The democratic case against fissuring work

Abstract

This article explores what is problematic about the “fissuring” of work—the breaking up of large work organizations through subcontracting or franchising—from a democratic perspective. After describing the empirical evidence on fissuring, it explores two types of arguments. Arguments from “good work” ask whether fissured work is less likely to offer conditions of such good work. Here, however, one must expect a lot of variety; many arguments can be reduced to general criticisms of bad working conditions rather than offering specific criticisms of fissured work. A more successful type of argument stems from considerations of social cohesion, which concern citizens’ opportunities to get to know each other, fairness in remuneration, and chances for democratic experiences at work. For these reasons, fissuring is undesirable from a democratic perspective, even if one abstracts from its more obvious harms, for example, unjustifiable safety risks. The article concludes with some brief reflections on policy implications.

Introduction

The future of work is an omnipresent topic in public discourse, but has only recently become more widespread in normative theorizing (e.g. Breen and Deranthy, 2022; Jonker and Rozeboom, 2023). While wage justice has long been part of debates about distributive justice in general (e.g. O’Neill, 2017; Piketty, 2014), the lack of accountability and workers’ voice have also become issues of discussion (e.g. Anderson, 2017; Ferreras, 2017; Frega et al., 2019). But in recent decades, many individuals have suffered not only from low wages and lack of voice, but also from worsening working conditions. One dimension of the problem is the precarity of work: more and more forms of work come with low security, fluctuating wages, and unpredictable working times (e.g. Azmanova, 2020; Standing, 2011). Employers often do not offer standard employment contracts anymore, but instead short-term contracts or contracts in which workers are not treated as employees at all, but as independent contractors. The most extreme form is platform workers whose work is mediated by algorithms (e.g. De Stefano, 2015; Dubal, 2020).

But this is just the tip of an iceberg, the underwater part of which is the broader practice of “fissuring” work: breaking up what had traditionally been integrated workplaces for the sake of efficiency (Callaci and Vaheesan, 2022; Goldman and Weil, 2020; Weil, 2014, 2019). Fissuring describes the practice of companies outsourcing work that functionally belongs to its core operations—for example, cleaning in a hotel—to save labor costs and avoid liability, through mechanisms such as subcontracting or franchising. ¹ So far, fissuring has not been discussed in the philosophical debate about the future of work. This article explores its normative dimension from the perspective of democracy.

Some theorists in the liberal-egalitarian or neo-republican tradition have turned to the topics of precarity and “gig work,” considering them as a problematic form of “risk shift” that threatens to dominate workers and destroys their “planning agency” (Bieber and Moggia, 2021). Precarious work undermines workers’ individual and collective agency, and hence also their solidarity (Herranz, 2024). Moreover, by misclassifying workers who should be employees as independent contractors, many companies escape their fair responsibilities (Halliday, 2021). In this article, I contribute to this discussion by, firstly, focusing on the broader phenomenon of “fissuring” work, and secondly, by discussing it from the perspective not only of justice, but also democracy (see also Näsström and Kalm, 2015). This perspective asks, specifically, what is wrong about fissuring happening in democratic societies in which work has, traditionally, played an important role for supporting social cohesion.

I start by providing a more in-depth description of the phenomenon of fissuring, based on the groundbreaking work by David Weil (2014, 2019, 2020) and contributions by other social scientists. Fissuring can be wrong in obvious ways: for example, it often leads to violations of safety standards and puts unjustifiable risks on workers (in ways that often are not only immoral but also illegal, but which can still happen for lack of enforcement). To abstract from these dimensions, I construct a scenario that contains fissuring, but not in an obviously problematic way. This allows for exploration of some of the subtler ways in which fissuring is in tension with democratic principles. Doing so raises the question of what it would mean to evaluate the organization of work not just from a perspective of efficiency, but also from a perspective of democracy.

I first explore what I call “arguments from good work.” Authors have suggested various conceptualizations of “good,” “meaningful,” or “unalienated” work. What might be particularly relevant are arguments around the development of skills (“excellence”) or social recognition for work. But when explored in more detail, they fall back into general criticisms of bad working conditions, rather than specific arguments against fissuring.

I then explore what I take to be a more successful democratic critique of fissuring: the argument from social cohesion. In more integrated companies, individuals have greater chances of getting to know fellow citizens from other backgrounds and other occupational groups. Moreover, the fact that in integrated companies fairness considerations limit wage inequality contributes to bringing different groups closer to each other in a material way as well. Thirdly, the chances of experiencing democratic values at work are likely to be greater in more integrated companies, even though more research is needed to explore these mechanisms. These three arguments speak in favor of less fissured work even in a hypothetical scenario in which the obvious wrongs of fissuring, as we currently see them, could be prevented. The conclusion reflects on some policy options that could help reintegrate work along the lines suggested by these arguments.

What is fissuring?

It is not a new phenomenon that labor gets subcontracted or that different organizational forms are used to put pressure on workers (see e.g. Hobsbawm, 1962: 50, on subcontracting during the industrial revolution). Since the 1980s, however, fissuring has seen a massive upswing. ² I follow the account by labor law scholar David Weil (2014), who has done crucial work to understand these developments in the US, and has also introduced the term “fissuring.” ³ As Weil discusses, in many industries—hotels, telecommunication, food production, etc.—workers are no longer directly employed by the “lead companies” that carry well-known brand names. ⁴ What gets subcontracted are not specific tasks, for example, occasional repair services, but tasks that need to be undertaken constantly and that concern the core functionalities of these businesses (e.g. room service in hotels) (Weil, 2014: 121).

