Wage-Earner Funds and the Market Turn, 1983–1994.” Enterprise & Society 24(4): 1014–1037. https://doi.org/10.1017/eso.2022.23.

Wright, Erik Olin. 2010. Envisioning Real Utopias. London: Verso.