Plural Cooperativism. The Material Basis of Democratic Corporate Governance

1 Introduction

In recent political theory, many powerful normative arguments have emerged for democratizing corporate governance much more strongly, understood here as the collective self-governance of those who are subject to corporate authority, based on elections, votes, and prior deliberation. But there is still a relative lack of institutional models that outline how exactly democratic corporate governance ought to be structured.

The numerous normative reasons for expanding and deepening democratic corporate governance are much clearer now than they were a few decades ago: The argument from the analogy between firm and state holds that firms are similar to states in their organizational structure: Just like governments govern citizens, firms govern employees, and they do so within hierarchical structures, backed by a vast arsenal of sanctioning powers, from demotion to dismissal. If being subjected to state government requires democracy, so does being subjected to corporate government (Landemore and Ferreras 2016). The republican argument demonstrates that freedom as non-domination requires protection from arbitrary, unaccountable power, which is particularly urgent in the context of management hierarchies within companies (Gourevitch 2015; Anderson 2017). The argument from democratic virtues points out that firms are an especially important site of forming habits of thought and action, so that democratic firms may contribute to strengthening the democratic skills and virtues of citizens, while non-democratic firms are in danger of fostering authoritarian and fatalistic attitudes (Pateman 1970; O’Neill 2008; Kuch 2021). The argument from internal regulation underscores that external regulation of market institutions through laws and regulations is often ineffective and structurally intransparent, warranting a complementary regulation from within major market institutions, particularly by means of democratized firms (Hussain 2023). This is only a small fraction of the many arguments on the table pressing for a stronger democratization of corporate governance.^[1]

But there is a striking imbalance between our knowledge of the normative arguments for democratic corporations on one hand and our knowledge of institutional models of democratic firm governance on the other. In terms of institutional models, many important political theorists, such as Elizabeth Anderson (2017) and Waheed Hussain (2023), content themselves with quickly pointing to Germany’s codetermination system, in which the seats on the supervisory board of large corporations are composed equally of worker and shareholder representatives. German codetermination seems to demonstrate the separability of ownership and control, a possibility that many view as a sufficient starting point for expanding and deepening democratic control over corporations more extensively.

Drawing on recent work on the relationship of ownership and control, I argue that simply separating ownership and control rights, without burdening oneself with changing the property relations themselves, is insufficient. Democratic control over the corporation must be underpinned by social ownership of the corporation, either in the form of cooperative or public ownership. But this immediately raises the question: which form of social ownership exactly? How precisely should it be organized? This is the major challenge I aim to tackle in this paper. I analyze various models of social ownership to get a clearer picture of the merits of each model. The overview reveals not only a better understanding of the strengths and weaknesses of the various models, it also suggests that these models can be combined, in an attempt to complement one another in their strengths, while offsetting their particular weaknesses. Quite organically, the overview leads to the idea of combining the various forms of social ownership in a hybrid model that I will call ‘Plural Cooperativism’.

Plural Cooperativism envisions a society where cooperative ownership is universalized, so that business corporations basically operate as cooperatives. At the same time, this cooperativism is pluralistic, and in a specific sense: It allows for, and systematically combines, cooperative ownership with various other forms of social ownership, including limited private stock ownership, complemented by public stock ownership, as well as a more democratic reallocation of control rights over private ownership. Certainly, Plural Cooperativism is not a ready-made blueprint, it is an early-stage exploration. The fundamental point is to open up the fundamental conceptual space for thinking about a society-wide model of cooperative ownership.^[2]

Following this introduction, the paper is organized into six sections. Section 2 clarifies the methodological basis for the following analysis of democratic corporate governance and introduces an important substantive assumption. Section 3 reconstructs the promises of unbundling ownership and control yet ultimately rejects this approach for suffering from a legalist illusion. The conclusion of this section is that democratic firm governance requires a particular material basis. Section 4 analyzes three important forms of social ownership that could potentially provide material support for democratic firm governance, namely cooperative ownership, worker self-management and socialized stock markets. Section 5 analyzes important challenges of cooperative ownership. Section 6 introduces a new model of social ownership, outlining the contours of Plural Cooperativism. In Section 7, the paper concludes by discussing challenges and objections to the model.

2 Realistic Utopias: Methodologically and Substantively

A preliminary remark on the approach to analyzing the design of democratic corporate governance is in order, followed by the introduction of a substantive assumption. Both the method and the substantive assumption are closely linked to what John Rawls refers to as the concept of ‘realistic utopias’, which deals with the following question: “What would a just democratic society be like under reasonably favorable but still possible historical conditions, conditions allowed by the laws and tendencies of the social world?” (Rawls 2001, 4) For the matter at hand, corporate governance, the question of realistic utopias can be rephrased as follows: What would a just structure of democratic corporate governance be like under reasonably favorable but still possible historical conditions, conditions allowed by the laws and tendencies of the social world? In posing this question, I reiterate that I assume here that the essential issue is not whether corporate governance should be democratic – it should be – but rather what the “institutional content” (Rawls 2001, 136) of a just structure of democratic corporate governance should look like.

As I see it, the key point of focusing on realistic utopias is to set aside concerns about whether certain institutional proposals might appear politically unfeasible under given social and political conditions. The primary question is how a just structure of democratic corporate governance would be under “reasonably favorable” (Rawls 2001, 4) conditions. Unfavorable political majorities, obstructive power structures, or ideological barriers must not be allowed to constrain our conception of a just democratic corporate governance; otherwise, we risk committing a naturalistic fallacy, letting the ought to be dictated by the given, no matter how bad it may be. We should begin by focusing fundamentally on the normative desirability and institutional feasibility of specific proposals, as long as they are “allowed by the laws and tendencies of the social world” (Rawls 2001, 4). This is crucial for establishing a point of orientation on the path to meaningful political transformation (Simmons 2010). Only in a later step should we turn to more pragmatic concerns about transformation strategies, taking into account questions of political feasibility. However, this later step is not the focus here.