The workers who carry out these tasks are employed by subcontracting companies (and sometimes further sub-sub-etc. companies) that compete for contracts. Wages and working conditions are worse than they had been, historically, when workers were employed by lead companies directly (see also Freeman, 2014 for data on hotel workers). A related phenomenon is franchising (Weil, 2014: chap. 6), which means that branches of businesses—for example, fast-food restaurants belonging to a large brand—are run by independent owners, but according to rules and protocols strictly and one-sidedly determined by the lead company, and with a fixed percentage of profits going to the latter.

The fissuring of workplaces would not be possible without the use of modern communication technologies. Pre-fissured workplaces were integrated because—as Coase (1937) has famously argued—integrating certain tasks into “hierarchies” was more efficient than buying them in markets: doing so reduced transaction costs (see also Williamson, 1975). Companies could reap productivity gains through economies of scale (producing the same products or services on a larger scale) and economies of scope (integrating other products or services into an existing portfolio) (Weil, 2014: 33). This also meant that many companies developed internal labor markets, “governed by rules that increased skill acquisition, stable methods of advancement, and employee loyalty” (Weil, 2014: 38). Coincidentally, large integrated workplaces also created good opportunities for unionization, because many workers with aligned interests worked on the same premises (Herranz, 2024).

But the transaction cost calculus changed when new technologies created alternative tools for coordinating work: digital tools such as barcodes, GPS, electronic sensors, and other opportunities for ensuring quality standards without directly employing individuals (Weil, 2014: 12, 60–62). ⁵ As Weil summarizes the development (2014: 44): “it has become far less expensive to contract with other organizations—or create new organizational forms—to undertake activities that are part of producing goods or providing services. That alters the calculus of what should be done inside or outside enterprise boundaries.”

In addition to technological changes, there were also legal changes that made subcontracting and franchising possible. As Callaci and Vaheesan (2022) argue with regard to the US, from the 1940s to the 1970s, antitrust law was understood as preventing “vertical constraints” between companies that would, for example, restrain a franchisee to only sell products by the franchisor. Such restraints were, then, understood as “unfairly limiting the freedom of small and medium-size firms and unduly restricting business rivalry at one or more levels of a supply chain” (Callaci and Vaheesan, 2022: 29). In the late 1960s, businesses began to argue against this logic, holding that the use of brand names offered marketing advantages and supported the continued existence of small, privately owned businesses. In the 1970s, the Supreme Court began to relax antitrust rules (Callaci and Vaheesan, 2022: 42). Another relevant area is labor law, and specifically the question of who counts as an employee, which remains notoriously contested (see Weil, 2014: 21–22, 183–199). Even in cases in which labor law is violated, however, the chances of detection are low, and even lower in fissured workplaces (Weil, 2014: 22). This allows companies to get away with widespread practices of misclassification of employees as contractors, often at the cost of vulnerable employee groups, for example, migrants (see also Halliday, 2021).

Companies, driven by capital market pressures, use fissuring for various reasons, all related to the cutting of costs. They can shift the risks of fluctuations in labor demand to other companies or independent contractors; they can benefit from the downward pressure on wages and working conditions in the competitive markets for outsourced services; and they can avoid liabilities and benefit payments connected to employment relations (Weil, 2014: 10–14, see also 77–78). The—typically smaller—companies that provide subcontracted work are often not unionized (Weil, 2014: 77). And crucially, by fissuring work, companies can cut through concerns about wage fairness that exist among the employees of one company. In an integrated company, employees typically compare their wages among each other, and work motivation could be harmed if some wages were lower than others. Historically, this pulled up the wages of low-paid workers, which was not in the interest of employers (Weil, 2014: 20, see also 82–90, see also Bewley, 2002). Instead of lowering the wages of certain groups and thereby undermining morale, companies preferred completely outsourcing such activities—as Weil puts it, “with fissuring, the fairness problems are less acute and wages can be pushed downward” (2014: 90). An example are the low incomes of independent contractors who deliver packages: at Amazon, they receive payments that amount to $5.30 per hour, compared to wage payments as employees, which amount to $23.10 at UPS (which is unionized) or $14.40 at Fedex (which is not) (Weil, 2019: 154–155).

Fissuring thus constitutes a massive shift in the power relations between workers and companies, increasing the profitability of the latter at the cost of the former (Weil, 2014: 8–9). This holds, in particular, if the power between lead companies and other parties—for example, between franchisors and franchisees—is unequal from the start, as when franchisees are recent immigrants who get locked into unfair contracts (Callaci and Vaheesan, 2022: 46–48). Franchisees are often in turn forced to minimize labor costs to remain profitable, which can lead to further violations of labor standards. Those who benefit are the shareholders of lead companies, but also, to a certain extent, consumers who benefit from low prices (Weil, 2014: 158). This is part of a more general pattern: much economic policy in the last decades has focused on benefits for investors and customers, while leaving questions about the impact on workers outside the picture (e.g. Khan, 2017 on antitrust policies).

