Abstract
There is a sense of fairness that is distinctive of markets. This is fairness among economic competitors, competitive fairness. We regularly make judgments of competitive fairness about market participants, public policies and institutions. However, it is not clear to what these judgments refer, or what moral significance they have. This paper offers a rational reconstruction of competitive fairness in terms of non-domination. It first identifies competitive fairness as a distinctive claim, advanced within markets in turn characterized as antagonistic, instrumental and procedural. It distinguishes competitive fairness from a number of familiar ideals with which it might be confused: legitimate expectation, equality of opportunity, sporting fairness and economic efficiency. While many exponents likely assume competitive fairness can be explained in terms of one of these ideals, in each case there are significant objections to doing so. Instead, the paper argues that the most promising justification of competitive fairness is under the republican ideal of non-domination, which can reconstruct many of the intuitive judgments of competitive fairness that we make in particular cases. However, it concludes, this explanation makes it difficult for exponents to continue to emphasize competitive fairness, given diverse other risks of domination, and to other values, in markets.
Keywords
There is a sense of fairness that is closely associated with markets. This is fairness between competitors, competitive fairness. This is the sense of fairness that we invoke when we complain that a competitor has received an unfair advantage, or is using unfair means, to succeed in the market at our expense. Many of us regularly make judgments about competitive fairness, both in our own name and on behalf of others. However, it is not clear to what these judgments refer, or what if any moral significance they have. This is the ‘puzzle’ in the title.
Such judgments have wide relevance for economic law and institutions. They are particularly evident in anti-trust and international trade, but also arise in the international and domestic laws governing employment, taxation, investment protection and real property, and in competitive social institutions generally. My goal in this paper is twofold: first, to identify and problematize these judgments; and, second, to suggest a route to their rational reconstruction through the republican ideal of non-domination.
I proceed in three parts: the first diagnostic; the second critical; and the third reconstructive.
In the first part, I distinguish the principal characteristics of competitive fairness. Working from a set of examples that I hope will strike readers as familiar, I sketch the structure and content of competitive fairness claims, and the domain over which they are made. I do not seek, in this first section, to advance an argument for or against competitive fairness, but merely to identify it, with a view to its critical evaluation in subsequent parts.
The second part is critical: it establishes what competitive fairness is not. Competitive fairness bears a passing resemblance to various familiar moral principles, including sporting fairness, equality of opportunity, legitimate expectations and efficiency. However this resemblance is misleading, as its distinctive structure, content and domain distinguish it from each.
These first two parts, then, set up the puzzle. We recognize competitive fairness as familiar from our everyday legal and moral lives; but it does not seem to readily reduce to any other principle that is plausibly morally basic.
The third part seeks to reconstruct competitive fairness in terms of the republican ideal of non-domination. The archetypal examples of competitive unfairness, I will suggest, each involve the kind of arbitrary power that republicans identify with domination. An explanation in terms of non-domination avoids the problems identified with the principles considered in the second part, and offers potential insights into unclear or contested cases.
My method throughout is interpretive, in the sense advanced by Ronald Dworkin (1986: Ch. 2). We begin with a pre-theoretical account of a practice, identifying this inter alia through a number of paradigm instances. We next ask whether there is a principle or set of principles that could both explain and justify the practice, so conceived. To be successful, the required principles must both account for prominent features of the practice, and explain the value realized thereby. We can then return to our paradigm instances, and to any unclear or marginal cases, reassessing these based on our new, principled, understanding of the nature and function of the practice under examination.
Defining (and distinguishing) competitive fairness
The goal of this part is to identify and offer an initial specification of competitive fairness. I begin with a number of examples to motivate the claim that competitive fairness is a distinctive ideal to which we often appeal. Thereafter I identify some salient characteristics of markets, as the domain over which judgments of competitive fairness are made. Finally, I describe the structure that complaints of competitive fairness take.
Some examples of competitive fairness
This paper seeks to make sense of judgments and intuitions that I believe many of us share. This section discusses a number of examples, in the hope of evoking such judgments on the part of the reader, as a basis for subsequent discussion. My goal is not to establish a set of uncontroversial fairness judgments that readers will necessarily share. Rather, I imagine many readers will disagree about the extent to which each of the examples below is or is not unfair. My hope is only to show that we are inclined to make and contest judgments about the fairness or unfairness of these cases. Having established that these cases raise questions of fairness, we can move to the subsequent question of the criteria by which these should be settled, and the conclusions we should reach about each specific case.
Consider the following examples:
Subsidy: The United States subsidizes cotton farmers. As a result, these farmers can sell their products cheaply on world markets, reducing prices and in turn revenues and market shares of unsubsidized producers in other countries, including developing countries. (Arsenault, 2014) Regulation: The European Union's Emissions Trading Scheme imposes additional costs on certain enterprises, who must either reduce carbon emissions or buy permits to cover these. Enterprises in other countries without such schemes do not bear these costs, allowing them to offer their products at lower prices, or earn greater profits, than would otherwise be the case. (Nastu, 2008) Dumping: Korean manufacturers allegedly sell washing machines in the United States at costs below their cost of production, or below the costs at which they sell similar models in Korea. This leads to lower prices for washing machines in the United States, making it harder for domestic producers to maintain market share and profits. (WTO, 2013) Exclusivity: In the United Kingdom, nearly half of all bars are ‘tied houses’, subject to exclusive supply arrangements with particular breweries. (Heys et al., 2013) This makes it difficult for newer, smaller or independent breweries to enter the market, as they cannot sell their products through these bars. Scale: The Coca-Cola Company spends over USD$4billion per year on advertising and marketing globally, while its nearest rival, PepsiCo, spends around USD$1.5billion. Independent manufacturers cannot match this marketing. This makes it harder for new entrants to establish themselves in the soft-drink market. Labour: The minimum wage for garment workers in Bangladesh is USD$95 per month (Paul, 2013). In the United Kingdom, the equivalent figure is USD$1741.
1
As garment manufacturing is labour intensive, it is almost impossible for UK-based garment manufacturers to compete.
Without prejudging whether there is anything objectionable in any or all of these examples, readers will hopefully recognize them as cases where complaints of unfairness would not seem out of place, even if we might disagree about whether they were sound. We would not be surprised to hear complaints about unfair competition coming from the unsubsidized cotton manufacturer, the regulation-burdened European business, the troubled US washing-machine manufacturer, the struggling independent brewer or soft-drink manufacturer, or the failing British garment maker (or any of their employees). In some cases, the law (whether domestic or international) will specifically address the relevant situation or behaviour. In others, the law is silent, but we still feel there is something to be said, one way or the other, about the fairness or unfairness of the example. Even those who reject fairness claims in particular instances will usually explain that rejection in terms of competitive fairness – by arguing that there is nothing unfair about the particular situation or behaviour complained of. The disagreement is about what competitive fairness demands, rather than about its existence or relevance. It assumes that competitive fairness is itself a relevant and coherent standard, which can be applied to distinguish genuine from bogus complaints.
We might however wonder whether this is simply a mistake. Perhaps complaints about competitive fairness are simply “sour grapes”, complaints open to – and made by – almost anyone who loses out from economic competition? Why take seriously the special pleading of profit-motivated agents with substantial vested interests, whose object in making these complaints is plainly to advance those interests, regardless of the rights and wrongs involved? While defending the coherence of competitive fairness is the task of Part III below, let me foreshadow it now through offering one further example:
Quality: In the early 1970s British Leyland Motor Corporation constituted the bulk of the British motor industry. Its products had a reputation for poor quality and reliability. As a result, more reliable imported cars took a significant share of the British market, leading to British Leyland's decline.
