1 Introduction

Liberalism has shaped the world as a political philosophy that guided countries in designing their governments. Even beyond liberal countries, the market is appreciated for its efficiency in coordinating society and allocating scarce resources. The moral philosopher Adam Smith coined the famous metaphor of the invisible hand that leads people as economic agents to do what is good for society by doing what is good for themselves, which means for firms to maximize profits (Friedman 1970). After years of retreat, liberalism experienced a remarkable rebound after the implosion of the Eastern bloc, which is widely taken as evidence for its superiority over the communist ideology. Margaret Thatcher went so far as to claim that “there is no alternative” to liberalism Helleiner (2003) and Fukuyama (1992) proposed that it is “the end of history.”

Liberalism has never been that popular among business ethicists. The corporate scandals in the early 2000s and the 2008 financial crisis reassured its critics in claiming that profit maximization is exactly what plunges us into crises. The liberal paradigm also hinges on national governments to support the invisible hand with laws in ensuring that individual profit seeking is for the good of society. Globalization has reduced the regulatory power of governments. This led critics to denounce a “normative deficit” (Margolis and Walsh 2003; Scherer and Palazzo 2011), which is exemplified by social or environmental dumping (Hay et al. 2005). Globalization thus exposes the weaknesses of liberalism, and some argue that liberalism therefore can no longer afford ethical guidance to companies (Hanlon 2008; Gond et al. 2009).

This article sets out to show that liberalism remains a valuable ethical resource. Trivially, liberalism is primarily about liberty, not profits. It advocates the market because the market allows free contracting, and under sufficiently favorable conditions, it is indeed enough for companies to maximize profits and abide by the rules. However, liberalism does not warrant, let alone advocate unrestricted profit maximization; instead, the concept of freedom both allows and confines profit seeking. We suggest a broad reading of freedom, which encompasses both negative and positive freedom, and derive criteria to establish that contracts are based on voluntary consent. From the viewpoint of liberalism, corporate action is ethically legitimate if it builds on voluntary consent, which should guide contracting. We discuss the example of sweatshops for illustration.

The main contribution to the literature is the conceptual clarification of liberalism, along with the implications that we derive from it. Our argument reveals the potential contribution of liberalism to the development of a theory of CSR, which, so we believe, should not be pitched against liberalism. Liberalism is often reduced to and condemned as a profit-seeking imperative. The benefits of the market taken aside, this line of reasoning fails to appreciate the concept of liberalism and to see that, properly understood, liberalism infers sensible and non-trivial restrictions on contracting. Regardless of the legal environment, contractors should ensure that the other party’s consent is voluntary. The fact that the other party consents does not mean that consent is voluntary. This is particularly true if one contractor is more powerful than the other.

We proceed in six steps. We first summarize the rather critical view of the business ethics literature on liberalism (Sect. 2). Next, we discuss the role of freedom in liberalism, which liberalism prioritizes over profit (Sect. 3). We then infer that contracts must be voluntary to be legitimate and, inspired by the notions of negative and positive freedom, we develop criteria to test for whether they are (Sect. 4). We go on to explore the potentially controversial minimum resources to guarantee positive freedom (Sect. 5). Sweatshosp labor serves us as an example to illustrate the implications of our criteria for contracting (Sect. 6). Finally, we conclude with a summary and outlook (Sect. 7).

2 Liberalism and business ethics

While most scholars of business ethics disagree with Milton Friedman’s (1970) claim that “the social responsibility of business is to increase its profits,” his seminal essay in the “New York Times Magazine” remains the reference point for much of the literature (Jahn and Brühl 2018; for examples, see Scherer and Palazzo 2007, p. 1106; Freeman and Phillips 2002, pp. 331–332). In a nutshell, Friedman claims that business firms should maximize profit while abiding by the rules. His argument relies on the assumption of a clear separation between the public and private domain, where the rules belong to the public and business to the private domain, and the rules can be relied on to enforce legal or ethical behavior (Djelic and Etchanchu 2015). The business ethics literature challenges this argument mainly on two grounds.