The substantive assumption I make is a system of universal inheritances that provides an equitable starting point for all citizens. Each citizen is granted a significant financial endowment in the early phase of their adult life, financed by a strongly progressive inheritance taxation. This is at the heart of Rawls’s realistic utopia of a property-owning democracy (Rawls 2001; O’Neill 2012). As we shall see, Plural Cooperativism critically depends on universal inheritances. In addition to Rawls, there are many significant voices in contemporary political theory highlighting the importance of universal inheritances for all, financed by a much more progressive taxation of inheritances (Gosepath 2022; Piketty 2022). The reason behind this idea is that in Western capitalist societies, inheritances significantly contribute to the skyrocketing social inequality.^[3] More fundamentally, stark disparities in inheritances violate basic principles of justice, including equality of opportunity and the merit principle.^[4] If, however, one has reservations about taxing inheritances strongly, capital grants could serve as an alternative to universal inheritances, and they can be financed through other means than inheritance taxation, such as wealth taxes (Alstott and Ackerman 2000).

It should be emphasized that even when assuming a system of universal inheritances (or capital grants), the question of how democratic corporate governance should be structured is by no means self-explanatory. The implications for democratic corporate governance are anything but trivial, quite the opposite, only under the assumption of universal inheritances will some important issues of democratic corporate governance really come into focus. All this means that assuming universal inheritances is not only justified for good reasons, it also opens up a significant, nontrivial level of debate.

3 Democratic Corporate Governance and Private Property

Democratizing corporate governance without changing property relations is alluring in its simplicity, which is surely one major reason why it finds so much resonance in contemporary political philosophy.^[5] The idea is to alter the bundle of property rights, without troubling oneself with changing property relations themselves. The traditional understanding of ownership encompasses not only the right to the profit, but also the right to control property.^[6] In the capitalist conception of ownership, these rights are intertwined in a single bundle, yet they can be separated conceptually and, more importantly, they can be separated in practice. At the least, this is where the unbundling approach places its hope. Its wager is that the bundle of property rights can be un-bundled.

This rebundling of property rights has been at least partially achieved by German codetermination law, which applies to larger stock corporations. The right to control property has been decoupled from the right to profit and the right to buy and sell; the latter two rights remain with shareholders, while the former is partially granted to employees. Half of the seats of the supervisory go to worker representatives, the other go to representatives of shareholders. Control over the supervisory board is important since this is the highest-ranking decision-making body within the corporation: It has the formal authority to appoint, dismiss, and supervise top management, and in addition, it decides on major corporate policy issues.

However, power over the supervisory board is only seemingly shared between capital and labor on equal terms, because in case of a stalemate between capital and labor, the chairperson has a tie-breaking vote. Guess who elects the chairperson? Yes, the shareholders. Yet German codetermination still rose to a surprising prominence in political theory in recent years. In Private Government, Elizabeth Anderson refers to it as a „highly successful model“ (Anderson 2017, 70), and Waheed Hussain, in his Living with the Invisible Hand, uses it as the major model to explain his idea of an intermediated market arrangement (Hussain 2023, 193–6).

Others are more critical: In view of the tie-breaking vote of the chairperson, Isabelle Ferreras called out codetermination as “something of a sham” (Ferreras 2017, 173). As shareholders retain ultimate control over major corporate decisions, the actual power of labor representatives is limited, despite the formal voting rights they have on paper. This structural feature perpetuates the dominance of capital in corporate decision-making while simultaneously obscuring it. As both shareholder and worker representatives are at least implicitly aware of the power asymmetry between them, the latter are pressured into proactively complying with the profit expectations of shareholders in order to avoid open conflicts that they are only bound to lose. Codetermination, then, is a recipe for co-optation and pacification: “The history of capitalism shows that labor organizations have, at best, been co-opted into comanaging firms.” (Ferreras 2017, 172)

Still, the fundamental outlook may seem sufficiently attractive: Why not keep pushing the unbundling of property rights forward, by strengthening control rights of workers? Recall the basic strategy: Simply decouple control rights from the right to profit. It seems surely possible to deprive shareholders of their control over corporate decision-making entirely, while significantly strengthening the voice of workers. Ferreras (2017) developed a model of corporate governance where capital and labor are both represented in two separate chambers within the company: Management requires the approval of both chambers, so unlike in the German codetermination model, employees have an effective veto right, thus their control rights are sharply strengthened. Quite some time ago, Jay (1980) developed an even more radical proposal, carrying the decoupling of control rights and profit rights to the extreme: up to decoupling them completely. While shareholders in Jay’s proposal enjoy their rights to profits unrestrictedly, control rights over the corporation are shifted to workers entirely. In Jay’s model, firms are completely financed through shares, and although shareholders retain the right to the company’s profits, control rights are entirely in the hands of the firm’s workers.

The unbundling approach is certainly important: It demonstrates that control rights and profits rights are not welded together in an ironclad way, their relationship is subject to historical transformations. German codetermination showcases the possibility and feasibility of this approach, and from Ferreras’s up to Jay’s models we can see the great extent to which control rights and profits rights can be decoupled – up to their complete separation, where workers enjoy full control rights over their workplace, while profit rights remain with the shareholders.

Yet the prospect of simply unbundling of property rights, while leaving the basic property relations intact, is limited. This approach stands under the spell of a legalist illusion: Purely formal legal changes are intended to fundamentally alter the power of property in its materiality; yet if control rights over property do not extend to the right to buy and sell it, they have little real power. This is because merely strengthening control of workers does not change that fact that the productive assets the enterprise needs can be withdrawn at any time.^[7] Yet enterprises desperately depend on the supply of capital for their goal to produce and sell goods. They need financial resources to acquire productive assets such as machines, buildings, technical equipment, and so on. The formal right to control someone else’s property wields little real power if the owner retains the ability to withdraw the property at whim. Disinvestment is a real and effective threat. As capital is withdrawn, the company may face bankruptcy, and its employees lose their source of livelihood. Therefore, ownership potentially involves power over others, even if formal rights seem to counter this power. In this sense, property is potentially “the power of disposing of the labour-power of others” (Marx and Engels 1976, 46). So, the formal right to control property that workers enjoy can be counteracted by the threat of disinvestment that property owners can yield. Keep in mind here that in advanced capitalist economies, corporations are typically highly dependent on exceedingly large capital resources, for example, due to productivity-improving technology and the scale and scope of production, which magnifies the threat of disinvestment that property owners have at their disposal.