British Leyland, its shareholders and employees were presumably unhappy about this result. However, assuming it was simply the result of consumers choosing the superior imported products, none could complain that their troubles were the result of competitive unfairness. Ceteris paribus, there is nothing unfair about consumers choosing someone else's product. This just isn’t a kind of claim that makes sense in these circumstances. (Other fairness claims could of course be made, about the ways the car industry and its employees were sacrificed for the benefit of other sectors of the British economy, but these are complaints of a very different form to those above.)
The contrast between Quality and the other examples above highlights a first important point about competitive fairness claims. Not every market outcome is susceptible to challenge in terms of competitive fairness. More generally, the fact that market outcomes benefit some and not others is not in itself a cause for complaint. Complaints of competitive fairness are complaints about specific features of, or conduct in, markets. They are not complaints about markets as such. 2
This gives us a first route into making sense of competitive fairness. Competitive fairness assumes that there are at least some fair market outcomes. In consequence, whatever competitive fairness is about, it cannot be about necessary features of markets. An account of what competitive fairness is about therefore requires that we first offer an account of what markets are about. That is the task of the next section.
Markets as antagonistic, instrumental, and procedural
Markets can be variously defined. Herzog (2017) characterizes them as “institutions in which individuals or collective agents exchange goods and services. They usually use money as a medium of exchange, which leads to the formation of prices.” The market is distinguished from other institutions, on this view, by the role of prices, exchange and competition. Polanyi (2001: 59) similarly emphasizes exchange and prices, and the market as the context in which enough buyers and sellers come together for competition over the former to produce the latter.
Others place more focus on the voluntary character of market exchanges, as distinct from administered transactions. Thus Friedman (2002: 14–15) identifies markets as a domain of “voluntary cooperation of individuals … individuals are effectively free to enter or not to enter into any particular exchange, so that every transaction is strictly voluntary”. Miller (1989: 18) tells us that “[a]s participants in a market, people's relationships are those of voluntary exchange. Each obtains what he wants by offering some equivalent benefit in return.” Or as McGurn (2003: 66) puts it: “Markets mean many things, but primarily they mean the voluntary exchange of goods, services, labor, and capital”.
Something else that is implicit, if perhaps not necessarily so, in these descriptions is that markets are characterized by relatively anonymous relations, and an emphasis on direct as opposed to diffuse reciprocity. This distinguishes them from private, local or pre-modern systems of division of labour along kin, community or household lines. In the market, we pursue our own goals, and we give only so that we may receive. (See generally: Polanyi, 2001, Ch. 5: Miller, 1999: 108)
Building on these general characterizations, let me suggest three features of markets that seem particularly relevant for making sense of competitive fairness. Markets are antagonistic, instrumental, and procedural.
Markets as antagonistic
Markets are antagonistic. The interests of different market participants are – at least to some extent – opposed, and each seeks to advance their interests at the expense of others. 3
There is of course a significant cooperative aspect of market exchanges, but there is never perfect harmony of interests: buyer and seller have a shared interest in finding a mutually beneficial transaction, but conflicting interests in how the cooperative surplus is divided. 4 Both want to transact, but one wants to buy cheap, and other to sell dear. However, the relationship between buyer and seller is not the only, or even the distinctive, relationship in the market. Rather, markets are competitive. They involve multiple purchasers seeking to buy, and multiple vendors seeking to sell, the same – or somewhat interchangeable – products. 5 And this latter relationship – between different buyers and between different sellers – is wholly antagonistic. We may both seek to make a particular sale, but only one of us can do so: our interests (in making this sale) are therefore directly opposed. 6
We can thus understand markets as involving relations along two dimensions, both somewhat antagonistic: between buyers and sellers; and amongst different potential buyers, or different potential sellers. Let's call these respectively the vertical and horizontal dimensions. This distinction is important because we can, and frequently do, analyse the fairness of market interactions along both dimensions. The Marxist concern with exploitation, for example, focusing on the division of benefits between buyers and sellers of labour, addresses the vertical dimension. (See generally: Zwolinski and Wertheimer, 2017) Similarly, the FairTrade movement addresses the vertical division of benefits between buyers and sellers across different levels in multi-stage production chains. (World Fairtrade Organization et al., 2009) Competitive fairness, by contrast, addresses the horizontal dimension, examining relations between agents competing for the same opportunities, and aspiring to fulfil the same roles and offer the same benefits to the same counterparties. 7
Markets as instrumental
Second, markets are instrumental. They are institutions for the allocation of goods whose value is not itself intrinsically linked to the manner of their allocation. (Miller, 1999: 27)
This distinguishes markets from other competitive institutions that allocate valuable benefits. Romantic relationships are often allocated through social institutions (e.g. dating) involving competition between participants. Similarly, many societies place surprising importance on contests of beauty, skill or athletic prowess in which many participants compete for some prize or honour. However, both are distinguished from markets by the close connection between the value of the goods distributed and the mechanism through which they are distributed. This is clearest in the latter case. The social recognition of beauty or skill can only be valuable – indeed can only be what it purports to be – if it is allocated through a mechanism that tracks, with some degree of accuracy, the quality that it claims to recognize. (Sandel, 2013: 94) An Olympic medal is only of value as such if it is won in bona fide sporting competition, as opposed to being bought, or given, or won in a contest where the other competitors were not also seeking victory. The former example is less clear: an arranged marriage, including one brokered through a professional matchmaker, might be the foundation of a fulfilling relationship. However, at least in many contemporary Western cultures, the value of a romantic relationship is understood to derive at least in part from its quality as being chosen by the participants thereto, and from the reasons behind those choices. 8
Contrast this with the relationship between markets and the goods distributed thereby. A shiny new car is a shiny new car, whether it is bought, gifted, inherited, or won at poker. This does not mean that we might not prefer to obtain it by one means rather than another. There is (I imagine) a particular joy to be had from driving a car that you won through great skill, instead of paying for it like everyone else. Equally, some may find satisfaction in saying that they bought their car with money they earned for themselves, rather than being gifted it by a rich parent. But in either case, this satisfaction is secondary to the value of the good itself, and this latter value is not contingent on the manner of its acquisition. Markets are tools for allocating goods whose value is not derived from the manner in which they are allocated. Some goods may not be appropriate for distribution in this manner; but for those that are, the manner of their distribution is itself irrelevant to their value. 9
Markets as doubly procedural
Third, markets are procedural, in at least two senses.