Proponents of political CSR and similar concepts (Matten annd Crane 2005; Scherer and Palazzo 2007) take issue with the assumption that there are rules to ensure the ethical conduct of business. In the absence of a strong political environment, they argue, companies must step in and accept political responsibility. Modern western nation-states used to offer an environment which largely fulfilled Friedman’s assumption; however, as the authors point out, the regulatory power of states has eroded as a result of the globalization. Economic transactions expand beyond national jurisdiction and enforcement to oppressive or failed states. They mention global warming, deforestation, the regulation of capital markets, and sweatshops are examples of challenges that cannot be addressed by states individually. Moreover, migration and the disintegration of societies undermine the political authority of national governments (Scherer and Palazzo 2011).

As an explicit challenge to liberalism, Scherer and Palazzo (2007) draw on Habermas’s theory of deliberative democracy to develop their concept of political CSR. Under this concept, firms engage in a discourse with civil society, which is typically represented by NGOs to solve ethical issues. An example is the Forest Stewardship Council, which brings together wood-working corporations (e.g., IKEA, Home Depot, and OBI) with activists and NGOs to develop principles of sustainable global forest management. The council was founded at the beginning of the 1990s, after governments failed to develop shared standards in the United Nations’ conference on Environment and Development and thus illustrates how firms step in and take political responsibility where governmental regulations fail.

When firms turn into political actors, the question of democratic legitimization naturally arises, which Scherer and Palazzo (2007) address by founding their concept on Habermas’s deliberative theory. However, other questions remain controversial, such as why firms should step in or how much political responsibility they should take. Specifically, there is a risk of overwhelming firms morally as the responsibility to address market failures is easily shifted to them. Incidentally, Smith (2019, p. 133) notes that Habermas himself argues in this sense. Rather than turning firms into political actors, one might seek to develop international institutions to reestablish the conditions for the liberal division of labor between government and business on the international level and avoid the pitfalls of leaving it to firms (Mäkinen and Kasanen 2016).

The advocates of political CSR do not oppose markets as a mechanism to allocate scarce resources. Instead, their concern is that a key assumption of the liberal argument for markets is not fulfilled and therefore call for adjustments. However, there is also a more fundamental debate about the moral limits of markets, led by a heterogenous group of political philosophers, legal scholars, economists, and business ethicists (Wempe and Frooman 2018). The most prominent “market moralists” are Michael Sandel and Michael Walzer, who started the debate in the 1980s as a critique of John Rawls’ (1971/1999) “Theory of Justice” (Sandel 1982; Walzer 1983). The “market moralists” claim that the market logic has expanded into spheres of life where it does not belong.

Sandel (2012, 2013) puts forward two main objections to markets. First, we might see parties seemingly agree on a deal, but consent might not be voluntary. The standard example is the desperately poor peasant who “agrees” to sell his kidney because he does not really have a choice. Second, market practices can “corrupt” the good or value traded. Some of the examples that Sandel discusses are paid blood donations, which produce results that are empirically worse than those of unpaid donations; residential approval for nuclear waste sites, which decreases when residents are compensated; late pick-up of children from daycare, which increases when parents are fined. In these and other cases, Sandel argues, market mechanisms are applied to situations where they “do not belong,” sometimes resulting in unintended and surprising effects.

Sandel claims that the expansion of the market mechanism has been gradual and unnoticed. He seeks therefore to direct people’s attention to the problem and spark a public debate, but he does not suggest a readymade solution. Walzer (1983), in turn, distinguished different domains, such as business, politics, education, and the like, which are governed by different values and use different allocation mechanisms. While these domains help organize the debate, they remain naturally controversial. Wempe and Frooman (2018) revisit Walzer’s and similar works to derive their own domains and discuss allocation mechanisms, such as lotteries, queuing, rationing, or assessment based on merits, needs, or other attributes. Markets turn out an appropriate mechanism to support the value of efficiency, whereas a lottery is a mechanism to support equality.

Unlike liberalism, there is no single closed theory underlying the market moralists’ criticism or proposals. They are often labeled “communitarian” (Bell 2016) to emphasize that they give more weight to the community relative to the individual than liberals, but neither Sandel nor Walzer nor leading representatives of communitarianism have subscribed to this label. This does not mean, of course, that the criticism is wrong or unsound. Many of Sandel’s examples are supported by evidence (e.g., Frey and Oberholzer-Gee 1997; Gneezy and Rustichini 2000; Roth 2007). That said, there are also examples where liberal-minded readers, with all arguments on the table, might not follow Sandel’s moral intuition but still go with the arguments for the market. Hence, the judgment about the issues on hand is quite subjective in some cases (Küpper 2011; van Aaken et al. 2011).