To be sure, in the case of share-issuing stock corporations, the unbundling-approach seems to have better prospects of success. The reason is that shareholders do not really ‘own’ the stock corporation, they merely own their shares. The capital itself is owned by the company as an independent legal entity, something rightly highlighted by David Ciepley and others (Ireland 1999; Ciepley 2013). On this basis, stock corporations are protected from a sell-off of their productive assets by the so-called asset lock-in (Ciepley 2013, 144–5). This means that even though shareholders have every right to sell their shares, they are not permitted direct access to the material wealth of the company, for example its technical equipment. Shareholders cannot walk into the factory and lawfully take a robot with them and sell it. Asset lock-in seems to greatly diminish the threat advantage of shareholders, since they appear to have lost the power to withdraw important resources.

Yet the threat of disinvestment remains. Shareholders retain the full right of selling their shares. Selling shares leads to a decline in value on the stock market when demand doesn’t keep up with supply, and selling shares on a massive scale can lead to a significant loss. At the extreme, shareholders even have the right to dissolve the company completely in order to assert their residual ownership claims. Iñigo González-Ricoy points out the serios consequences:

As capital suppliers, shareholders retain the right to sell their stocks and to collectively terminate the firm. This is public knowledge: workers know this and shareholders know that they know it. Now, while employees may in principle exercise such control rights as they deem more appropriate for their interests, their decisions are likely to be importantly influenced by shareholders’ implicit yet ever-present threat of selling their shares en masse or terminating the firm. (González-Ricoy 2020, 420)

So, suppose shareholders (illegitimately) push for one or more of the following profit-maximizing, cost-saving strategies: squeezing wages, neglecting due care obligations for public safety, or externalizing costs by polluting the environment.^[8] The democratic corporation (legitimately) opposes these strategies, and luckily, shareholder representatives have lost their power to overrule all other stakeholders. However, despite having lost their ultimate authority, shareholders still have the right to sell their shares. Given the resistance of company board, shareholders sell off shares, the share price plummets.

Where is the harm in a large-scale stock sell-off? One might think that this doesn’t pose a threat to the stock corporation, since it does not lose any money as a result – only shareholders do, as the value of their shares dwindles. And obviously, just because there is a stock sell-off doesn’t mean that the corporation loses any physical assets. However, a decline in stock market valuation diminishes the stock corporation’s future prospects of raising capital through issuing shares again. More importantly, a loss of stock market value lowers the corporation’s creditworthiness dramatically, which in turn results in higher interest rates and a greater difficulty obtaining loans. Note here that capital-intensive industries like the manufacturing industries typically depend heavily on debt financing, as their debt is often twice as high as their equity. Thus, the threat of disinvestment remains real, and this would still be the case even in most radical version of separating ownership and control, as in Jay’s model, where shareholders have no control rights whatsoever over the corporation. Yet if shareholders retain the right to sell their shares, their threat advantage over workers remains.

Sure, under a system of universal inheritances, the workers of a formally democratic firm (along the lines of Jay’s model) sustaining a large-scale stock sell-off could draw on their own financial resources to save their firm. Yet if only one heroic employee decides to save the company, it amounts to a wasted effort: The isolated purchase of shares by a single employee would make little or no difference. However, attempting to get all employees to save their company with their own resources comes with almost insurmountable difficulties, particularly massive collective action problems.

It follows that the legalistic hope for achieving democratic control over the corporation by means of unbundling property rights is largely mistaken. Merely handing over formal control rights to workers, without giving them real power over property, is largely a paper tiger. It puts workers in a situation where they always have to fear their decisions will be punished by a loss in the value of their company and a greater difficulty obtaining loans, which ultimately threatens their company with bankruptcy and a loss of their source of livelihood. Even if shareholders have lost access to the company’s boardroom, with no formal authority over workers, they can rely on the shared knowledge of the threat of disinvestment. As workers depend on investments, they are typically forced to preemptively fulfill the presumptive goals of the shareholders. The workers of a formally democratic corporation are thus structurally in position to be compelled to almost read the capital suppliers’ wishes of the lips. It is almost as if they were still working in a capitalist company: Day in, day out, they are compelled to toil to satisfy the shareholders, either by maximizing dividends or increasing the stock market value.

However, although this section raised concerns about the separability of ownership and control, it does not foreclose the possibility that, under certain conditions, some degree of unbundling ownership rights might be both viable and desirable. I return to this idea below, not to endorse it wholesale, but to explore whether and how it might be adapted to serve more democratic ends.

4 Three Models of Social Ownership

If democratizing the corporation is to live up to its claim, if it is to become more than an empty phrase, then democratic decision-making within the corporation must be immune to the disinvestment threats of private shareholders. Given that private ownership is a major source of the problem, social ownership of corporate assets might offer a way out. In fact, social ownership can function as a material basis for protecting and asserting democratic self-governance within the corporation. There is a more defensive and a more offensive aspect to this: Defensively, social ownership provides a safety buffer against the threat of disinvestment, offensively it can function as a direction device for steering resources towards important public goals and values, particularly those that go beyond efficiency, such as distributive justice, human rights, or sustainability.^[9]

Social ownership is not equivalent to central planning or state socialism, it is compatible with market coordination of economic transactions, and it doesn’t necessarily imply direct state ownership of enterprises, since there are many varieties of socialized ownership below the level of state ownership. In particular, there are three important forms of social ownership: cooperative ownership, worker self-management and socialized stock markets. This not an exhaustive list, there are many more forms of social ownership, but these are the most important and promising ones.^[10] As the brief review of all three models will make clear, even though they each have distinct flaws, they bear the potential to complement one another in a hybrid system.

(i) Cooperative Ownership: Cooperative ownership is one of the most important forms of social ownership. In cooperative ownership, the company’s capital is jointly provided and controlled by its workers, which effectively socializes capital from the bottom up. At the same time, this usually requires every firm member to purchase cooperative shares, often paid as membership fee, which constitutes an essential basis of the company’s capital. Building on the Spanish Mondragon cooperatives, Ellerman (1990) developed one of the most well-elaborated models in this tradition.^[11] Cooperative ownership provides a far-reaching material foundation for democratic self-governance, as democratic decision-making power can extend to a wide range of crucial domains, including working conditions, production goals, profit objectives, hierarchies within the firm, income differentials, and more. It provides a material basis for democratic self-governance within the corporation, because it disables the threat advantage of external investors, and it does so quite simply: There are no external investors anymore who could threaten to disinvest. In cooperative ownership, democratic control rights and ownership converge to one of the highest possible degrees. Cooperative ownership also means that the opportunity for capitalists to exploit workers is overcome from the outset, almost a priori, since the division and asymmetry between capital and labor is sublated.^[12] Finally, cooperatives are also less profit-driven than stock corporations, as their production activities are more strongly oriented toward use-values, and they are more deeply embedded in local communities, since control rights over property do not rest with distant shareholders. All of this means that cooperatives typically fare better than stock corporations in terms of their ability to take into account important public goals and values, such as sustainability.