First, no agent is entitled to a particular substantive outcome from the market. When a buyer enters the market, he might hope to find a seller willing to sell at a price that he can afford. However, no particular seller is obliged to sell to him. (Market transactions, recall, are voluntary, and that voluntarism cuts both ways.) He may find, whether because others are willing to pay a higher price, or simply because sellers would prefer not to deal with him, that he is unable to buy what he wants. The market has left him without what he seeks, but this does not in itself give him grounds for complaint about the market's fairness. In markets, some win and some lose; and he has simply lost (Cf. Mill, 1989: 95)
Second, no agent has – at least in standard cases – a right to be judged in the market by any particular objective criteria. There are no generally applicable rules requiring market participants to make decisions on particular grounds. Markets respond to agent preferences without concern for the origin, propriety or wisdom of those preferences. If most consumers prefer to avoid hormone-treated beef despite there being no evidence of its posing health risks, then we can expect producers of hormone-treated meat to lose out; and it is no objection to their losing out that their product is in fact as good or better than the more successful alternative. Consumers prefer the alternative; and in the market, preferences are king. For many observers, this lends markets an attractive democratic quality. 10
This latter aspect distinguishes markets from professional examinations, public tenders and court trials. In each of these, the process is competitive, but the criteria for success are given by the nature of the institution: a decision based on individual taste or caprice would provide obvious grounds for complaint. Markets produce winners and losers; and it is not necessarily the case that those who win are the most deserving, by any standard independent of their success in the market. 11 The market's structuring principle is, in Nozick; (1974: 160) formulation, “from each as they choose, to each as they are chosen”. 12
These, then, are the key features of markets for my purposes: they are antagonistic, expressing an opposition of interests both vertically (between buyers and sellers) and horizontally (amongst buyers and amongst sellers); they are instrumental, allocating goods whose value is not intrinsically linked to the manner of their allocation; and they are procedural, guaranteeing participants neither a particular substantive outcome, nor an outcome tracking any particular objective criterion. 13
We might be skeptical of markets, so understood, whether in general, or in particular circumstances. We might worry, for example, that markets play an excessive role in determining the life prospects of individuals in capitalist societies. Because markets produce winners and losers, they often seem to license inequality. (Rawls, 1999: 242–51 and passim; Dworkin, 2000: 66) Because they do not guarantee any particular outcome, they may seem inappropriate for the distribution of goods that we regard as essential, whether for material subsistence, self-respect or community membership. (Walzer, 1983: Ch. 3–4) Because they render economic relationships wholly instrumental they epitomize the impersonal, socially isolating aspects of liberal society. (Walzer, 1990) We might prefer that benefits were distributed by some more egalitarian or communitarian mechanism. At a minimum, we will likely want markets to be backstopped by administered public provision of a social safety net. 14
However, my concern here is not the justification of markets, or their proper scope, but rather to understand certain judgments that we seem inclined to make that are internal to markets. I thus assume that it is through markets, having the characteristics outlined above, that at least some valued goods are distributed; and ask what it might mean to complain about unfair competition in those markets.
Competitive fairness as determination by inappropriate factors
Let me start with a very general conception of procedural fairness, as the determination of an outcome exclusively by factors of a kind that are appropriate to determining that outcome. This is the basic model of fairness underpinning John Rawls’ influential account of Justice as Fairness, and in particular the original position thought experiment. (See generally Rawls, 1999: Ch. 3) Obviously, so stated, this is largely formal. We don’t know what is fair or unfair until we know what kinds of factors are appropriate to determining a particular outcome. This is why, for Rawls, the original position itself matters less than the detailed arguments for the inclusion or exclusion of particular knowledge and features in it. 15 Knowledge of inter alia our native talents is excluded because these are ‘morally arbitrary’ – they are not the sort of factors that are appropriate determinants of the particular process Rawls is describing. (Rawls, 1999: 63–65, 116) It is our conception of the person as free and equal, in Rawls’ account, that gives substantive content to this formal ideal of fairness. (See most clearly Rawls, 1980)
Now let us assume, by way of initial approximation, that each of the examples numbered 1–6 above is a candidate case of competitive (un)fairness. Not that each is necessarily unfair, but that the complaints raised in each case are at least of the right kind, such that once we have identified the appropriate criteria for judging competitive fairness or unfairness, we will be able to apply them to adjudicate these cases.
Note, first, that each is an example of market relations. Each is also, at least structurally, a complaint of unfairness as described above. The complaint is not that a particular result has been reached, but that the factors leading to that result are objectionable, rendering the process, and hence the outcome, unfair. It is not simply that US cotton farmers are beating out their West African competitors, but that their success is because of public subsidies. It is not that European manufacturers are losing out to foreign competition, but that their competitors are winning because of their less demanding regulatory environment. And so on for each of the other examples discussed.
In each case, these complaints – whether we accept them or not – only make sense given two assumptions: first, that market competition is an appropriate mechanism for determining success or failure and distributing the relevant opportunities and benefits; and second, that when certain factors or practices affect that competition, this renders it unfair to those who lose out.
Let me therefore suggest, as a first cut, that competitive fairness is concerned with the particular factors that determine outcomes in economic competition. In Fair Competition outcomes are determined exclusively by ‘fair’ factors, which might include the skill and energy of the competitors, and their capacity to meet consumer needs. In Unfair Competition outcomes are to some extent determined by ‘unfair’ factors, which might include government subsidies, relationships with third parties, or particular strategies such as below-cost selling or slandering competing products.
This general notion, of outcomes determined by permissible and impermissible factors, captures, I believe, a very broadly shared conception of what it means for competition to be fair or unfair. This need not mean that, in particular cases, we will agree on which factors or strategies fall on either side of the line. Some may feel that differential pricing, for example, in which a product is sold at different prices in different markets, is unfair, while others will see it as unobjectionable. However, the disagreement is about whether this particular factor or strategy is fair or unfair, rather than the general idea of competitive unfairness, which I suggest will have the structure sketched above. We know what we mean when we say something is competitively unfair; we just disagree about what is and is not unfair in these terms.
Structurally, then, competitive fairness complaints look like other fairness claims. However, structure cannot take us very far. The model of fairness described above is purely formal. Adopting it simply shifts our focus, to ask what makes some factors ‘fair’, and others ‘unfair’? In Rawls’ thought experiment, it is the conception of the person, whether in turn grounded in substantive moral reasoning or a public political culture, that gives these ideas content. But whatever else it is, competitive fairness is not Justice as Fairness, and something else must fill the equivalent logical space.
The question of competitive fairness then becomes what explains why, as a general matter, a particular determining factor is fair or unfair? 16 In so far as competitive fairness claims apply within particular social institutions, markets, it seems implausible to regard them as morally basic. Rather, their explanation is likely to in turn invoke one of a number of familiar moral principles. However, as I explain in the next section, the most obvious candidates struggle to account for competitive fairness, even allowing for quite substantial divergences from our considered judgments.
But this is to jump ahead. Before we get there, three further points of clarification, and one caveat, are required.
First, claims of competitive fairness, like fairness claims generally, are rights claims. By this, I mean that they are claims that specific agents have on their own behalves against other agents or social institutions. They are not – or at least they are not presented as – aggregative claims, about the kinds of rules that will best serve social goals, or claims of general benevolence without any specific correlative rights holder. (For this conception of a right, Dworkin, 1977; Waldron, 1988: 87) This distinguishes them from consequentialist claims, which seek to maximize some impersonal, aggregate good. It also distinguishes them from various instrumental liberal views, which advocate specific institutional, including economic, arrangements because of their expected social consequences. 17 Competitive unfairness is presented as a wrong done to the individual competitor in their own right, rather than to the society as a whole.
Second, claims of competitive unfairness are not contingent on the violation of any other legal or moral right. This does not mean that antecedent rights violations will not render competition unfair: they may. However the two complaints are distinct. If another merchant steals my stock, and then sells it cheaply, undercutting me, I will likely have two distinct complaints: both that he has violated my property rights (stealing my goods) and that he has violated my right to fair competition (underselling me with goods he obtained for free). That the two wrongs are distinct is clear if we imagine that, instead of stealing from me, he had stolen from a third party. I then have no complaint of property rights violation, but my complaint of competitive unfairness presumably remains. In other cases – including the various examples set out above, there is no plausible antecedent rights violation: the complaint is about competition itself, and without that competition, there would be no complaint.