The extent of fissuring has increased over time, in line with the general trend toward more precarious work. For example, in the US hotel industry, over 80% of staff are today not employed by the lead company (Weil, 2014: 7). “Non-standard” working conditions concern over 25 million US workers (Weil, 2020) and might hold for as much as 19% or more of the workforce (Weil, 2019: 151). While estimates for other rich countries are lower (e.g. Greer and Doellgast, 2017: 199–200 on Germany, or Christophers, 2022: chap. 4, on the UK), there seems to be a wider upward trend. ⁶ Beyond its direct effects on the wages and working conditions in fissured workplaces, there might also be systemic effects, such as downward pressures on wages and working conditions in sectors that are alternatives for relevant groups of employees. For example, employees in an integrated restaurant for whom the only outside alternative are precarious jobs in fast food chains have a weaker bargaining position vis-à-vis their employer than if they had better outside options. Fissuring is likely to be one of the drivers of overall increases in income inequality, which are caused to a considerable extent by wage inequalities between firms (Freeman, 2014: 40; Tomaskovic-Devey et al., 2020).

As it exists today, fissuring is connected to many obvious and normatively uncontroversial wrongs. Danaher (2018) lists it as one of the reasons for why he takes modern work to have become “structurally bad.” Weil (2014, 2019) focuses mostly on violations of health and safety standards, which put individuals in fissured workplaces at increased risks of accidents and even death. Also, as noted above, because of the low rates of labor inspections, many forms of illegal treatment of workers go undetected. And while the details of the concept of “exploitation” are controversial, the low wages and bad working conditions in many fissured workplaces can be described as “exploitative” on different understandings of exploitation (e.g. Reiff, 2014; Vrousalis, 2023), in the general sense of workers’ weak position being instrumentalized to reap disproportional benefits from them. I take many of these issues to be normatively overdetermined, and thus theoretically (though unfortunately not practically) straightforward to address.

To allow a more specific focus on the democratic wrongs of fissuring work, it is useful to work with a hypothetical scenario in which fissuring takes place, but in which the most obvious and egregious harms are avoided. For example, we can imagine that health and safety regulations and their enforcement are identical across companies, so that all workers—from employees in lead companies to self-employed contractors at the end of the chain—are exposed to the same risks. We can also imagine that in such a scenario, extreme forms of one-sided market power are prevented by antitrust legislation, and the misclassification of employees (as franchisees, or as independent contractors, etc.) is overcome by stringent legal rules and consistent enforcement, also making wages more uniform. The question is whether under such idealized circumstances, there would still be something wrong with fissuring. In what follows, I discuss two sets of arguments—one less successful, the other more so—to show that fissuring is indeed problematic from a democratic perspective, at least ceteris paribus.

Arguments from good work

A first set of arguments against fissuring might be thought to come from considerations about “good,” “meaningful,” or “unalienated” work. The background assumption—and connection to democracy—is that democratic societies should ensure decent working conditions for all their members. A democratic society can be understood as one in which free and equal citizens cooperate in mutually beneficial ways (e.g. Rawls, 1971: 313), which requires that all of them can contribute under decent working conditions. The question about “good” or “meaningful” work also stood behind the large public debate about “bullshit jobs” (Graeber, 2013, 2018), even though broader social scientific analysis later showed that few workers actually find their own work useless, and that the greatest discontent with jobs can be found in jobs that have bad management and toxic workplace environments (Soffia et al., 2021).

Philosophers have conceptualized the quality of work in different ways. There is a literature on “meaningful work” (e.g. Muirhead, 2004; Roessler, 2012; Veltman, 2016; Yeoman, 2014) that emphasizes the value of meaningful work for individuals’ subjectivities and sees it as a distribuendum of justice. Gheaus and Herzog (2016), in a pluralistic account, distinguish four categories of “goods of work,” in addition to the instrumental value of providing an income: excellence, social contribution, community, and social recognition. In a recent criticism, Tyssedal (2023) has argued that social contribution should be prioritized in the understanding of good work, while Kadiyali has suggested a revival of a Marxist account of non-alienated work as “self-realization through others—by helping them satisfy their needs” (2020: 555; on alienation see also Jaeggi, 2016). ⁷

A critic might here ask to what extent such accounts of “good work” are marred by problems of perfectionism (for an overview see Wall, 2021). Isn’t it unduly perfectionist to expect all individuals to want to engage in work which, for example, requires complex skill development to achieve “excellence” or “self-actualization”? What about those whose talents or interests do not lend themselves to such work—shouldn’t they have the right to choose work that does not live up to such a standard (but which might have other advantages, e.g. higher pay) (e.g. Kymlicka, 1990: 186–192). In response, one can admit that different accounts may be more or less in danger of perfectionism. But it is worth emphasizing that they all describe standards that are formal, not substantive, and could be realized with regard to very different activities, from carpeting to office work or artistic endeavors.

Secondly, there is objective evidence, for example, from psychological research, about the negative effects of work that does not fulfil these standards (which Murphy quotes, see e.g. 1993: 4, 30, 44–45). ⁸ A certain degree of paternalism might thus be justified, in analogy to the paternalism of health and safety standards, because good work preserves individuals’ well-being and autonomy in the long term.