In principle, all of this seems to speak in favor of this corporate form, which naturally leads to the expectation that cooperative ownership would be all over the place. Yet, this is not the case, in capitalist economies cooperatives are quite rare. Sure, a significant part of western capitalist markets is organized cooperatively, ranging from credit cooperative via housing and consumers’ cooperatives through to processing and utility cooperatives, but especially worker cooperatives are relatively seldom, hence the title of a Jon Elster’s seminal paper: ‘If Cooperative Ownership is So Desirable, Why are There So Few Cooperatives?’ (Elster 1989). At this point, the systematic worry must be that this is due to internal flaws of this corporate form, to which I will return in the next section. In addition, it is telling that there is only very limited number of proposals for generalizing this form of ownership, that is, society-wide models of cooperative ownership.^[13] One of the most daunting challenges in this respect is that cooperative ownership might require every member of society to buy cooperative shares, a notion which immediately seems profoundly illiberal.

(ii) Worker Self-Management: Worker self-management is the best-known model of social ownership in the market socialist tradition, most prominently elaborated by Schweickart (2011), and sharing similarities with Yugoslavia’s economic system.^[14] In market socialist economies, corporations compete on the market, but they are not in private ownership, and there are no external shareholders, instead they are publicly owned. Under worker self-management, enterprises secure their funding from public banks, while paying a leasing fee for their funds, proportionate to the capital assets they oversee. In a certain way, the fee is similar to a “flat-rate tax on the capital assets of all productive enterprises” (Schweickart 2011, 57). While belonging to society collectively, the socialized enterprises are governed by its employees, not by the state, and certainly not by shareholders, since stock ownership is abolished. Here, too, the threat advantage of external investors is neutralized. In contrast to comprehensive planning, the socialized enterprises of the market socialist model engage in market transactions, they operate on the market to buy the inputs they need and to sell the products and services they offer.

Without going into detail, it can be said that worker self-management comes with several problems, particularly false incentives both for workers and for the state/public banks. A key issue in market socialism, in terms of false incentives for the state or public banks, is the concentration of power. While enterprises are nominally owned by society as a whole, in practice, this means that the entire productive assets of society are under the control of centralized institutions, such as public banks or the central government, which are managed by a small group of experts. The legitimate liberal concern is that power corrupts, and that control over productive resources is one of the most dangerous sources of such power (Friedman 2020, 26–33). Additionally, the well-known problem of soft budget constraints, identified by economist Kornai (1980), occurs when governments protect failing enterprises, especially large ones, to avoid social instability or political fallout. This reliance on bailouts reduces the incentive for enterprises to improve efficiency, while also straining public finances and misallocating resources. In terms of false incentives for workers, there is the issue that under worker self-management, workers benefit from public resources without being personally affected by business failure or the loss of capital. This can lead to overly risky or negligent business conduct. Thus, the public banks, as major lenders, have reason to be concerned that workers will use their decision-making authority opportunistically (Thomas 2017, 223–46).

(iii) Socialist stock markets: Socialist stock markets might sound like a contradiction in terms, and even outside of a socialist framework, there are numerous reasons for skepticism toward stock markets. However, there are influential models of socialist stock markets, including those by Roemer (1994) and Corneo (2018), which are based on the assumption that there are important reasons for allowing some form of stock ownership: Stock markets gather and convey information about the economic performance of companies in a decentralized way, which reduces the need for bureaucratic monitoring and control by a centralized institution.^[15] The concentration of economic power in centralized institutions, particularly public banks, is substantially reduced, something that I highlighted as a key disadvantage of the worker self-management model. In addition, the seductions of overly soft budget constraints are held in check, since the stock market puts discipline on unprofitable and inefficient firms to improve their performance.^[16]

One of the most well-known models was developed by the economist Roemer (1994). It operates with dual-currency system, introducing a distinct ‘currency’ for stock shares. Since the model is somewhat complex and technical, I focus on another model that retains the stock market in its current form, avoiding the complex institutional overhauls required by Roemer’s model. Giacomo Corneo developed a model for socialist stock markets that places special emphasis on public ownership (Corneo 2018). Corneo’s model requires the state, in the form of the ‘Federal Shareholder’, to hold a majority stake in all publicly traded stock corporations of a given society, that is, a stake of at least 51 % of the total equity of each stock corporation. By holding a majority stake, the democratic public, represented by the Federal Shareholder, would have a substantial influence over decision-making processes within stock corporations. Although private shareholders still wield significant influence over corporations – because of their ever-present threat of selling shares – their threat advantage is diminished to a considerable degree. Moreover, the Federal Shareholder works as a direction device, steering corporate activities toward the common good and aligning them with important public values such as distributive justice, sustainability, or human rights. Fundamentally, society would have much broader opportunities to decide on the surplus generated in the economy as a whole.^[17]

Again, without going into detail, it can be said that the idea of socialist stock markets comes with several problems. Fundamentally, it allows for too much scope both for stock markets and for centralized state power, at the expense of worker ownership and democratic decision-making at the workplace. Giving the state a majority stake and control rights in all stock corporations leads to an enormous concentration of power in the hands of the state. At the same time, the logic of stock markets remains untouched to a considerable degree, which is troublesome because of the many normative defects of stock markets: Among these normative defects is the tendency of stock markets to foster a narrow focus on bare numbers among shareholders, that is, on maximum dividends and stock value increases, which pressures management of stock corporations to deliver immediate returns to shareholders and to prioritize profits over a whole range of important public values.^[18] Stock markets are also said to encourage speculative tendencies and herd behavior, which leads to market bubbles with strong price fluctuations upwards and sudden crashes afterwards. Finally, stock markets subject those who toil in the factories to the dictates of blind market forces. Their fate depends on the narrowly self-interested decisions of an amorphous mass of isolated shareholders, rather than allowing workers to take democratic control over the conditions in which they work.