The purpose of this article is not to rebut or downplay the concerns of the market moralists. Instead, we will show that liberalism, with a proper emphasis on individual freedom, imposes stronger ethical restrictions than some of its critics—and probably also many of its proponents—commonly think. Our argument thus relates more to Sandel’s first concern about coercion, as we explore when consent can be considered voluntary. Conversely, we do not seek to address his second concern about the application of markets to values that are thus corrupted. In terms of the debate about political CSR, our argument will highlight moral obligations and criteria that business firms can use to complement legal rules and substitute missing regulation.

3 Liberalism and freedom

“A liberal, I suppose one could say, is a person who believes in liberty, as a nudist is a person who believes in nudity” (Cranston 1967, p. 47). As liberalism is about liberty, it inherits also the ambiguity of this concept, to the point that it means opposite things to different people. For economists and philosophers like von Hayek and Nozick, liberalism is first and foremost about freedom from government intervention. Presuming an antagonism between individual freedom and organized social action, they conceive of government as the ultimate source of coercion. We refer to this view with Samuel Freeman (2001) as “libertarian” rather than “liberal.” Far from embracing the socialist idea of the welfare state, liberalism does not subscribe to the unfettered libertarian distrust against government. It recognizes that freedom requires certain goods, such as food, healthcare, or schooling, and that the government may help provide these (Rawls 2005; Sen 2011).

Liberal conceptions require, first, equal freedom for everyone and, second, that freedom has intrinsic rather than instrumental value. That is, freedom is an end in itself rather than a means to other ends (Peter 2004; Sugden 2010). These requirements are broad enough to comprise an otherwise heterogeneous field of philosophers like Immanuel Kant and John Rawls and economists like James Buchanan and Amartya Sen. Most approaches to business ethics are liberal in this sense. The theories of Donaldson and Dunfee (1999), Homann and Blome-Drees (1992), Küpper (2011), Pies (2011), or Scherer et al. (2014) differ in many ways, but they all consider the individual to be the ultimate source of value without categorically refuting government intervention. Freeman and Phillips (2002) argue thus that even stakeholder theory, which is often opposed to liberalism, is liberal indeed.

The requirement that freedom has intrinsic rather than instrumental value is crucial. It implies that freedom cannot be overruled by other values (van Aaken 2012). For example, a liberal society would not sacrifice freedom to wealth, and market economies are superior to planned economies not because they create more wealth, but because they allow more freedom. Of course, liberalism does not oblige individuals to maximize their freedom, but requires that they be free to decide on their own values (Buchanan 1986). Whether someone wishes to take an important position in business or polity, amass wealth, or fight poverty, freedom does not substitute any of these goals or the values underlying them; it is the sine qua non to pursue them. The more people’s discretion is confined by coercion, the less they can live up to their own values (Feinberg 1984).

Freedom has intrinsic but not absolute value, and there are two potentially acceptable reasons to restrict freedom. First, an individual’s freedom is necessarily limited by the principle of everyone else having the same freedom. Without this restriction, interference by others might prevent individual choices from being realized. This principle has a long tradition and was prominently stated by Kant, who argues that everyone can pursue his or her life goals in a way that he or she likes it to do as long as nobody infringes upon “the freedom of others to pursue a similar end which can be reconciled with the freedom of everyone else” (Kant 1970, p. 74). In other words, if everyone is supposed to be free to pursue their goals, his or her freedom must be compatible with others having the same freedom (Gewirth 1978).