This second point, however, could (and should!) be taken care of by general laws that protect workers which are, by assumption, in place in the idealized scenario of fissuring that I have described at the end of the previous section, and which excludes the worst effects of real-life fissuring on workers.

The question thus becomes how different dimensions of good work are affected by fissuring. ⁹ This, however, will depend to a great extent on the concrete jobs in question, and on how fissured work is put into practice. I therefore can only offer some general points that concern likelihoods rather than logical necessities.

With regard to self-actualization or the development of skills, it is not immediately clear that fissured jobs are worse than comparable jobs that are integrated into companies. Murphy (1993), for example, argues that dignified work requires a division of labor that does not reduce them to mindless executors of other people's ideas. ¹⁰ To develop complex skills, individuals need to be able to combine “conception” and “execution” (Murphy, 1993: 9). While the price of giving up all division of labor would be unrealistically high, a division of labor in which workers cycle through different tasks (e.g. through batch work) can be both efficient and allow for greater flourishing. Murphy argues against any “naturalness” in the division of labor, emphasizing its conventional and intentional character (Murphy, 1993: esp. Part III).

While a full realization of this ideal would require more institutional changes than just the end of fissuring—for example, self-governing working teams (Murphy, 1993: 229–230)—it seems likely that in many fissured companies, there are fewer opportunities for skills development and the combination of “conception” and “execution.” But ultimately, this is an empirical question. Some opportunities could lie in the fact that companies to which work is outsourced tend to be smaller and less bureaucratic, which could allow, for example, for people to swap tasks in ways that enrich their work. What seems a greater challenge, however, is that the opportunity to develop one's skills by moving from one job to the next may be reduced because there are no internal labor markets in large companies anymore, thus keeping people stuck in rote jobs for much longer than they would like. This connects to questions of inequality, to which I will return below.

Another dimension of good work that many authors refer to, in some formulation or another, is the social recognition of work; the appreciation by those who benefit from one's work. ¹¹ For example, Kandiyali speaks of “producers and consumers completing one another in the production and consumption of the producer's product” (2020, 564). A related thought is that in integrated companies, workers who make an indirect contribution that may not seem very meaningful can nonetheless identify with the final product, as in the cliché of the secretary in a car company who proudly speaks of “our” cars.

To be sure, sometimes workers may prefer not to be seen by others because it is so painful to experience the disdain, and lack of recognition, for their work. Costas (2022, chap. 5), in an auto-ethnographic study of cleaning work, describes the way in which cleaners in a large office and apartment complex avoid encounters with the clients they work for, to prevent what they experience as attacks on their dignity, and instead withdraw into the “invisible underworld” of underground corridors and storage rooms. But if they were part of the same company, the relation between these individuals could, potentially, be different. Individuals could get a better sense of the complementarities between activities and develop respect for the contributions of others. It is one thing to theoretically acknowledge that the work of cleaners, clerks, drivers, etc., is crucial for keeping society going, but another to meet individuals in these jobs on a regular basis, for example, in the canteen or at company events, and to see them as part of a “we.”

Nonetheless, one needs to ask whether this point really speaks against fissuring—or against the devaluing and invisibilization of certain forms of work, and hence the lack of social recognition for these workers, in a more general sense. It seems undeniable that there can also be social recognition for work that cuts across the boundaries of companies: an office worker might gratefully acknowledge the contribution of the cleaner who tidies up her office whether or not the latter is employed by the same company. ¹² When work is fissured, there might, paradoxically, be stronger incentives to fight for its recognition. Costas (2022: chap. 2), in her study, describes how her employer pushed for cleaning to be seen as a trade, to counter stigmatization and to appeal to employees’ pride. It remains unclear to what extent there was an element of genuine care for workers in these attempts, or whether it was merely a cynical strategy for further exploiting them. But we cannot exclude the possibility that the social recognition of workers outsourced into special companies might end up being higher than if this work is done as part of integrated companies, but marginalized within them.

The same probably holds for many other dimensions of fissuring that one might make from a perspective of “good work”: a lot depends on the details of specific cases. It may be true, as a general trend, that the more work is fissured, the more it is subject to market pressures and hence the more we must expect any considerations about “good” work to be neglected by companies to cut costs. But note, again, that there is likely to be a lot of overlap with general criticisms of bad working conditions that are worthy of critique from many normative angles, rather than being specifically tied to fissuring. The more successful types of arguments specifically against fissuring are likely to come from considerations of social cohesion, to which I turn in the next section.

Arguments from social cohesion

Social cohesion is a complex concept, which gets defined and operationalized differently in different lines of research (for an overview see Friedkin, 2004). I here understand it broadly as positive attitudes that fellow citizens have toward each other and that lead to solidaristic forms of behavior, for example, mutual help. While not per definition positive for democracy—for example, certain forms of in-group cohesion may reinforce rejection of other groups—democratic societies arguably need a certain degree of social cohesion. Citizens need to see themselves as bound by the same set of rules, and solidarity is often crucial for them to engage in political action, for example, in social movements.