5 Challenges of Cooperative Ownership

The comparative analysis of the three most important forms of social ownership has shown that cooperative ownership is the most promising model, both compared to capitalist firms and to competing models of social ownership. It fares best in terms of enabling democratic corporate governance, particularly because it effectively disables the disinvestment threat of private investors and ensures a strong convergence between control rights and ownership. Moreover, it socializes capital from the bottom up, thus avoiding over-centralization of control over collectively owned assets, a problem that affects both Schweickart’s worker self-management model and Corneo’s model of socialized stock markets. At the same time, members of the cooperatives acquire a personal stake in the company, preventing the emergence of false incentives, which plagues worker self-management.

Now the problem is that, despite its many advantages, cooperative ownership remains relatively scarce (Elster 1989). If cooperative ownership is so desirable on normative grounds, why is not the preferred form of ownership for most empirically existing corporations? At this point, the systematic concern must be that the scarcity of cooperatives results either from inherent flaws in this corporate model or from legitimate aversions to it.

The response must be sufficiently nuanced. There are in fact unfair disadvantages for cooperatives under capitalism, but it must also be acknowledged that there are inherent problems with this particular corporate form. One obvious form of unfair disadvantages for cooperatives is capitalism’s “wealth inequality constraint” (Bowles and Gintis 1993, 96): Those who could benefit most from cooperative membership can afford it the least. Under conditions of class inequalities, most people simply don’t have the money to afford to buy the cooperative shares, which, as noted before, may cost up to two annual incomes (Thomas and Logan 1982, 149). So, there is a sad irony: workers with no assets, who are forced to sell their labor to corporate giants, have the greatest objective interest in becoming cooperative members, yet they are the ones least able to realize that interest.

Under a system of universal inheritances, things would be different, it would give people the real opportunity to become a member, that is to acquire a member share. Yet, even then, there are obstacles to cooperatives: One important obstacle is that company founders usually reject the cooperative form for reasons of self-interest. They neither want to share control nor profits (Ben-Ner 1988, 290). In capitalist societies, the vast majority of company founders chooses a corporate form that allows for hiring and firing employees easily, which, from their perspective, is important in times of strong company growth or sudden economic slumps. So there is a “hiring bias” (Malleson 2015, 83) in capitalists markets, a bias in favor of hiring employees rather than recruiting co-members or co-owners. There is no reason to think this would be any different in a property-owning democracy. This means that capitalist companies tend to be more prevalent than cooperatives simply because they confer special advantages on a particular group of individuals, namely company founders who can monopolize profits and control, even if capitalist companies and cooperatives are on equal economic footing, for example, in terms of productivity. Therefore, the paucity of cooperatives is not necessarily due to their own flaws or the superiority of their capitalist counterparts; rather, it may be the result of self-interested decisions made by a minority. This is not meant to criticize the self-interested behavior of capitalists as inherently unfair, but rather to highlight how the actions of a tiny elite can produce substantial and unfair disadvantages for cooperatives.

Yet cooperatives not only suffer from unfair disadvantage, there are also problems intrinsic to this particular corporate form. A major problem is financing. Usually, cooperatives rule out issuing shares to external investors, since they would assume partial ownership of the company, and thus partial control over it, yet without working in it. Cooperatives primarily raise capital by gaining new members or by retaining their earnings. But gaining new members is much more difficult than issuing shares. Cooperatives face the difficulty of having to find new members who are willing to pay the membership while at same time being willing to work in the company. This is obviously miles away from the situation a stock corporation faces (Dow 2003, 249–53). It can easily find people from around the globe to buy shares, and buying a share can be done almost without moving an inch, for example, simply by clicking on a trading-app.

Apart from raising capital by attracting new members willing to buy cooperative shares and using retained earnings, cooperatives finance themselves through debt, either in the form of loans from banks or tradable bonds. But compared to issuing shares, this form of financing comes with many drawbacks as well. The most obvious reason is that debt has to be paid back, while issuing shares invites new co-owners into the business, thus requiring no repayment. Moreover, stock ownership provides greater flexibility: Issuing shares does not come with immediate financial obligations for the company, as dividend payments are determined by the company itself, while debt requires fixed interest and repayment rates. Given typical market fluctuations and uncertain profit expectations, the greater flexibility that stock ownership provides is a significant advantage (Williamson 1985). Finally, banks are often hesitant to offer loans, depending on the availability of a collateral to secure the loan, while issuing shares does not require collateral (Dow 2003, 186–9). Stock corporations, therefore, enjoy a much greater flexibility to secure their financing compared to cooperatives. As a result, the prevalence of cooperatives significantly depends on capital intensity: The higher the capital intensity, the less likely the cooperative form, a pattern which is confirmed empirically (Dow 2003, 118). Limited access to a wide range of sources of financing is a distinct disadvantage for cooperatives.

There is yet another intrinsic disadvantage of cooperative ownership, which is known as the portfolio problem (Jensen and Meckling 1979): Cooperative membership requires investing in cooperative shares. The members must invest a large portion of their assets where they are employed. Their wealth is largely tied to a single investment, which in turn is tied to their workplace. If the cooperative fails, members lose both their financial investment and their job, their source of regular income. It is, of course, in the legitimate self-interest of individuals to avoid such a heightened risk, that is, the risk of simultaneously losing both their savings and their livelihood. Therefore, there are at least to some extent legitimate aversions against becoming a member of cooperative.

6 Plural Cooperativism

We’ve seen that, in a capitalist context, cooperatives suffer from unfair disadvantages, but we’ve also seen that there are problems intrinsic to this particular corporate form. Given that cooperative ownership is still strongly desirable on many grounds, most of all in terms of democratizing corporate governance, is it possible to find an institutional arrangement where cooperativism is universal? Is there a model that universalizes cooperative ownership while also mitigating its internal flaws? Plural Cooperativism is an attempt to provide an affirmative answer. It is a hybrid model that combines the three forms of social ownership, namely cooperative ownership, worker self-management and socialized stock markets, while also leveraging the specific advantages of the unbundling approach. The goal is to offset the particular weaknesses of each model while mutually reinforcing their distinctive strengths.