Second, there is a liberal case for prohibiting people from dealing away too much of their freedom. This case is more subtle because freedom is both the sine qua non of trade and a tradable good (van Aaken et al. 2014). A work contract, for example, naturally restricts the contractors’ freedom (Ostermaier 2013). In particular, work contracts allow employers to determine some of their employees’ actions, who thus give away freedom in return for other goods that they value and that their wages can buy them. That said, liberalism does not allow contracts where employees sell themselves into slavery or serfdom. Put simply, the intrinsic value of freedom implies that no one is free not to be free or, as Mill (1859/2001, pp. 345–346) famously put it,

“When persons have bound themselves by a contract, not simply to do some one thing, but to continue doing something forever or for a prolonged period, without any power of revoking the engagement, the presumption which their perseverance in that course of conduct would otherwise raise in favor of its being advantageous to them, does not exist; and any such presumption which can be grounded on their having voluntary entered into the contract … is commonly next to null. The practical maxim of leaving contracts free, is not applicable without great limitations in the case of engagement in perpetuity.”

Practically speaking, restrictions on contracting can also be warranted when the basic requirements of voluntary and informed consent, which we will explore at more length in the next section, are not fulfilled. Of course, any such restriction should possibly remedy the obstacles to making a voluntary and informed choice. For example, the liberal case for compulsory schooling is not so much that education increases social welfare, but that (certain) students lack the ability or information to fully appreciate the benefits of schooling and that schooling enables students to make more voluntary and informed choices later in their lives (Oreopoulos and Salvanes 2011). For similar reasons, liberalism does not conflict with public health care and other public services, such as the advancement of science, art, and culture (Knight 1939).

Summing up, liberalism is characterized by assigning intrinsic value to individual freedom. It is then left to every individual to choose the values to pursue, as long as that individual does not violate other people’s freedom and does not give away his or her freedom for good. When people interact or, economically speaking, contract with each other, their contracts must therefore be based on voluntary and informed consent, no matter the implications that this may have for their wealth. Liberalism is primarily about liberty, which legitimizes markets to the extent that they ensure more liberty than other coordination and allocation mechanisms. Hence, liberalism cannot be reduced to an ideology of wealth maximization, although wealth maximization is, of course, a legitimate value to pursue for individuals.

Proponents of political CSR are concerned about weak regulatory environments that fail to ensure that factual consent is voluntary. However, if one accepts that contracts must be voluntary to be legitimate, this requirement is universal and applies regardless of the society where a company operates and that society’s norms. If a legal system nullifies involuntary contracts, contractors do not need to worry about this ethical requirement. In the absence of such a system, it remains their responsibility to make sure that contracts respect each other’s freedom; otherwise, the contract may be ethically illegitimate from a liberal standpoint. For contracts to be legitimate, liberalism may thus require more than complying with the law. This is particularly true for multinational companies that operate in poorly developed and non-democratic legal environments.

4 Voluntary consent

The requirement of voluntary consent seems uncontroversial at first glance. In fact, the liberal preference for the market as a social coordination mechanism derives from the argument that it gives people the freedom to contract as they please. Friedman (1962, p. 13) argues thus that a “household … need not enter into any exchange unless it benefits from it. Hence, no exchange will take place unless both parties benefit from it.” The example of the poor peasant who sells his kidney, however, raises the question of whether consent can always be taken to be voluntary. Another example is the so-called “Summers memo,” which proposes that toxic waste be exported to poor countries that accept it for disposal in exchange for money (Pellow 2007, pp. 9–10). What is considered voluntary depends on the underlying concept of freedom (Hoßfeld and Schmiel 2015).

An individual is free to the extent that he or she can put his or her intentions into action or, more technically, on the actions available in a given situation. It is helpful to think of these actions in terms of whether that individual is allowed and able to take them, or may and can take them. In his seminal essay, Isaiah Berlin (1958/1970) popularized the terms of negative and positive freedom to distinguish these two concepts of liberty, where freedom to do something is positive, whereas freedom from obstacles or coercion is negative. It is easy to think of actions that someone is allowed but unable to take in some situation, as well as actions that he or she is able but not allowed to take. Hence, while both negative and positive freedom are necessary for an action to be available, the concepts are independent and can be meaningfully distinguished.

Libertarians like von Hayek and Friedman tend to reduce freedom to negative freedom. Max Weber (1922), however, already pointed out that freedom from coercion is not valuable in itself, because an option is vain if there is no real chance to use it. Similarly, Berlin (1970, p. 48) asked: “[f]or what are rights without the power to implement them?” People need the resources to take options to be free. Negative and positive freedom are thus two sides of the same coin (Bavetta and Navarra 2012; Hoßfeld and Schmiel 2015). For example, wealth is useless if the good someone wants to buy is prohibited. In turn, if the purchase is allowed but the good is unaffordable, wealth does not increase that person’s freedom either. Hence, an individual is free to the extent that he or she is allowed and able to realize his or her life choices (MacCallum 1967).