In modern societies, the workplace is a crucial space for creating social cohesion—for example, by creating “weak ties” that go beyond the “strong ties” of families and small groups of friends (Granovetter, 1973). As sociologists since at least Durkheim (1933 [1893]) have argued, differentiated work can bring individuals together and strengthen social cohesion, because different tasks and roles complement each other. These complementarities are organized either in exchange relationships or in hierarchies, that is, firms (Coase, 1937). Markets, however, do not offer a promising setting for strengthening social cohesion. While they can sometimes create complementarities and hence positive social bonds between buyers and sellers (Bruni and Sugden, 2008), they are organized by the principle of competition among those on the same side of the market. This competition pits people against each other, which is a threat to solidarity (Hussain, 2020).

Hierarchies—or integrated workplaces in general—offer better conditions for building social ties among individuals. ¹³ Herranz (2024: 8–9) provides a useful set of conditions for solidarity to develop: “shared purposes,” “non-rivalry,” and “common interest.” Employees within one company often fulfil these conditions (even though certain competitive elements or divided interests may remain). The term “colleague” describes the specific relationship of individuals who encounter each other on a regular basis—sometimes spending more time together than with family members—and who need to work together to achieve specific goals. In many cases, sharing the same management is an additional experience that ties people together, because they are united by a common opponent. Unionists and other organizers have long known that the close ties and shared interests of colleagues offer the potential for stable and long-term organizing (e.g. McAlevey, 2016).

If the structures of work organization change, then the social ties created by work must also be expected to change, in direct and indirect ways. Bieber and Moggia (2020: 292–295) discuss the effects of gig work on individuals’ ability to form stable social relations. Their focus, however, is on private relations: gig workers are less able to care reliably for their families, they cannot join collective efforts on a regular basis because of changing shifts, and their reliance on welfare may lead to resentment on the part of other members of society (Bieber and Moggia, 2020: 294–295). While these are important concerns and apply to many workers in fissured workplaces as well, they are about social relations outside of work. Herranz (2024: 11–13), in contrast, also considers direct effects, focusing on the problems caused by labor market dualization for solidarity: “dualisation erodes solidarity: the risk profile among various groups is differentiated, making it easier to fracture their interests or impeding them from realising them together” (Herranz, 2024: 12–13). However, fissuring has broader effects than just creating a division between labor market insiders and outsiders (“dualization”). This is why I consider the effects on social cohesion more broadly along three dimensions: epistemic effects, inequality effects, and effects on potentials for democratic experiences. On all three counts, fissured workplaces are likely to score lower than more integrated workplaces.

Epistemic effects

This first dimension concerns the question of what individuals get to know about their fellow citizens through the social connections they make at work. ¹⁴ Do they only get to know others with a similar background, or do they have the chance to meet a wide array of diverse individuals? It is well established that workplaces bring together individuals who would not meet each other in associations that are completely voluntary (e.g. in religious groups or civil society associations), because the latter allow individuals to stay within their own bubble (Estlund, 2003). But this positive epistemic effect is reduced if workplaces are fissured: individuals then tend to have colleagues who are more similar to them in many respects, notably in terms of education and social class. Cleaners have other cleaners as colleagues, accountants, lawyers, and so on.

As such, the fissuring of workplaces has most likely contributed to the lack of mutual understanding of individuals from different groups, and especially lack of understanding for the situation of the poor, which sociologists have identified as one of the great challenges of our societies. As Lamont, for example, writes (2023: 35) “Many have a poor understanding of the world beyond their familiar and homogeneous neighborhood, which also leads them to underestimate the problems facing the poor.” And while she notes that “diversity trainings and supposedly inclusive hiring practices have often had disappointing results” (Lamont, 2023: 161), she recommends mentoring and genuine efforts (accompanied by financial incentives) for diversifying the workforce as one of the strategies to counter these tendencies (Lamont, 2023). But again, fissured workplaces here present obstacles: if mentoring programs take place within them, they will only create ties between individuals who are in similar occupational situations (even though, admittedly, their racial, religious, political, etc., background may still differ).

In highly fissured labor markets, individuals have no opportunity to learn about the situation of others from different social classes in their daily routines (and the lack of geographical integration and of means of mobility that often comes with it does not help either). For example, they may have no idea what it is like to live without a bank account, a challenge that many precarious workers experience (Baradaran, 2018). This may also contribute to a lack of awareness of own privileges on the part of members of the middle class, who may feel poor compared to the managers whose lifestyle they are aware of, while they are not confronted with the realities of, say, the canteen workers who prepare their food, but whom they never encounter in shared spaces or at company events.

If one assumes that workplaces are also places in which individuals meet friends and partners, then the harmful effects of fissuring on social cohesion, across different social classes, becomes even more visible. Often, friendships between parents also lead to encounters between children, who get to know different potential role models who might influence their own occupational choices. If circles of friendship remain tied to people from similar backgrounds, this is unlikely to have a positive effect on the social mobility of the next generation. Here, a critic might object that as long as wages remain as unequal as they currently are, integration into the same organization might not be sufficient to draw people toward each other also in their spare time. But in response, the next argument becomes relevant, namely that more integrated workplaces might in fact also lead to more material equality.