The term ‘Plural Cooperativism’ already suggests that cooperative property is at the center of the hybrid. But the model is also ‘plural’: It is plural in a particular sense, allowing for, and systematically combining, a variety of ownership forms. Plural Cooperativism is based on cooperative ownership, but it may include limited forms of private stock ownership, complemented by public stock ownership.^[19] It envisions a structured diversity of ownership forms: (i) The Cooperative Foundation: The center of the model is cooperative ownership. The cooperative form is universalized, which means that business corporations basically operate as cooperatives. (ii) Strong Public Banks: Cooperative firms are supported by strong public banks, similar to Schweickart’s model of worker self-management. (iii) Conditional Private Shareholding: There is the limited possibility of capital supply by private outside investors. But this possibility depends on two conditions: (a) Public Counterbalance: As soon as private shareholders participate in financing cooperatives, the public sector joins in as well, modelled on Corneo’s Federal Shareholder. (b) Limited Control Rights: The power of private shareholders is additionally restricted by curtailing their formal control rights, following the reallocation of control rights advocated by the unbundling-approach.

In the following, I will go through features (i)–(iii) step by step. But already at this point it is evident that under Plural Cooperativism, plurality is something else than amorphous arbitrariness; it is the opposite of an anything-goes pluralism that would allow for any number and combination of ownership forms. Rather, plurality means a structured and systematic form of diversity that brings the relevant forms of ownership into a desirable and stable balance. This blocks the possibility that the cooperative form is undermined or crowded out by conventional capitalist firms, which would inevitably happen in a free-for-all scenario due to the disadvantages cooperatives face compared to capitalist firms, disadvantages that include, as we’ve seen, significant unfair disadvantages.^[20]

(i) The Cooperative Foundation: The center of the model is cooperative ownership. The cooperative form is universalized, private businesses basically operate as cooperatives. This does not mean there are can be no other forms of socially owned businesses. There is surely room for publicly owned corporations, as well as corporations under steward or trust ownership. However, capitalist enterprises, whether privately held businesses or stock corporations, are abolished. They are replaced by cooperative forms of business.^[21] Cooperative ownership should be the regulatory ideal because it allows for the strongest form of democratic self-governance of the firm, most importantly by eliminating the disinvestment threat posed by external investors. Compared to worker self-management under market socialism, cooperative ownership comes with a number of advantages: An obvious advantage is that we are already familiar cooperative ownership since it exists and functions sufficiently well in the midst of our capitalist market economies; in comparison, the track record of worker self-management under Yugoslavian market socialism is mixed at best, and quite difficult to assess because of the unique historical and political circumstances (Estrin 2010). Another important advantage of cooperative ownership is that capital is supplied by the members themselves, so that property is socialized from the bottom up; in the self-management model, by contrast, workers largely depend on the favor of public banks for their capital supply. Moreover, within the cooperative framework, members of the cooperative acquire a personal stake in the company, a direct involvement, which prevents the emergence of false incentives that the self-management model provokes. Members of cooperatives are personally affected by business failure and the loss of capital, which means they have a direct incentive to avoid overly risky or negligent business conduct.

Now, when the cooperative form is universalized, a gigantic challenge arises. It seems that this requires every worker to buy cooperative shares, something many may perceive as deeply illiberal. This is the Achilles’ heel of universalizing the cooperative form. There is no way around making financial contributions to the cooperative fund mandatory. That is, each member of society is obligated to contribute parts of his or her personal wealth to the cooperative that he or she joins. As noted above, the required amount can be quite substantial, potentially ranging to as much as two full years’ income. The shares the members of a cooperative are required to purchase carry the risk of loss; if a cooperative fails, its members’ investment is gone. This means that underpinning workplace democracy with social ownership quite literally comes at a price.

As we shall see, Plural Cooperativism is specifically designed to reduce the required number of cooperative contributions. This is achieved through feature (ii), Strong Public Banks, and feature (iii), Conditional Private Shareholding, both of which will be explained in a moment. Nevertheless, substantial contributions to the cooperative fund are still necessary, and mandating these contributions continues to pose a significant burden.

Recall, however, that I presupposed a property-owning democracy as a minimal starting point, with a system of universal inheritances (or capital grants) at its core. Unlocking the potential of universal inheritances paves the way to overcome the conundrum. To reduce the burden of mandatory contributions, we can envision a simple mechanism for using the inheritances or grants: A significant portion of the universal inheritance that every citizen receives is earmarked from the outset. It can only be used for buying cooperative shares in the particular cooperative one is about to join. Earmarking a portion of the universal inheritance in this way should be seen as the proper form for society to pass down a share of its accumulated collective wealth to younger generations. As beneficiaries of this wealth, the younger generation is, in turn, entrusted with the obligation to contribute to the cooperative venture of social cooperation – and to do so, quite literally, within the framework of a cooperative enterprise.

However, despite the fact that contributions to the cooperative fund can be financed by universal inheritances, the price tag for underpinning workplace democracy with social ownership remains significant. But there is no way around universalizing the obligation of contributing to the funds of the various cooperatives. We must come to see it as a shared collective obligation that we impose on ourselves for good reasons. There is nothing intrinsically illiberal about self-imposed collective obligations, they are part and parcel of liberal democracies. They come in many forms, such as tax duties, mandatory education, mandatory health insurance, conscription etc., and they can be justified when they are necessary to provide significant public goods or realize fundamental values, such as a functioning statehood, education, universal healthcare, or military defense. Democratic corporate governance is such a fundamental value. When we impose collective obligations on ourselves for good reasons, they are not a restriction of freedom but rather an expression of it.^[22]

(ii) Strong Public Banks: The cooperative firms are supported by strong public banks, similar to Schweickart’s model of worker self-management; they make affordable loans available to everyone, which reduces the size of the required cooperative share significantly. Strong public banks are nothing outlandish; in Germany, the KfW serves as a prime example of a publicly owned investment and development bank. Since cooperative firms under Plural Cooperativism have access to various funding sources, including member contributions, private capital, and public funds, the role and influence of public banks diminish significantly, which limits the power of centralized institutions. Moreover, because members of cooperative firms have a personal stake in the company, their interests align more closely with its success, which discourages risky or negligent behavior at the expense of the public.