Third, albeit related, competitive fairness is one amongst a broader class of moral judgments we might make in and about markets. We might evaluate markets and market interactions by whether they respect property rights, satisfy consumer preferences, or facilitate economic growth, equality, or flourishing lives. A theory of competitive fairness is not a comprehensive account of the morality of markets. (Certainly, my approach prescribes neither a general institutional design nor a comprehensive business ethics.) Our economic institutions must meet a number of different standards, advancing plural goods, and subject to plural constraints, amongst which competitive fairness is only one. Complaints of competitive unfairness may thus, at most, identify pro tanto wrongs, not all things considered conclusions. 18
The caveat is methodological. Throughout this paper I take seriously intuitions many of us have, and judgments we make, about fairness in markets. Appeals to intuition are not themselves remarkable: as Williams observes, a critical test of any moral view is whether it “rings true”. (Williams, 1993, xv) However, we must be particularly careful with shared intuitions and judgments about market interactions, given the temptation for participants to invoke moral language to clothe selfish interests, challenging others’ pursuit of profit while justifying our own. Certainly, much fairness talk in markets is presumably in bad faith. Nonetheless, the fact that it is effective in challenging and justifying, and that we do try to distinguish valid from bogus uses of it, suggests that there is something here. More generally, by seeking a principled account of our judgments, I try to link them to some more general value, remaining open to revising or rejecting them if this cannot be done. 19 To the extent we should care about competitive fairness, it is not simply because many market participants and observers seem to. Rather, I argue, it is because it can be explained as a concern for freedom as non-domination, and that is something we should care about.
This, then is what we are trying to understand. There is a widely shared sense, amongst those who otherwise endorse market competition, that certain factors or practices affecting that competition render it unfair to those competitors adversely affected. The question that I seek to pose in this paper is, why should this be the case? What are we concerned about when we worry about unfair competition? In the next section I consider a number of familiar moral principles that bear a passing resemblance to competitive fairness, examining whether and to what extent they can make sense of it.
What competitive fairness is not
The previous part sought to identify, in an admittedly schematic and intuitive manner, competitive fairness, as an ideal that is commonly invoked to evaluate relations amongst market participants. At this point readers should hopefully both understand the kinds of judgments to which I am referring, and recognize them as judgments of a kind that they themselves are inclined to make (or at least to hear made).
In this part, I want to problematize those judgments. The hunch motivating this section is that many of those who make these kinds of judgments assume – to the extent they reflect on them at all – that they express one of a number of familiar moral principles, whether legitimate expectation, sporting fairness, efficiency, or equality of opportunity. My strategy is thus to consider each of these in turn, showing why they fail to explain judgments of competitive (un)fairness. It is this failure that renders competitive unfairness puzzling, a puzzle I will try to resolve in the final part.
Competitive fairness is not fair play or legitimate expectation
As a first approximation, we might try to understand competitive fairness as appealing to explicit or implicit shared understandings of appropriate market behaviour. When we complain about unfair competition, our complaint – on this interpretation – is that there is a ‘way that things are done around here’, and that the challenged behaviour conflicts with this.
This recalls the familiar ideal of fair-play, characterized by HLA Hart (1955: 185) as follows: “when a number of persons conduct any joint enterprise according to rules and thus restrict their liberty, those who have submitted to these restrictions when required have a right to a similar submission from those who have benefited by their submission.” Rawls (1999: 273) labels as legitimate expectation the slightly broader claim that “as persons and groups take part in just arrangements, they acquire claims on one another defined by the publicly recognized rules”. Perhaps there is nothing mysterious about competitive fairness. Perhaps it is simply legitimate expectations or fair play applied to the particular context of market relations. The particular features we identify as constituting unfair competition need not derive from any prior principle; rather, the principle is simply that those who play by the rules, whatever those rules may be, can demand that others do likewise, and that they reap the rewards those rules promise. And around here, this response suggests, the rules include restrictions on those things we characterize as competitively unfair.
This, however, doesn’t fit with the ways competitive fairness is invoked. In particular, this interpretation renders incoherent the use of competitive fairness as a critical standard, by which existing institutions are judged and calls for reform are made. Historically, for example, it is not the case that the practices prohibited by anti-trust and state aid rules have always been prohibited. Rather, by stages legislatures, courts, and states through multilateral treaties, have prohibited those various behaviours on the grounds, inter alia, that they constituted unfair methods of competition. (See e.g. Finger, 1991; Gerber, 1993, 2001) Explaining competitive fairness in terms of legitimate expectation thus risks circularity. Once these prohibitions are in place, it might make sense to understand them in terms of legitimate expectation; but this cannot explain the normative force of demands for prohibition in the first place. Conversely, other examples noted above are not currently prohibited by either domestic or international laws, but they are still discussed in terms of competitive fairness. If competitive fairness was simply an appeal to prevailing rules these critical invocations would seem deeply puzzling. The fact that they do not, and that even those who reject them do not do so by exclusive reference to existing rules, suggests there is more going on here.
This does not dispense with this interpretation, however. While it is clear that competitive unfairness has been prohibited by stages, and that not all candidate competitive unfairness is prohibited today, we might explain this as the law imperfectly tracking the shared understandings of market participants. I referred above to the way that things are done ‘around here’. This need not refer to legally enforceable rules. Rather, the same principle can be applied to the implicit understandings of traders, provided those understandings are sufficiently widespread, and that participants not only themselves respect those understandings, but expect that others do likewise. 20
However, the ways competitive fairness claims are applied in practice undermines this latter interpretation. Most obviously, those claims are frequently made in the context of international competition, between enterprises in very different communities with very different implicit understandings about how international commerce proceeds. This is perhaps clearest in disputes arising from the integration of China into the international trading system. Whatever other arguments we might advance for the application for fair competition principles in Chinese law, or towards Chinese enterprises, we cannot plausibly claim that these are already implicit in the relevant practices. Further, as a matter of interpretation, claims of competitive fairness are not generally advanced simply by reference to prevailing mercantile norms. 21 They are claims, not simply about how things are, but about how they should be, implying an appeal to some standard that is independent of prevailing norms. This, I suggest, is how those who make such claims understand them; so we must look for an explanation that reflects that understanding.
Competitive fairness is not sporting fairness
Another promising avenue might be to understand competitive fairness as expressing the same values that underpin fair competition in sports. Certainly, there is plausibly a genealogical relationship between fairness in market competition and fairness in sporting competition. The popularity of sporting metaphors in commerce (“the level playing field”, “a team player”) underlines this connection. 22
However, closer examination shows how these ideals come apart in practice. Showing this requires a detour here, to consider what sporting fairness is about. It is only once we have a handle on this that we can enquire whether and to what extent it might be helpful in understanding the competitive fairness that is our focus.