These two concepts of liberty have implications for the legitimacy of business dealings. First, contracts are only voluntary and thus legitimate if neither contractor has been forced into contracting (van Aaken 2012). Even a contract that seems to benefit all parties is therefore not necessarily voluntary. For example, a robber who makes you choose between your money and your life offers you a deal that holds benefits for both of you: the robber gets your money, you keep your life. However, you would not even consider this deal if you were not forced into it. The robber confines your action space to two actions, both of which make you worse off, and thus restricts your negative freedom. Hence, consent to a deal that reduces ex ante the options available to any contractor cannot be assumed to be voluntary.

Second, a contract can even be illegitimate when it creates rather than removes options (Meyers 2004). For an example, imagine you go shipwrecked and are about to drown when another sailor happens to find you and offers to rescue you for money (Nozick 1969). Unlike the robber, the sailor does not force you to contract. Instead, he exploits your situation, which forces you to contract. The deal creates rather than removes options and thus does not decrease freedom: while you have no choice but drowning without the offer, you can now choose to live. However, the cost of declining the sailor’s offer (i.e., drowning) is prohibitive, which makes it a “freedom enhancing coercive” offer (Feinberg 1988). Summing up, consent to a contract that creates an option which cannot be refused for lack of alternatives cannot be assumed to be voluntary either.

Of course, we do not claim that a deal may only be offered when people have the resources to accept it. For example, there is nothing wrong with offering a fancy new product to consumers that many cannot afford. The point of the example is not that the shipwrecked person may be unable to raise the amount demanded by the sailor. The point is whether the cost of refusing it is prohibitive. Potential consumers do not need to purchase the new product; the shipwrecked person, in turn, must accept the sailor’s offer. If a potential contractor lacks the resources to accept an offer, it is sometimes possible to adapt it. If an offer does not pass the test for positive freedom, a hypothetical test can reveal how that offer would have to be adapted. This thought experiment will usually result in a more advantageous offer than what the other contractor would accept involuntarily.

While the requirements for consent to be voluntary are controversial, it clearly cannot be taken to be voluntary if it is based on incomplete or false information (Friedman 1962). Informed consent is a practical issue for multinational corporations, which contract across language or cultural borders, so that misunderstandings and misinterpretations of the terms of contracts are common. There are mainly two reasons why consent is potentially not informed: lack of information and contractors’ incapability of processing information (Feinberg 1986; Trebilock 1993). A situation where fully informed individuals interact is ideal; in real settings, the question is rather whether contractors are sufficiently informed. Several methods have been proposed to test for whether contractors have sufficient information (e.g., Epstein 2006).

Some groups of potential contractors, such as children or the mentally challenged, are widely considered incapable of processing information well enough to give informed consent, either to particularly complex or consequential deals or in general. The capacity of processing information can also be lost temporarily (e.g., when people are sick), and situations can be arranged to hinder people from processing information (e.g., by time pressure). Many jurisdictions codify rights that protect vulnerable groups of consumers and discourage the creation or exploitation of such situations. For example, the right to withdraw from door-to-door contracts, which gives consumers extra time to process information, is law in various jurisdictions. To legitimize their contracts, business firms should play by such rules even when they are not legally binding.

Intuitively, the requirement of informed consent precludes contractors from deceiving each other and withholding information requested by the other party. Legal requirements go sometimes further, though. For example, EU countries require website providers to obtain users’ explicit (“opt-in”) consent to their data policies. Large multinational corporations enter often into asymmetric contracts, where they are the stronger party: they have more information and the contract matters more to the other party than to themselves. This increases their responsibility to obtain informed consent. Mediation or questionnaires to cross-check the terms of the contract are means of reducing the asymmetry in power, information, and importance of the subject-matter, which pose a threat to the legitimacy of the contract (Feinberg 1988).