(iii) Conditional Private Shareholding: Plural Cooperativism allows for the limited possibility of capital supply by private outside investors, depending on two conditions, namely (a) A Public Counterbalance, as well as (b) Limited Control Rights for private investors. Before explaining (a) and (b), it must become sufficiently clear why to allow for private outside capital in the first place. As we’ve seen, despite the normative defects of stock markets, there are several good reasons to provisionally accept this institution, provided that it is democratized by a significant degree of public share ownership: The concentration of economic power in centralized institutions is substantially reduced, and the temptations of overly soft budget constraints are contained.

Moreover, we should take into account additional reasons for allowing the possibility of private outside capital, reasons that are particularly important because they are grounded in the interests of the members of cooperatives themselves. It should be quite obvious by now that allowing some form of stock ownership makes it much easier for cooperatives to raise capital. Those who invest in the firm are not required to also work there, so their investment can come from almost any place at any time, unfettered by geographical constraints and social ties to a particular place. Allowing some form of stock ownership also means that the members of cooperatives can pass some business risks on to shareholders. They can pool their business risks much more broadly, which means their own risk decreases (Corneo 2017, 168–97). Conversely, the portfolio problem faced by members of cooperatives is significantly reduced. Since allowing external shareholding decreases the required number of cooperative shares held by members, they can invest a larger part of their resources elsewhere in order to spread, and thus reduce, risk. These are some of the reasons that speak in favor of stock markets even from the perspective of cooperatives, and in fact, in some countries cooperative law has been amended to allow for minority equity-holding by external non-members.^[23]

Let us now turn to the two limiting conditions for allowing private outside investments:

(a) Public Counterbalance: As soon as private shareholders participate in financing cooperatives, the public sector joins in as well, modelled on Corneo’s Federal Shareholder. Public shareholding by the Federal Shareholder counterbalances the power of private shareholders, and it ensures that the public interest is asserted against the narrow interests of private shareholders. Thus, cooperative ownership forms the backbone of the model, and when private shareholding occurs, public shareholding establishes a balance of power. Social ownership in its various forms maintains superiority over private ownership.

(b) Limited Control Rights: The second condition for allowing private shareholding pertains to control rights: The power of private shareholders is additionally restricted by curtailing their formal control rights. This is where the unbundling-approach comes back into play. Private shareholders are granted only minority control rights. The percentage of shareholder representation in corporate governance could lie somewhere between Ferreras’s and Jay’s models; between Jay’s proposal of a complete withdrawal of control rights and Ferrera’s two-chamber model, there is ample scope for different possibilities: In principle, minority representation of shareholders could range somewhere between one and 49 percent. So, in case a cooperative’s equity consists of 40 % from private shareholders, it would be counterbalanced by 40 % public share ownership. Yet still, under the model, it would be entirely possible to permit only a 20 % minority representation of shareholders on the cooperative’s board. As a result, shareholders would have lost their formal dominance in the company, yet at the same time, they would not have lost any kind of control right. Contrary to what Jay’s model suggests, granting at least some control rights to private shareholders is important: Private shareholders surely have a legitimate claim to have some say in what happens with their wealth.

Plural Cooperativism demonstrates that cooperative ownership can be complemented by other forms of social ownership, both by assigning a major role to public banks, and by allowing a well-balanced combination of private and public stock ownership. To recap the argument: the major point simply is that there are strong reasons for democratizing corporate governance, yet in order to achieve its immanent claim, democratic corporate governance must be underpinned by social ownership. The analysis of the various forms of social ownership made it clear that cooperative ownership must be the major way to realize social ownership. Fortunately, however, nothing requires cooperative ownership to carry all the load on its own.

The fundamental rationale for underpinning democratic corporate governance with social ownership was to shield democratic businesses from the disinvestment threats posed by private shareholders. Ultimately, the question is whether Plural Cooperativism is achieving the goal it has set for itself: Since Plural Cooperativism does not completely abolish stock markets, private shareholders still wield material power over the firms in which they invest. They can still threaten to sell their shares. Given the persistence of this threat, can Plural Cooperativism hold up in the face of disinvestment threats from private shareholders? The answer is Yes. When private shareholders threaten to disinvest, the cooperative and public capital with which corporations are equipped serves as a safety buffer. Even when private shareholders withdraw and the share price plummets, cooperatives have much less to worry about in terms of securing access to new financing. They can rely on their own and public resources, making their creditworthiness and ability to borrow funds far less dependent on the whims of shareholders. As a result, shareholders largely lose their threat advantage.

At the same time, in Plural Cooperativism, stock markets still exist. They still have a function to play: And this is precisely to convey information about the economic performance of companies, and to put some discipline on unprofitable and inefficient firms. This means that a sell-off by private shareholders might give cooperatives reason to change direction. Therefore, a massive stock sell-off leads to a zone of ambiguity. Is it a sell-off for illegitimate reasons, e.g., because the cooperative rejects exploitative behavior? Or does the sell-off occur because shareholders have identified genuine inefficiencies within the cooperatives?

The point of Plural Cooperativism is to politicize this ambiguity in the right way. In the event of a widespread stock sell-off, it enables a democratic response grounded in normative principles. Two scenarios can be schematically contrasted:

1. Sell-off for illegitimate reasons, e.g. because the cooperative rejects exploitative behavior: The company still has a solid capital stock, and the Federal Shareholder might have to step in as countervailing force by providing additional resources.

2. Sell-off due to internal inefficiencies: The cooperative has reasons to change course, and the Federal Shareholder has reasons to push for it.

This does not mean that finding a democratic response is easy. The situation will be complex and contested, with many voices contradicting one another. However, the point is that Plural Cooperativism makes it possible in the first place to challenge and push back against the anonymous power of investment decisions by private owners, allowing for the possibility of shaping the market dynamics democratically, not simply from the outside and ex post, but from within.

7 Challenges and Objections

One thing is almost crystal clear: Plural Cooperativism implies certain limitations on personal autonomy and is quite likely to lead to some efficiency losses compared to capitalist corporations, normative drawbacks that might be viewed as particularly troublesome, since personal autonomy and efficiency are often regarded as among the most important values for justifying the market economy. Even more troublingly, these values are also central to several influential theories of justice, most notably Rawls’s theory of justice, which remains one of the most widely discussed frameworks in contemporary political philosophy. I use it here as a minimal starting point, providing the basis for my argument that certain limitations on personal autonomy and some efficiency losses do not, in themselves, constitute decisive objections, at least within the framework of Rawls’s theory of justice.