Inequality effects

As described above, one of the reasons for why fissuring is attractive for companies is that it allows them to break through the fairness norms about the relation between wages for different jobs, the violation of which within one organization can undermine work morale (Bewley, 2002; Weil, 2014: 82–90). From a democratic perspective, however, such fairness constraints are valuable: they indicate that citizens have certain expectations about the fair remuneration of different activities that contribute to a joint project. Existing forms of fissuring are often in deep tension with democratic values, in the sense that individuals get treated in ways in which the members of a democratic society should never treat each other: even basic work standards are violated and individuals are exploited in material terms. But even an idealized scenario in which the worst excesses are absent looks problematic from a democratic perspective: individuals’ work is remunerated exclusively according to their market power, without any attention to, for example, the special burdens of certain forms of work.

It is by now well known that the relation between gains to labor and gains to capital has moved in the direction of the latter over the last decades (e.g. Piketty, 2014). This has contributed to the high working time that individuals, especially in the US, need to invest to cover the costs of living (e.g. Eichner, 2020; Schor, 1992), which also leaves them with less time to engage in civic associations or other social activities. Recent empirical work shows that in many countries, wage inequality between rather than within firms is an important driver of overall inequality, with the authors explicitly mentioning the role of “organizational strategies of outsourcing, franchising, subcontracting, and the like, as well as more individual and firm-level wage bargaining” (Tomaskovic-Devey et al., 2020: 9281, see also Rosenfeld, 2024). ¹⁵ While one may disagree on some details, it seems clear that overall, ever-growing inequality is problematic from a democratic perspective: poverty makes it more difficult for some parts of the population to get engaged in political activities at all, while those with high amounts of income and wealth can exercise disproportionate influence on politics (e.g. Page and Gilens, 2020), undermining the legitimacy of democratic procedures. Moreover, the fear of falling into poverty and precarity can have wide-ranging effects on the lower- and middle-income strata of societies. It is one factor that has been identified as causing the recent rise of right-wing populism in many democracies (Häusermann, 2020).

In more integrated workplaces, low wages would probably move up again—in what are higher “costs” for companies from a purely economic perspective, but which mean less exploitation and better life chances for workers in those positions. One might here object that a better strategy would be to let companies decide about fissuring strategies, according to principles of efficiency, and to redistribute incomes through taxes and transfers; the traditional strategy of social democratic welfare states. ¹⁶ But this approach encounters a number of problems: it meets the political resistance of those who reject taxation and perpetuates a problematic narrative about those who are “unproductive” and need “help” from the state. Instead, as part of a “predistributive” strategy (Hacker, 2011), the very ways in which low wages come about should also be addressed. Given that fissuring is here a key factor (though certainly not the only one), it needs to be included in the discussion.

The increase in inequality caused by fissuring also concerns the distribution of risk, with many individuals working in fissured industries today experiencing far greater insecurity (see also Bieber and Moggia, 2021: 284). Authors have noted the “great risk shift” that has made the life of the average individual and family far more uncertain than in the past (Hacker, 2019, see also Näsström and Kalm, 2015 for a critique from the perspective of democracy). This shift, however, concerns not only the decline of welfare state provisions, but also the structure of companies. Large, integrated companies could serve as buffers of risk in the past, in the sense of compensating for fluctuations in demand by, for example, shifting workers into other departments (see also Sennett, 1998: 122–125). By fissuring work, lead companies have shed this insurance function; the (typically smaller) companies that now employ many workers are much more prone to dismiss them (or go bankrupt). The irony is that the very principle of incorporation—allowing companies to be legal persons of their own, separated from the private fortunes of their owners—is that they can carry greater risks (e.g. Ciepley, 2013). But corporations are, at the same time, “externalization machines” (Bakan, 2004) that do whatever they can to externalize costs and risks—including those that come with permanent, stable employment

In addition to the formal side of wage structures and the distribution of risks, social norms and fairness perceptions might also play a role on a more informal level, between individuals who see each other as colleagues. For example, many individuals probably subscribe to the principle that if one can help a person close to oneself without great costs, one should do so. There are many forms of spontaneous, low-cost support that colleagues offer each other, both on the job (such as help with specific tasks, knowledge sharing, etc.) and off the job (such as helping out in family emergencies, passing on information about an apartment available for rent, etc.). If workplaces integrate individuals from different backgrounds, more resourceful individuals can more easily help those in greater need than if work organizations are stratified by class, because of fissuring. An important form of such mutual support are those that can increase the social mobility of disadvantaged groups, for example, by helping out with tips for how to present one's CV, providing mentoring, or lending “decent” clothes for an interview. Such informal forms of support, in addition to the epistemic effects discussed above, can be especially valuable for the children of disadvantaged workers who want to enter higher education or better jobs.

Again, a critic might say that it is not the role of workplaces to offer such opportunities. Rather, this should be ensured by the educational system. This argument makes sense at the level of ideal theory: if one were to design an ideal society, one would want it to have an educational system that offers all children true equality of opportunity, independent of background. Insofar as “soft” factors such as habitus, social connections, and so on play a role for access to opportunities, the educational system should provide mentoring or other ways of supporting children from disadvantaged backgrounds. This is true, and improvements of the educational system in this direction are obviously to be welcomed. But in many countries, this system is itself in bad shape, and additional mechanisms are needed. Insofar as workplaces—the place where grown-up individuals spent most of their waking hours, outside of their home—can play a role in this respect as well, this should be embraced, at least on the level of non-ideal theorizing.