An important device to ensure informed consent are accounting and reporting standards. These are enacted and enforced by legislators and regulators to give (potential) stakeholders information and enable them to make voluntary and informed decisions about whether to provide funding to a corporation, buy its products and services, work for it, or interact with it in any other way. This holds both for the IFRS and national GAAPs, although these vary in whom they address in the first place and which decisions they are supposed to facilitate (Böcking 2007; Küpper et al. 2013). In the last decades, standards, frameworks, and guidelines to cover areas other than the financial position and performance have multiplied, such as the triple bottom line framework or guidelines of the Global Reporting Initiative (Kannenberg and Schreck 2019; Küpper 2011).

Summing up, liberalism favors markets to coordinate people because it gives them more freedom than other social mechanisms. However, the fact that market participants agree on a deal does not imply per se that this deal reflects the contractors’ freedom. A deal that confines the contractors’ choices ex ante or that cannot be refused for lack of alternatives should never be taken to be voluntary. It may be possible, though, to improve deals that do not pass the second test. Other than that, legitimate contracts require that contractors have all relevant information and are capable of processing it. Ideal regulatory environments ensure that any contract that complies with the rules is legitimate. In less-than-ideal environments, in turn, it is up to the business firm to ensure that these requirements are fulfilled to legitimize its operations from a liberal perspective.

4.1 Standards for positive freedom

Positive freedom is the most controversial requirement for voluntary consent. Imagine a multinational corporation that wants to hire workers for offshore production who have no choice but to work for that corporation to feed their families. Is the lowest wage that these workers are willing to accept “fair”? Clearly, this is not necessarily the case. The corporation might therefore want to conduct a hypothetical test to determine a “fair” benchmark. Such a test requires an assumption about the resources that a worker should have to be free to accept or reject the corporation’s offer and is, in a sense, independent of the market. This is another version of the question whether and which resources a society should grant its members. Social philosophers have made various attempts to derive criteria (e.g., Rawls 1971; Nussbaum 2004; Sen 2011).

Rawls (1971) conducts a famous thought experiment to answer this question. Which principles would rational and mutually disinterested (i.e., egoistic) individuals agree on to allocate wealth in a hypothetical state of nature, where no one “knows his place in society, his class position or social status, nor does anyone know his fortune in the distribution of natural assets and abilities, is intelligence, strength, and the like” (p. 12)? Behind this veil of ignorance, Rawls argues, the members of the society would agree on two principles: first, primary social goods are equally distributed; second, any inequality must advantage the least advantaged in society most. By contrast, they would not agree on the principle of welfare maximization, which readily sacrifices someone’s individual welfare to increase social welfare.

Primary social goods are resources that everyone wants to have, regardless of his or her personal preferences. Of course, these preferences are hidden behind the veil of ignorance: the members of the society do not know their own preferences when deciding about the principles of justice (Rawls 1971, p. 92). As the available quantity of any good is, again, behind the veil of ignorance, primary goods are such that everyone will want more rather than less of them (Buchanan 1975, pp. 398–405). Specifically, primary goods include individual rights and liberties as well as income and wealth. Financial resources figure prominently among these goods. This is unsurprising because money is fungible and it thus enables everyone to pursue their (yet unknown) preferences, no matter which these turn out to be (Sugden 2010).

An intuitive choice for the allocation of primary social goods would be some minimum living income. This minimum income varies, of course, with the cost of living. Returning to the example from the beginning of the section, the multinational corporation might take the local minimum living income as a benchmark to determine a fair wage level rather than go for the lowest wage workers are willing to accept. The OECD, the International Labour Organization (ILO), and other organizations propose guidelines to determine this minimum income. For example, the ILO Minimum Wage Fixing Convention (No. 131) accounts for the needs of workers and their families, relative to the general level of wages in the respective country, the cost of living, social security benefits, and the relative living standards of other social groups.

Other philosophical approaches come to similar conclusions. Zwolinski (2015) develops a libertarian argument for a basic income, which relies on the Lockean proviso. Locke (1689/1952, sec. 27) claims that an item of common stock may be turned into private property if the new proprietor leaves “enough, and as good, in common for others.” The common reading of the proviso is that everyone must be compensated individually (not just on the aggregate level) for the loss in common stock. Zwolinski concludes that a basic income is the least oppressive and probably only feasible way to satisfy the proviso. As opposed to the common libertarian argument for basic income as an efficient solution (e.g., Zwolinski 2014; Friedman 1962), this argument derives from ethical theory. However, it should be noted that it remains controversial (Rallo 2019).