To start with personal autonomy: To what extent is it limited under Plural Cooperativism? Note, first, that Rawls – for good reason – explicitly excludes a right to private ownership of the means of production from the general right to private ownership (Rawls 2001, 114–5), and for him, freedom of occupation does not entail a right to wage labor in capitalist enterprises.^[24] Moreover, under Plural Cooperativism, everyone has the right to establish a cooperative at their discretion, as well as the right to choose freely which cooperative to join. All this means that Plural Cooperativism leaves significant room for personal autonomy and basic liberties. However, it is true that Plural Cooperativism constrains freedom of occupational choice in some way, since capitalist ownership is banned, and the democratic, cooperative firm universalized. The freedom to run a capitalist enterprise or to work in such an enterprise is abolished. In addition, mandating contributions to the cooperative fund is a considerable burden, which restricts one’s personal autonomy significantly. This is true despite the fact the mandatory contributions can be financed, in part, by the universal inheritances. Compare this to capitalist markets, at least when embedded in a property-owning democracy, where wealth and capital is widely dispersed among all citizens: Capitalist markets certainly allow for a greater elbow room to do as one pleases, for they allow, among many other things, to become a capitalist as well as a wage laborer, while not even ruling out becoming a member of a cooperative. In Rawlsian terms, this means that Plural Cooperativism constrains the first principle of justice to a certain extent, because, on the face of it, capitalist markets allow for a more “extensive basic liberty” (Rawls 1999, 53) than Plural Cooperativism.^[25]

Some efficiency losses will likely occur under Plural Cooperativism, since the allocation of capital is subject to the challenging normative demand of balancing cooperative, private and public ownership. This obviously hampers efficiency, while under capitalism, capital allocation runs comparatively unfettered, left to the impersonal forces of supply and demand. Another reason for efficiency losses is that democratic procedures in the cooperative enterprises may well be somewhat more time-consuming than the procedures in non-democratic, capitalist firms. In Rawlsian terms, however, efficiency losses affect the difference principle, which requires making the worst-off members of society as well off as possible; this goal may be constrained when economic efficiency is not maximized.^[26] So, in addition to constraining Rawls’s first principle of justice, Plural Cooperativism constrains the second principle of justice as well.

It is important to see, however, that Rawls himself is unequivocal in emphasizing that institutions must only be just, they must also foster a sense of justice.^[27] This is Rawls’s question of stability. Rawls underlines quite clearly that the requirement of cultivating a sufficiently strong sense of justice is of central importance and might even require revisions of the principles of justice. This idea must be connected to argument from democratic virtues mentioned at the beginning: Firms are an especially important site of forming habits of thought and action which may or may not strengthen the sense of justice. Consider two economic systems: One system does well in nurturing a shared of sense of justice, by fostering the habits of thought and action conducive to democratic justice, while the other system undermines it, by fostering attitudes such as market egoism, corporate authoritarianism, and fatalism in the face of blind market forces.

It is not hard to see that Plural Cooperativism resembles the first system, while the culture of capitalist corporations resembles the second (even in a property-owning democracy). Yet the relative strength of, and the threats to, the sense of justice under various economic systems must be taken seriously in view of the requirements of the stability of justice. If the sense of justice is undermined, the principles of justice, together with their institutional realization, have to be revised, as Rawls clearly stresses: In that case, “the parties [in the original position] must reconsider the principles agreed to and examine whether, on balance, other principles should be adopted” (Rawls 1999, 89). So, it is entirely possible that the limited and highly specified constraints to personal autonomy and economic efficiency under Plural Cooperativism are acceptable on the grounds that cultivating a sufficiently strong sense of justice depends on it. The long-term stability of the overall system of a just freedom for all may well take priority over maximizing personal autonomy and economic efficiency at any cost.

No doubt there are many additional challenges and objections. Among them are the following: Is there enough room for innovation under Plural Cooperativism? For example, does it allow for maverick entrepreneurs pushing for highly disruptive innovations (which might turn out to be beneficial eventually)? What about the pressure of global competition? If Plural Cooperativism is established in one society, does it require protectionism and capital controls? There is an especially important question that returns to the issue of personal autonomy, but now from a different angle: Is there any room for wage labor under Plural Cooperativism? Wage labor in capitalist firms is unavailable, since they are abolished. But what about wage labor in cooperatives? I conclude the paper with some remarks on this issue.

People might prefer wage labor over a full membership in the cooperative for many reasons. The most important one surely is that wage labor does not require contributing to the cooperative fund, and entrepreneurial risks are absent as well. Moreover, the opportunity for wage labor offers significant flexibility in acquainting oneself with various cooperatives before committing to membership in a specific one. These are certainly innocent and reasonable grounds for valuing wage labor. However, once some form of wage labor is permitted, a giant collective action problem arises (Elster 1989; Fleurbaey 2008). From a purely self-interested perspective, wage labor may well be preferable to cooperative membership, not only for individual workers (for the above reasons), but also for cooperatives. From the perspective of the cooperatives, wage labor allows for a greater flexibility in hiring and firing workers, and it prevents the dilution of control over the cooperative to ever more members. If both workers and cooperatives decide in a purely self-interested fashion, the cooperative form will be subverted over time. Inadvertently, however, this results in grave normative defects for everyone.

In the literature, the problem of the subversion of the cooperative form is discussed under the term ‘degeneration’: The cooperatives slowly ‘degenerate’ into quasi-capitalist firms, where a small elite of worker-aristocrats – the few remaining members of each cooperative – exploit the numerous wage laborers; in fact, many European cooperatives have a high proportion of wage laborers, sometimes up to 70 percent (Dow 2003, 224). To avoid the collective action problem, a collective commitment device must be in place, for example, allowing wage labor in each cooperative only up to certain percentage. Again, the Mondragon cooperatives may serve as an example: Here, the proportion of wage laborers was for a long time limited to a maximum of ten percent of the workforce (Dow 2003, 59). Yet this approach of tackling the problem merely offers a vague indication of a potential solution, it remains far from resolving it to a satisfying degree. Designing the details of a commitment device that allows for some wage labor without ultimately subverting cooperative ownership is a daunting task, since quite likely, tiny details can lead to major problems. But this is the topic for another paper.