What, then, is sporting fairness? By sporting fairness I mean the sense of fairness that distinctively applies to sports and other practices whose goal is either (i) to recognize and publicly honour some particular excellence 23 or (ii) to provide an opportunity for the development of such excellences. 24 These are two plausible accounts of the goals of sports. 25 However, they also readily describe many other practices, from beauty pageants to song contests to pub quizzes and Nobel prizes. Judgments of sporting fairness are, as a rule, closely tied to the purpose of the relevant practice. Something is unfair, in this sense, to the extent that it advantages one participant over another in ways not reflective of that purpose. 26
This model of sporting fairness captures the unfairness of straightforward cheating, in which one participant breaks the rules of the competition. The practice seeks to develop and recognize particular excellences, and that purpose is frustrated where winning instead results from one participant breaking the rules. The unfairness of cheating can of course equally be understood in terms of legitimate expectations and fair-play. However, a teleological explanation makes sense of something the latter explanation misses: the use of fairness as a criterion in the reform, rather than just the application, of the rules of a practice, and as a criterion for criticizing behaviour that is not actually contrary to those rules. A sporting contest can be unfair despite its publicly acknowledged rules being scrupulously followed, if those rules themselves do not track the relevant excellence. The rule that beach volleyball players must wear swimsuits, for example, might be variously criticized, but one concern might be that it makes a willingness to wear a revealing uniform a condition for success, despite such willingness bearing no relation to the excellences the sport (presumably) seeks to recognize or develop. The move towards bonus points for tries in rugby arguably made that sport not just more entertaining, but also more fair, by ensuring that it was the physicality, speed, agility and teamwork, rather than simply the fly-half's kicking ability, that would lead to success. 27 The use of many performance enhancing drugs would be unfair on this account, regardless of whether they are expressly prohibited, because a win achieved through their use would serve neither to identify and honour, nor to promote the development of, the distinctive excellences of the relevant contest Similarly, such underhand but legal behaviour as running down the clock to force a draw in test cricket, or serving underarm in tennis, make sense as examples of unfairness on this account. 28
An important part of judgments of sporting fairness is therefore an account of what the particular goal of a practice is, and of which excellences it recognizes and develops 29 Disagreements about purpose will lead to disagreements about fairness. Consider, for example, the role of “trash talking” and psychological mind games in professional boxing. If we think that boxing is a contest of speed, strength, skill, and endurance, then we will see a boxer who gains an advantage through such psychological tricks as acting unfairly. If, on the other hand, we think psychological toughness, concentration and strength of will are amongst the distinctive excellences that boxing realizes, then we will see these mind games as a valid part of the sport, and a boxer who complains about them as simply a “sore loser”. 30 Further reflection on this point might lead us to conclude that professional and amateur boxing were fundamentally different sports, emphasizing different excellences, and in consequence engaging different criteria of fairness. 31
It seems important, in making sense of sporting fairness, that it applies to practices whose intrinsic purpose lies in the achievements of particular participants (or groups of participants), rather than the realization of shared, or wider social, goals. The complaint of an employee who is passed over for promotion is not a complaint of sporting fairness, because the goal of the corporation's employment and promotion practices is not (despite the language fashionable in human resources departments) to recognize and reward the achievements of individual employees, but rather to advance the collective project of the company's success. The two goals will frequently coincide: a corporation will often flourish by recognizing and rewarding the achievements of its employees, making such recognition a sensible instrumental goal. However, where according such recognition runs contrary to the company's interests, few would expect the company to choose the former over the latter. The company can be expected to design its promotion schemes and reward its employees with an eye to its corporate goals, not their intrinsic merits. Equally, the best professional wrestler cannot complain of competitive unfairness if he does not appear in the title fight, or win the championship: the purpose of professional wrestling is dramatic entertainment rather recognizing sporting prowess; and if the script requires that the champion lose – and even lose in ways that break the “rules” – a complaint of unfairness seems out of place.
Sporting fairness, so conceived, looks a lot like competitive fairness. Both apply to competition, and to the ways that particular conduct or features advantage some participants over others in that competition, and in turn the distribution of the goods at stake in that competition. If this is what sporting fairness is, then why can we not understand competitive fairness as an expression of the same idea? I here first set out the principal objection to this move, and then consider various possible rejoinders to it.
The objection emphasizes the instrumental role of competition in markets. Whereas the purpose of sporting competition is to recognize, honour or provide opportunities for the development of particular excellences, the purpose of markets is to coordinate economic activity and allocate resources. And while the purpose of sporting competition implies standards for what counts as winning in the relevant practice, no similar implications can be derived for economic competition. It is implicit in the idea of a race that the person who runs fastest should win, because winning just consists in running fastest (Sandel, 2013: 93–4) But the reasons why success in a market should turn on some factors, but not others, cannot so readily be deduced from the idea of market competition itself. Of course, given a further account of the social goals (efficiency, equality, liberty, rights etc.) that particular markets serve, we can derive standards for those markets; but then it is those social goals, rather than anything distinctive about market competition itself, that provides these standards. 32 More importantly, according success or failure to particular persons, while perhaps an unavoidable aspect of economic markets, forms no part of their purpose: competition in markets is instrumental to their ultimate purpose of allocating resources and meeting individuals’ needs. The fact that someone might “win” in market competition is simply a tool for motivating pro-social behaviour, rather than a goal intrinsically linked to the practice's structure and purpose. And while we might hope, with Hayek, that the market would reward those who best meet others’ needs, it can do so only insofar as those others choose to reward them.
Some might challenge this contrast between sports and markets, characterizing sporting competition as similarly instrumental. Heath, for example, interprets competitive sport as an artificial prisoners’ dilemma, in which participants seek victory at minimum cost, but are precluded from colluding by rules that serve social goals, including providing “a riveting display, in which spectators can see the frontier of human achievement being pushed back year after year”. Competition recruits participants’ selfish motivations for social benefits: “The difference between healthy and unhealthy forms of competition is that, in the former case, the external benefits outweigh the losses incurred by the competitors, while in the latter case they do not.” (Heath, 2014, 97) If this is what we think sporting competition is, then it can certainly be analogized to markets, in which competitors’ pursuit of private profit is recruited in the service of consumer preference satisfaction. This interpretation of sporting competition, however, struggles to explain much non-elite sport, where there is no third party beneficiary of the competition between participants. If anyone benefits from the contest between two mediocre middle-aged tennis players on an empty court, it must be the players themselves, so the efforts they make must be explained from their perspective. 33 Further, it conflicts with the commonplace judgment that the value of sporting victory depends, in large part, in defeating an opponent who is themselves making every effort to win. Even a young child does not want to think that their parent has “let them win”. 34 And we readily recognize that winning by cheating, even if that cheating goes undiscovered, somehow “cheapens” the victory, depriving not just the loser but also this dishonest winner of something valuable. If being recognized as the winner, as opposed to actually winning in a true contest, was participants’ goal, then these judgments would appear mysterious. That they do not suggests that sporting competition is more than just an unresolved prisoner's dilemma. Without competition, victory is impossible for either competitor. It is because the value of winning depends on how the win was achieved that a sport's distinctive excellences generate principles for how competition should proceed. 35
Rather than instrumentalizing sporting competition, we might instead hope to avoid this challenge by re-describing market competition in non-instrumental terms. For many people, after all, economic life is not a purely instrumental domain, in which they participate to obtain the resources required to realize their life's projects. Rather, economic activity is an important part of their life's project. It is through having, finding or fostering new business ideas, building up an enterprise, and competing with others in the market, that the entrepreneur finds fulfilment and self-expression. (Knight, 1923: 602–3) If a life of achievement is a valuable life, then the achievements of the market may perhaps as readily add value as do those of the sports field or the writing desk. That this is not always acknowledged by scholars likely says more about the particular concerns of scholars than it does about human nature or value more generally. If scholars found their good in the cut and thrust of the market then they probably wouldn’t be scholars. (Cf. Friedman, 2002: 9; Tomasi, 2012: 71–3) But this should not blind us to the fact that others do find their good in this way. Perhaps the function of competitive fairness is to ensure that the rewards of the market go to those who best express the particular excellences that the market facilitates. 36
Three objections to this move occur.
First, it assumes an implausible consensus about the particular excellences at stake in, and the intrinsic goods realized through, market competition. This objection is foreshadowed by the discussion of boxing above: if fairness is a function of the purpose of a practice, then we need some agreement about what that purpose is before we can apply judgments of sporting fairness to it. This is not of course a one-way process: it may only be through making and contesting claims about fairness that we clarify our understandings of that purpose, or that disagreements about fairness help us to identify an underlying disagreement about purpose. 37 However, it is at least plausible to aspire, in the context of sports and similar contests, to a shared account amongst participants of the purpose of the relevant practice. 38 It seems doubtful that the same is true of markets. 39 Even restricting ourselves to accounts that understand the value of markets in intrinsic terms, the goods realized might include the development of particular professional skills, meeting the needs of others, outwitting one's commercial opponent, building a lasting legacy, and so on. Each of these will likely point towards different criteria for what should count as fair or unfair conduct in the market. 40 (And this is ignoring the preponderance of participants who likely do understand their market participation in purely instrumental terms.) And of course, unlike sports, it is not open to those who disagree to exit and establish their own economy, as they can their own sport or sporting federation.