Likewise, Nussbaum (2007, p. 58) points out that the outcome of Rawls’s (1971) thought experiment resembles the Kantian position that each person possesses an inviolability that is founded on justice and that even society as a whole cannot override. Taken together, it is important to note how liberalism differs from utilitarianism, which is blind to individual freedom. From a utilitarian viewpoint, there is no reason to question the legitimacy of any contract. In one way or the other, it must make the contractors better off, or at least they expected so, because otherwise it would not have been concluded. Hence, any such contract increases social welfare by definition. Liberalism, in turn, prioritizes individual liberty—both positive and negative—over everything else, which implies harder requirements for deals to be legitimate.

5 The example of sweatshop labor

Work contracts offer real-life examples of challenges similar to those of the sailor, albeit less colorful. In particular, sweatshops have received much attention in the business ethics literature (e.g., Kates 2019; Powell and Zwolinski 2012; Sollars and Englander 2018). Sweatshops are exploitative environments, where people work under poor conditions, such as extremely low wages, excessively long work hours, or health and safety hazards (Radin and Calkins 2006). Moral intuition tells us to condemn sweatshops as unethical, and this intuition is also supported by theoretical arguments. From a Kantian perspective, for example, sweatshops can be argued to treat workers as means only rather than ends (Arnold and Bowie 2003). However, the ethical debate is more controversial and has raised interesting arguments on both sides.

Maitland’s (1997) defense of sweatshop labor is the common point of reference for the sweatshop debate. Maitland refers to a New York Times article by Larry Rother (1996), who claims that people work voluntarily in sweatshops as “anxious onlookers are always waiting, hoping for a chance at least to fill out a job application.” Maitland takes this and other empirical evidence to argue that sweatshop labor complies with what he calls the “classical liberal standard”: it is ethically acceptable because it is voluntarily chosen by informed workers. From an ethical viewpoint, managers thus need not pay higher wages or provide better working conditions. Quite the contrary, Maitland argues that “it may be ethically unacceptable for [a company] to pay wages that exceed market levels [which] raise labor costs to the point of discouraging foreign investment” (p. 607).

Similarly, Sollars and Englander (2007) argue against legal regulation of wages, working hours, and safety requirements because these disrespect workers’ choice. They endorse Maitland’s conclusion that managers should not pay a wage above market price as higher wages contribute to unemployment and harm the least-advantaged workers. Zwolinski (2007) argues that it is wrong to interfere with the choice of people to work in sweatshops even if it is made from among a severely constrained set of options. Sweatshop labor must be regarded as those workers’ most-preferred option, and any interference will harm them by further confining their options (similarly, Powell and Zwolinski 2012). Flanigan (2018) points out benefits for the workers, who “take pride in their ability to provide educational and economic opportunities to their families by working” (p. 81).

While this line of reasoning has its merits, it is important to note that it reduces freedom to negative freedom. The test for positive freedom may lead to a different conclusion. Whiles sweatshops add an option for workers, the cost of rejecting it may be prohibitive. For example, if the alternative to working in the sweatshop consists in starving to death, the situation is very similar to that of the shipwrecked sailor. That is, sweatshops also create a freedom enhancing but coercive offer (Zimmerman 1981; Feinberg 1988), which exploits the asymmetry of power among the contractors (Carson 2013; Kates 2015). The conclusion that people work voluntarily in sweatshops just because they have accepted to work there is therefore fallacious and does not ethically legitimize sweatshops, at least not from a liberal point of view.

Conversely, there are good reasons for skepticism about whether the interference of third parties makes the potential contractors or society better off. In addition to the ethical argument in favor of freedom, interference has economic effects, which should be considered even in an ethical debate (Schreck et al. 2013; Flanigan 2018). Economists warn that any regulation has unintended results that can make the situation even worse (e.g., Bhagwati 2007). Numerous empirical studies document the negative impact of minimum wages and other regulations (Sollars and Englander 2018). While the liberal argument in favor of the market as a social mechanism derives primarily from its respect for people’s freedom, the efficiency of markets in making people better off is thus another reason why they are so widely accepted, even beyond liberal countries.