Potentials for democratic experiences

What a democratic society would ultimately want is for workplaces to be democratic: to allow citizens to gain democratic experiences at work, which can “spill over” into the political realm (e.g. Pateman, 1970: chap. 3; Schlachter and Arsaelsson, 2024). This would directly strengthen a democratic form of social cohesion, shaped by democratic habits and skills. Arguments for workplace democracy can be made from a number of perspectives, for example the need to control economic power or the greater likelihood of meaningful work.

I will not try to summarize the debate about workplace democracy here (see e.g. Anderson, 2017; Ferreras, 2017; Frega et al., 2019), nor is my claim that there is a logically necessary connection between democratization and fissuring. In principle, non-fissured companies can be run non-democratically, while fissured workplaces can also be run democratically, especially under the idealized scenario I am assuming, which excludes the worst cases of exploitation or risk imposition. One could, for example, imagine a fissured economy consisting entirely of small worker cooperatives.

There are, nonetheless, some considerations that suggest that the two questions are not completely unrelated. For thinking about them, it is helpful to take up a point made by Frega (2020), who distinguishes between formal democratic structures, and “democratic patterns of social interaction.” While the likelihood of realizing the latter is greatest if there are also democratic formal structures (e.g. worker representation on boards, etc.), it is worth considering the possibility that a democratic ethos can also exist within formally non-democratic workplaces. Especially among colleagues on the same level of hierarchy, it seems possible that the “constant, everyday, experience of working together, of being actively involved in the same social tasks, striving for the same goals, cooperating together to form and achieve shared goals” (Frega, 2020: 46) helps develop a democratic ethos even if formal structures are not democratic.

With this distinction in mind, what can be said about the likely effects of fissuring? Let me turn to formal democratic structures first. One obvious point is that insofar as fissuring prevents unionization (cf. Weil, 2014: 77), non-fissured work, by definition, would be more likely to included democratic elements. In countries in which unionization needs to happen at the level of companies, organizing workers in one large company is easier than organizing them if they are splintered into numerous subcontracting firms or franchises. ¹⁷ Moreover, if fissured labor markets lead to companies with more precarious, quickly changing working conditions, it again becomes more difficult to organize workers (see also Herranz, 2024: 13).

In a country like Germany, where a relatively high degree of formally democratic work governance exists, fissuring might be a contributing factor to the erosion of these structures (imperfect as they may be considered, from the perspective of an ideal of workplace democracy). Democratic structures such as co-determination and works councils are typically well established in large companies, but there is increasing resistance to it from the side of employees—and particularly so in medium-size companies with 51–200 employees (Hertwig and Thünken, 2024: 3, quoting evidence from Behrens and Dribbusch, 2018). ¹⁸ While not all medium-size companies are the result of fissuring, this points to a possible causal pathway in the weakening of democratic structures: if large companies typically have such structures, but smaller companies are less likely to have them, then fissuring might contribute to the erosion of these structures. More research would be needed to see whether this is indeed the case.

What about the informal side of work, and the chances of experiencing a democratic ethos on that level? Again, this is an empirical question. A defender of fissuring might hold that it is precisely because it creates companies with more similar occupational roles (e.g. cleaners only) that the likelihood of being seen as an equal by one's colleague might be greater in fissured than in integrated companies. The shared experience of disadvantages (though, by assumption in my idealized scenario, not of the worst kind) might bring together groups of people who have shared interests that it is worth fighting for, motivating them to organize and engage politically.

On the other hand, the logic of fissuring means that the resulting companies typically stand under great cost pressure because there is no longer the “insurance function” of larger integrated companies, as described above. The bosses of fissured workers (the owners of franchises or subcontracting firms) might be correct if they tell workers that they can do little to change working conditions, because they are in turn under pressure from the lead company. This might make the very process of organizing a non-starter, as employees realize that there is little to gain.

Moreover, in fissured work relations, the working conditions are dictated by the contract with the lead company, often down to minute details. This means that there is very little space for a positive experience of working together toward a joint goal, as described by Frega (2020). But such experiences matter for the “democratic spillover.” Another study from Germany ask whether employees, at their workplace, feel overlooked (negatively coded), are able to speak freely about unions, can solve problems together with their colleagues, and feel that they can actively improve things (positively coded). These survey items were summarized in the construct “industrial citizenship”, which was positively correlated with more positive attitudes towards democracy (Decker and Brähler, 2020: 119–147). Beyond the formally democratic governance of companies, what also matters for democratic spillovers are thus opportunities for agency and an egalitarian and collaborative culture among colleagues. While it is not a logical necessity, one must assume that fissured work, even under idealized conditions, often does not provide workers with such opportunities. More research into the informal side of fissured work could uncover whether this hypothesis holds true.

In summary, effects that concern the opportunities for members of society to get to know each other, inequality effects, and—somewhat more tentatively—effects concerning the likelihood of making democratic experiences at work all point to a negative effect of fissured work on social cohesion. Lack of social cohesion, in turn, is likely to weaken democratic life because citizens cannot develop ties of trust with each other, which are needed to engage in democratic politics from below and to hold the powerful to account (e.g. Herzog, 2023: chap. 10–11).