Second, it fails to explain why the appropriate institution through which to realize these goods is, and must be, the economic system as a whole, given its pervasive impacts on the capacities of all participants to realize their own projects, whether economic or otherwise. We might accept that the excellences of the market were equally valuable, and so equally in need of a venue for their realization, as the excellences of the playing field, the writing desk, the artists studio and so on. However, these latter excellences can be, and generally are, developed and exercised within discrete domains, both within the lives of their exponents, and within society. The boxing gym is a space apart from social life generally, and for those uninterested in it, it need not impact them, or impose particular costs on them. Why can the excellences of economic competition not be similarly realized through distinct, and limited, institutions? (Perhaps tournament poker might sate some of their needs?) And if, in fact, they can only be realized by organizing our economic lives around markets, and making all subject to them, then this seems an unreasonable demand to make simply because some find their good realized in this way. 41 Just as the barbarian's desire to realize a life of plunder, pillage and glory cannot justify the costs this imposes on others, so the capitalist's desire to realize his particular conception of the good cannot justify his demanding society's economic life be organized to facilitate this. (Cf. Knight, 1923: 612, 624)
Third, it ignores the fact that markets are not simply domains of achievement, but also mechanisms for the distribution of instrumentally valuable resources. We might reasonably think it worthwhile to constitute an institution through which individuals cultivate and demonstrate their capacity for entrepreneurship. Such an institution would accord recognition to those who successfully develop that capacity, including – assuming it is a competitive institution – to those who do so to the highest degree. However, the market, even if we understand its purpose as tracking this particular excellence, does more than simply recognize its achievements: it also confers significant economic rewards. And in doing so, it generates inequalities which, perversely, deprive the less fortunate of the very economic freedoms, to own, invest, and produce, than any such defence of markets must be built on. (Cohen, 1980. Cf. Waldron, 1988)
This highlights the limits of the argument from sporting fairness, even in a sporting context. It is an argument that an institution whose purpose is the cultivation and demonstration of particular excellences should track those excellences in the rules by which it determines what counts as winning. It does not tell us anything further, about the consequences of winning. In European professional sports, for example, ‘winning’ generally confers huge financial rewards on the most successful teams and players; whereas in American sports, economic rewards are more evenly distributed. An exclusive focus on sporting fairness cannot tell us whether one or other of these systems is preferable, because they are downstream of – and not necessarily connected to – the question of how we choose who wins and loses. 42
Similarly, it is a very big step from recognizing that some find purpose and a chance to develop and display distinctive excellences through participation in markets, to the claim that wealth should be distributed in proportion to their success in doing so. The former recognition provides no warrant for the latter distribution. Relevant here is Walzer's (1983) suggestion that the distinct meanings of different social goods imply different principles for their distribution. Sporting fairness is plausibly relevant to the distribution of honour and recognition; but it has no necessary connection to the distribution of income and wealth.
The upshot, then, is that there are echoes of sporting fairness in competitive fairness, and there are institutions that we might imagine which would make it relevant to the pursuit of distinctively economic excellences, but it has little relevance to actually existing markets.
Competitive fairness is not efficiency
A key objection to understanding competitive fairness in terms of sporting fairness was the fundamentally instrumental quality of market relations. We might then look to an instrumental ideal, that of economic efficiency, to make sense of fairness judgments in this context.
There is of course a close connection between competition and efficiency in contemporary economic thought. Given free competition in the market, consumers can choose amongst a range of possible suppliers the one that best meets their needs, and at the lowest cost. Fully efficient markets require to have various features, including: prices that accurately reflect the true social cost of production (no externalities); resources that can be readily allocated to their most productive use (total factor mobility); and sufficient producers and consumers, or at least few barriers that would prevent others entering the market, such that none can unilaterally determine the market price (negligible market power). Efficiency thus provides various criteria by which we can judge the quality of market competition, and the factors that should or should not be allowed to determine market outcomes. More generally, efficiency is, on many accounts, the point of the market. The reason we organize economic life around markets is because they are more efficient than the alternatives. We might therefore hope that we can construct an account of competitive fairness on the (broadly shared) recognition of the value of economic efficiency. (See e.g. Heath, 2014 for an account of business ethics in these terms)
There are, however, a number of problems with this move.
First, many of the judgments we in fact make about market fairness do not track the prescriptions of economic efficiency. In the domestic context, for example, predatory pricing is a prominent subject of fairness complaints; but it is usually only given significant barriers to entry that this will undermine long-run market efficiency. (Joskov and Klevorik, 1979: 227–31) Consumers need only worry about competitors being driven out of business if there are significant barriers to new entrants taking their place; but this is cold comfort for the victim of predation. In the international context, subsidies are a frequent target of complaints of unfairness; (e.g. WTO, 2001: para 149) but in a world where externalities (including, for example, environmental externalities) are a pervasive feature of economic activity, subsidies will often be the first best tool for internalizing these, and thus rendering markets more efficient. Complaints of competition from low-wage and low-regulation economies are similarly puzzling for anyone trying to understand competitive fairness in terms of efficiency: the gains from trading with economies with different relative costs form the crux of the economic case for trade. (Bhagwati and Srinavasan, 1995) This of course need not be a critical objection to reconstructing competitive fairness in terms of efficiency: it may simply be that these are mistaken invocations of competitive fairness. 43 Any reconstruction of an existing practice in principled terms will likely require revising some of our pre-theoretical judgments. However, the extent of the mismatch must surely count somewhat against this approach. If there is an alternative explanation that better fits our considered judgments then this will count in favour of that alternative. (Any alternative need not deny that efficiency is a central virtue of markets: there is a big difference between claiming that markets should be efficient and that efficiency is all that matters in their evaluation.)
Second, in the context of international markets, we might wonder from what perspective objections of inefficiency should be made, and in particular, whether our concern should be national or global welfare. From the perspective of an importing country, aggregate national welfare will be improved by importing products that are available on international markets at prices below the cost of domestic production. This is true, regardless of the reasons why goods are available at these prices. So, sticking with subsidization, if a foreign government chooses to subsidize its producers, an importing country will achieve precisely the same welfare gains by importing their subsidized outputs as they will by importing products that are cheap for other reasons. (Sykes, 2005: 92) It is only by adopting a global welfare perspective that such subsidized imports appear inefficient. Yet, in the context of debates about trade fairness, it is national rather than global interests that are generally at stake, and the language of unfair competition is frequently applied to challenge practices that clearly enhance the aggregate economic welfare of the complaining state.
This links to the third objection to understanding competitive fairness in terms of economic efficiency, namely the distinct perspectives that these two ideals adopt. By perspective here I mean the agents on whose behalf, or in whose name, the relevant ideal is invoked. Economic efficiency is typically invoked on behalf of a community as a whole. When we say that markets are efficient, we mean that the sum of the welfare of all agents affected will be maximized where an economic system is organized on a market basis. Like utilitarianism (to which it is closely aligned), efficiency adopts what Sidgwick (1874) memorably labels ‘the point of view of the universe’ or, in Thomas Nagel's (1986) terms, the ‘view from nowhere’. If our concern is efficiency, so understood, then we must deny any significance to the particular goals of particular agents, except in so far as these contribute to this welfare calculus. (Sen and Williams, 1982: 5–6)
By contrast, as noted above, claims of competitive fairness are rights claims, made from the perspective of the disappointed competitor. The complaint is not that the relevant behaviour has harmed the community as a whole, but rather that it has wronged the particular competitor. The competitor claims a right to participate in the market on fair terms, and that right has been violated. When we invoke competitive unfairness, we don’t simply assert that a wrong has been done: we claim that a wrong has been done to a particular agent. Of course, the advocate of economic analysis can reply that the competitor simply acts as a proxy for the society as a whole, or that our individualist and rights-based intuitions reflect a deeper aggregative and consequentialist moral reality. (See e.g. Smart, 1979: 113–7) However, that move seems more plausibly to motivate a rejection than an explanation of competitive fairness. Certainly, where our intuitions about competitive fairness conflict with the teachings of economic efficiency, it suggests that we should reject the former in favour of the latter. That may indeed be the better view: however, it undermines any claim that efficiency provides an explanation of competitive fairness.