The purpose of this article, however, is not to argue either for or against legal regulation, but to develop liberal ethical criteria, and sweatshops create a freedom enhancing coercive offer. People who work there are likely to lack the positive freedom to decline this offer and thus their consent is not voluntary. Sweatshops labor is then not ethically legitimate from a liberal viewpoint, which requires consent to be voluntary both in the negative and positive sense of freedom. Regardless of legal regulation, corporations need to consider whether the deals that they offer comply with ethical standards, and liberalism turns out to be a resource to derive such standards. If a contract fails the test of hypothetical voluntary consent, we propose that decision-makers consider revising it such that it passes this liberal standard.

6 Conclusion

Much of the literature in business ethics replies to Milton Friedman’s (1970) narrow definition of the social responsibility of business firms. In the light of corporate scandals and economic crises, critics argue, for one reason or the other, that liberalism does not offer firms sufficient ethical guidance and that it is not enough for them to care about profits. The common identification of liberalism with markets, however, fails to capture that liberalism is, first and foremost, about freedom. While markets are the preferred mechanism to coordinate society because they support freedom, liberalism does not exempt market participants from respecting each other’s freedom in contracting. Properly understood, it turns out to be a powerful source of ethical guidance, which imposes serious but not excessive restrictions on firms.

Most criticism of liberalism derives from two arguments: first, contracts are not based on voluntary consent (i.e., freedom), but consent is forced; second, goods are corrupted when put up for sale. We explore the notion of freedom and follow Berlin (1970), who distinguishes negative and positive freedom, to respond to the first argument. We conclude that contracts are voluntary and thus legitimate if contractors do not restrict the other’s options ex ante; do not exploit the other’s lack of alternatives; provide the other with the relevant information and make sure he or she is able to process it. A contract exploits the other’s lack of alternatives if the cost of declining it (i.e., the opportunity cost) is prohibitive. In this case, the deal can possibly be adjusted to match a hypothetical situation where the other has the resources to decline it.

Clearly, it is often hard to tell what a “fair” deal looks like. For example, imagine a pharmaceutical company that develops a powerful drug that many people cannot afford. Dunfee (2006) argues that firms should devote substantial resources to “weaker” parties in such cases. Likewise, Hsieh (2004) infers from Rawls’s theory of justice that firms have a “duty of assistance toward those living in developing economies, even though the duty is normally understood to fall on the governments of developed economies.” That said, a drug, or any other good for that matter, is not just there to be distributed, but it needs to be provided. However, goods are not normally provided if the company cannot expect to recover the cost. While we cannot hope that our liberal criteria solve any ethical problem, we do believe that they offer guidance in most cases.

Democratic environments codify rights and liberties and provide citizens with resources. Ideally, it is enough for firms to play by the rules, which guarantee that consent to a contract may be considered voluntary. The argument for political CSR is that this assumption is not satisfied in a globalized economy, where multinational corporations can go offshore to dodge domestic rules. More generally, regulatory environments are rarely perfect. Under these circumstances the question might pop up why firms should voluntary forego profit for the sake of an ethical rationale. Without going in detail, there are economic, sociological, and institutionalist arguments for firms to respect norms that are not codified law (see e.g., Schreck 2009; van Aaken et al. 2013; Campbell 2006). Particularly, Dunfee (2006) showed that it does not matter why firms care about their ethical legitimacy. The key point is that they do take their responsibility seriously and walk the talk.

In conclusion, liberalism offers ethical guidance for companies, and the example of sweatshop labor shows that liberal requirements are not toothless. In turn, business firms should also be considered what they are, and they should not be subject to excessive burdens. From both viewpoints, liberalism is worth revisiting as a potential resource for business ethics, and we hope that our argument helps spark this debate. As an extension, we also noted that the results of Rawls’s thought experiment, surprisingly at first glance, match Kantian claims. Indeed, this is less surprising in the light of our argument for the ethical implications of liberalism. It would be interesting for future research to explore whether the requirements that we propose tally, in a similar fashion, with claims that are made toward firms from other approaches to business ethics.