Let me reiterate that these three lines of arguments about social cohesion require seeing work as something more than an input factor into an economic process, governed by a logic of efficiency. It takes seriously the fact that work is a major part of people's life and shapes their experience of society to a great extent (maybe in an ideally just and democratic society, this would be different, but this is another question). This is why counterarguments that want to shift the mechanisms of social cohesion I have described to other spheres of life, for example, civil society or religious associations, are not convincing: if work segregates people 5 days per week (or more!), then the creation of sufficient social cohesion, across these dividing lines, on the remaining days or during holidays is unlikely (at least, that is, if work times is not considerably reduced, which does not seem to be on the horizon). Rather, ceteris paribus, more integration should happen at work, and policies against fissuring could contribute to this.

The most central counterargument one here encounters is that doing so would bring massive welfare losses—maybe so massive that this could, in turn, become a threat to social cohesion and/or democracy. If all the “efficiency gains” through fissuring had to be given up, where would this leave the economic system? But this argument plays on an equivocation of the notion of “efficiency.” From the perspective of a company, efficiency can be increased by organizing processes in better ways, or by cutting costs in ways that are not an overall gain to the economy, but rather a mere shifting of rents from one group to another. Arguably, the “efficiency gains” from fissuring have had a lot to do with gaining at the cost of others (first and foremost, workers), rather than with creating genuine win-win solutions. An economic system that fights back against fissuring would certainly move to a new equilibrium—but whether that equilibrium would be less efficient, overall, would remain to be seen. For example, higher incomes on the part of low-skilled workers might increase demand for many goods and services and could thus in turn support other businesses. If one assumes, in line with welfare economics, that more purchasing power for poorer strata of society is a good thing, then anti-fissuring strategies need not raise fears about welfare—although they obviously cannot and should not be understood as the only steps that are needed to fight poverty and inequality in the current economic systems of many countries.

Conclusion

In this article, I have discussed the phenomenon of fissured workplaces from a democratic perspective. While many of the real-life harms of fissuring, for example, health and safety risk and exploitatively low wages, are normatively overdetermined, I have focused on two more specific arguments: arguments from good work, and arguments from social cohesion. The former provide general arguments against bad working conditions, but not necessarily specific arguments against fissuring. The second, by contrast, raise questions about the role that workplaces can fulfil for citizens to get to know each other across class and other divides, about fairness norms, and about the possibility of democratic experiences at work.

My goal has been mostly diagnostic: I have tried to analyze the reasons behind the intuition that something is wrong with fissuring. What is at stake are not only low wages and bad working conditions, but also the role that work can play for the social cohesion of democratic societies—at least, it has, to a certain extent, played this role in the past. If it no longer does so, but one still thinks that social cohesion is important, one can react in two ways (which can also be combined): one might look for other social spaces in which social cohesion could be strengthened, or one might ask in which ways the fissuring of work could be counteracted. At the very least, it might be said, fissuring should not be encouraged, for example, through a lack of enforcement of labor law that lets many illegal practices happen, often at the cost of the most vulnerable workers.

Weil (2014, 2019) suggests a number of steps for policy implementation, to “stop egregious forms of fissuring that are solely designed to end-run basic workplace obligations established through existing public policies” (2014: 204). He argues for ascribing greater responsibility for the enforcement of existing rules to lead companies: just as they insist on quality standards that they define and monitor, they should also insist on work standards (and be forced to do so through changed liability structures) (2014: chap. 9, esp. 205). Other reform proposals have to do with the definition of employment (including joint employment) (see also Goldman and Weil, 2020). Another option is to also draw on antitrust law, as Callaci and Vaheesan (2022: 50–54) suggest, especially for reducing the power of lead companies over subcontractors or franchisees. Bieber and Moggia (2021), finally, suggest a tax solution, the “Principle of Inverse Coverage,” which makes employers’ contributions to social security systems inversely proportional to the precarity of work, to create incentives against the social harms caused by precarization.

While I applaud these proposals, the question is how far they would go in reintegrating workplaces—and the broader question is whether such a reintegration could and should be pushed more strongly by policy measures. As mentioned above, social cohesion could maybe also be strengthened in other ways, so much will here depend on concrete policy circumstances. But work remains too important a social realm, at least in the near future, to completely neglect it. A more radical step could be to separate social security contributions from employment, and instead to tax companies for their profits to pay for social security no matter whether they employ people directly or not. This would reduce the pressure on labor costs and would be particularly relevant for companies that earn profits through the work of algorithms and robots (made possible by the past knowledge and skills of workers). In addition, by making collective bargaining at the industry (not company) level mandatory, governments could give workers more power vis-à-vis companies. At the same time, mandatory unionization (Reiff, 2020) could prevent freeriding on union benefits by non-unionized workers and level the playing field between unionized and non-unionized companies.

My central point, thus, is this: democratic societies cannot remain indifferent if a core social sphere, which had in the past contributed considerably to social cohesion, is restructured, for the sole purpose of corporate profits. Technological change has many positive potentials—but is all too often used in ways that benefit only companies, while workers are treated as factors of production. But they are also citizens, and as such, they can and should set the rules of the economic game differently, so that work is an experience that contributes to, rather than undermines, democracy.