Another response to this problem of perspectives might invoke Rawls’ distinction between the justification of a practice and the justification of moves within that practice. (Rawls, 1955) The justification of the practice of competitive fairness, on this view, would be efficiency, and its perspective aggregative; but once that practice is established, complaints adopt an individualist, rights-based perspective. 44 However this effectively reduces the efficiency explanation to the fair play explanation discussed earlier 45 – the competitor's complaint is about a violation of the prevailing standards, rather than about inefficiency per se. It thus faces the difficulties already canvassed there. We might think (for aggregate welfare reasons) that it would be better if there was a practice whereby all agents conducted themselves exclusively in ways that conform to economic efficiency; and if there was such a practice, then this might accord participants rights claims where others fail to do so; but no such practice actually exists, so there is currently no basis for these kinds of claims. 46 I can only sensibly complain that I in particular am injured by an inefficient subsidy if there is a general practice prohibiting inefficient government policies, to which practice I can appeal in making that complaint.
Competitive fairness is not (formal) equality of opportunity
Another widely accepted ideal to which competitive fairness might be linked is equality of opportunity, the idea that persons with the same native talents, or perhaps the same developed skills, should have the same chance to occupy valued positions or obtain particular rewards. Certainly, “fair competition” in education, employment, government office and many other settings will be a prerequisite for realizing equality of opportunity. More generally, both competitive fairness and equality of opportunity adopt a procedural approach to distribution, with outcomes judged by the process by which they were reached, and the factors that were relevant to determining them, rather than on the outcomes themselves. Under both equality of opportunity and competitive fairness we can expect, and accept, that different agents will experience different outcomes, provided those outcomes are a function of factors of the right kind. However, despite this structural similarity, competitive fairness and equality of opportunity are in fact quite different ideals. There are at least five reasons why we cannot understand one in terms of the other.
First, equality of opportunity depends on a prior account of the factors that should or should not be relevant to determining success in a particular context. In an employment context, for example, equality of opportunity can be operationalized through an account of the specific skills and capabilities that are relevant to filling a particular role. We move from a specification of the relevant role, to a specification of the ideal candidate for that role, and then judge the various actual candidates by this standard. This requires being able to specify, with some degree of precision, the criteria that are relevant to determining fitness for a particular role. However, as noted above, markets are not organized around formal criteria; the criterion for success in the market simply is success in the market. We can, and often do, seek to exclude certain bases for decisions in particular markets (most obviously, anti-discrimination obligations imposed on employers and service providers), but we do not impose a general standard against which competitors are judged, except the standard of success itself. We can try to ameliorate the impact of specific perverse or unjust preferences (racist, sexist, homophobic etc. preferences) on the market, but we cannot assure anyone that their success will necessarily depend on their capacities, measured against any particular objective standard. 47
Second, the specification of criteria that equality of opportunity assumes is not itself a morally neutral task. Taking a popular example (Friedman, 2002: 111–3), consider a shop that – simply because of the character of the local population – caters primarily to white, racist customers. In specifying the role of shop assistant, we would likely include an ability to relate effectively to customers. Given the attitudes of the customer base, it is at least plausible that we would conclude that a white person would be more effective in this role than would a non-white person. Nonetheless we are likely to reject a person specification that privileged white candidates as violating even formal equality of opportunity. The value of formal equality of opportunity is in expressing some sense of basic human equality; and when a purely functional specification of its demands so obviously contradicts that basic equality, we are likely to prefer a moralized over a non-moralized interpretation of the factors that are relevant. Formal equality of opportunity is not, therefore, simply the idea that those best fitted to a role should succeed; rather, specifying what it means to be “best fitted” itself depends on a prior account of which factors are or are not appropriately relevant to this inquiry. Yet, as we have seen above, the problem of specifying a conception of competitive fairness is precisely the problem of specifying these relevant and irrelevant factors; and – as explained below – it cannot be the case that the same factors are relevant to both.
A third difficulty lies in delineating the agent whose opportunities it is sought to equalize. When addressing individual persons, competing for employment or educational opportunities, we can roughly distinguish between characteristics of the relevant agent (skills, knowledge, experience, past achievements, future potential) which should be considered in making a decision, and characteristics external to the agent (family connections, wealth and status) that should not. 48 However, if we consider the features that determine agents’ success in the market, we recognize that many are external to the competing agents. Consider, for example, two mass-market high-street clothing manufacturers, one based in Dublin, and one in Dhaka. Assume that both have the same native talent, education, energy, innovation and entrepreneurial drive. Factors external to them, most obviously the cost of labour in their respective locations, will give one an insurmountable competitive advantage over the other. No matter how well you run your business, in today's world it simply is not possible to produce mass-market clothing at competitive prices in a high-income economy. The manufacturer in Dublin will almost inevitably fail in the market, through no fault of his/her own. Equality of opportunity is precluded by the facts of the international economy.
A fourth difficulty linking competitive fairness and equality of opportunity lies in their differing scopes of application. While by no means unanimous, exponents of equality of opportunity commonly apply it over a restricted population, most often members of the same political community. A French person should enjoy equality of opportunity with other French persons (or, perhaps, other French residents). It is no objection that they do not enjoy identical opportunities as a citizen of the United States, or Japan, or Burkino Faso. (Although contrast Miller, 2007: 62–68; Caney, 2005: 122–4) Yet complaints of competitive unfairness are not generally limited in this way; complaints of competitive unfairness are commonly made between agents in different jurisdictions who clearly do not enjoy equality of opportunities more generally. If equality of opportunity applies only within political communities, while competitive fairness can apply across political communities, then it is difficult to see how one could be derivative of the other.
A final objection, which seems particularly problematic, relates to the agents entitled to competitive fairness and equality of opportunity. Equality of opportunity applies to individual human persons. So applied, it is explicable as an expression of their basic moral equality. If we are all moral equals, then we should enjoy the same opportunities to access valuable benefits, unless there are good reasons to the contrary. Sometimes, variants of equality of opportunity might be applied with regard to particular group identities (e.g. geographical or nationality quotas in recruitment for federal, international or supranational bureaucracies), but in this case this is usually explicable in terms of either the equal status of members of the relevant groups or (more controversially) the equal status of particular morally significant collectives, including states, nations or peoples. Competitive fairness, by contrast, is applied to enterprises, whether corporations, firms or individual entrepreneurs. These may differ in their organizational and legal structures, their relations with other enterprises, their histories and internal cultures and practices, and their scales, whether in terms of assets or employees. The only thing they necessarily have in common is that they are competing in the same market. Given this, we might wonder what underlying feature or features they share, and share to the same extent, that would lead us to think of them as equal for these purposes. 49 When applied to persons, equality of opportunity supervenes on moral equality; but there is no shared feature on which it can supervene in the case of enterprises.
What competitive fairness might be
Here, then, is the puzzle. Demands for competitive fairness are a familiar part of our economic lives. They have features in common with a number of other familiar principles, but do not seem plausibly to reduce to any of these. At the same time, it does not seem plausible to regard them as morally basic. They are, after all, claims made in relation to a particular social institution, the market, which is itself contingent. It would seem odd if something were at once morally basic (not reducible to or explicable in terms of any other moral principle) and uniquely relevant to one contingent social institution, somehow emerging from nowhere to govern this particular institution and no other.
Rather, if competitive fairness is a genuine value, then we should expect it to be related to some more fundamental principle. That more fundamental principle need not look especially like competitive fairness. However, it should be possible, by combining that fundamental principle, whatever it might be, with the particular circumstances of market competition, to generate the principle of competitive fairness, as a specific principle appropriate to those circumstances. (Cohen, 2008: Ch. 7) I want to suggest, in this final part, that the republican ideal of freedom as non-domination can play this role.
From non-domination …
What, then, is non-domination, and why does it matter? (The presentation in this and subsequent paragraphs generally follows Pettit, 1997, 2012, 2014) Non-domination constitutes a distinctive interpretation of the value of freedom, distinct from either negative (non-interference) or positive (self-mastery) liberty. Non-domination is the status of being protected from subjection to arbitrary interference. It is the status of being a free agent, in a world of free agents. If another agent has the power arbitrarily to interfere with my choices, then I am to that extent dominated by them, and hence unfree. This holds whether or not they in fact choose to interfere: it is the possibility of interference, and the absence of protection therefrom, that places them in the superior, and me in the inferior, position. They may allow me my own choice, but I have that choice only because they allow it; and the fact that they might choose to interfere means that I must always bear them in mind in making my choices. Non-domination thus describes an ideal in which we interact as free and equal, possessed of a certain status and self-respect, each able to make our own choices, without fear or deference or dependence on the benevolence of another.
Two concepts here require a little elaboration before we try to connect non-domination and competitive fairness: interference and arbitrariness.
Interference denotes the capacity to alter the options available to the dominated agent, whether through directly closing off those options, or by altering the costs and benefits that attach to them, in ways that make things worse for that agent. (Pettit, 1997: 52–55) Further, it imports an intentional quality: merely accidental or incidental foreclosing of options does not constitute interference. Nor, on this view, am I subject to interference where an option is foreclosed, not through the intentional agency of some particular others, but through the impersonal combination of a host of unrelated and individually insignificant choices: “There is a big difference between the case where you deny me the option of reading the newspaper, as in invasive interference, and the case where that option ceases to be available because most other people come to rely on electronic sources of news, put the newspaper out of business, and thereby vitiate my choice of reading the newspaper.” (Pettit, 2012: 40) I have lost the option of reading the newspaper, but no particular others have taken it from me. For the exponent of negative liberty, I am either free to read the newspaper, or I am not; but interference is a relational concept, requiring that the foreclosing of my option is attributable to the intentional agency of another.
Even with this intentional agency condition, interference remains a capacious notion, capturing many ways that options might be foreclosed without raising any moral concern. In a world of finite resources, each of our choices necessarily forecloses some options from others; and in a world of social competition, foreclosing others’ choices is frequently our goal. Non-domination does not condemn all powers to interfere, but only asymmetric power, and the ways this enables arbitrary interference. Arbitrary here denotes the fact that interference is at the discretion of the interfering agent. The agent interfered with cannot counter that interference, nor is there any institutional mechanism to ensure that it tracks the interests of that agent. Many trivial examples of interpersonal interference can thus be discounted, on the basis that the agent interfered with can effectively respond, whether in kind or otherwise, to the interference. You can prevent me getting the last bottle of milk, but I can equally prevent you getting the last block of cheese. Our mutual capacities for interference are (approximately) symmetrical, and that symmetry prevents either of us dominating the other. (Pettit, 1997: 67; 2008: 108–10) 50 Where power remains unequal, as in the case of the individual/state relationship, we may rely on institutional protections against arbitrariness. (Pettit, 1997: 65–6) Thus, through democratic control, the separation of powers and constitutional checks and balances, we (ideally) ensure that the state's power is exercised only in ways that track the interests of the citizen. The state, through its laws, interferes with our choices, and the individual citizen has little capacity to directly check that interference, but the institutions through which state power is exercised (hopefully) prevent it dominating us, ensuring power is exercised only in ways and for ends that we share. (Pettit, 2014: 109)
We are dominated to the extent another has the capacity to arbitrarily interfere, regardless of whether they in fact do so. We are unfree when another has arbitrary power over us, even if their benevolence leads them to leave us well alone. It is only where that arbitrariness is checked, through individual counter-power or shared institutions, that we are free in the sense of being undominated.
Republicans defend non-domination as preferable on various grounds to competing positive and negative conceptions of freedom. (See generally Lovett, 2018) My goal is not to provide a new argument for non-domination, but rather to examine how far it can make sense of competitive fairness. Let me therefore highlight two implications of this concept that seem particularly relevant to this.
First, because it asks why and how particular choices are foreclosed, rather than simply comparing opportunity sets, non-domination allows us to distinguish between situations in which the same outcome results, depending on how that outcome came about. Interference is unobjectionable, unless accompanied by inequality and arbitrariness. We are regularly precluded from pursuing options that we might value. It is the circumstances under which others have the power to foreclose those options that republicans suggest should concern us. This allows republicans to identify unfreedom in the absence of actual interference. However, it also allows them to distinguish between instances of interference that are arbitrary, and hence troubling, and those that are not. Interference constrained by appropriate standards and institutions is not unfreedom.
Second, domination and non-domination are relations that may obtain between agents of different kinds. Persons might be dominated by the state, but also by their fellow citizens. Equally, it is not only persons that can be dominated: states may be dominated, whether by other states, by multinational enterprises, or even by particularly powerful individuals. (Pettit, 2014: Ch. 6) And while non-domination is, for republicans, the status of the free citizen, we might equally worry about a state dominating non-citizens outside its borders, or one enterprise dominating another. Non-domination thus has particular relevance for thinking about the global economy, characterized by diverse relations among diverse agents.
There are various ways we might link non-domination, so understood, with market relations. Some have emphasized the potential for domination in particular relationships, including especially employment. (Cabrelli and Zahn, 2018; Anderson, 2017. Cf. Pettit, 2014: 105) Others highlight ways inequalities resulting from markets might create opportunities for domination in political contexts. (Casassas and De Wispelaere, 2016; Dagger, 2006: 158–9) For others, markets (and in particular competitive markets) potentially realize the ideal of non-domination, through ensuring agents have choices, and options to exit particular relationships, and to transact as equals. (Pettit, 2006; Taylor, 2013, 2019) Anderson interprets the early-modern free-market thinkers as motivated as much by non-domination as efficiency, and the subsequent growth of industrial capitalism as fundamentally changing markets, from liberators to instruments of domination, particularly in the employment context. (Anderson, 2017) However, my suggestion here is that non-domination can also help us make sense of judgments that we make about relations between competitors, and in particular that it can generate a plausible account of competitive fairness.
… to competitive fairness
The general form of that explanation claims that instances that we identify as unfair competition are in fact cases of domination, with or without attendant interference. To be successful, it must explain why this is the case, and how it can help us distinguish acceptable and objectionable horizontal market relations.
First, however, note how understanding competitive fairness as non-domination avoids the principal objections against other explanations above. Unlike legitimate expectations, it explains how competitive fairness serves as a critical standard for judging existing norms and institutions. Unlike sporting fairness, it is not contingent on consensus about the particular values that the market realizes; whatever markets are for, domination between participants is objectionable. 51 Unlike efficiency, complaints of domination have the right structure: they are rights claims, that a particular agent is wrongfully subject to domination. Unlike equality of opportunity, they are not contingent on identifying a particular morally relevant shared feature among competitors: non-domination makes sense even between agents of very different kinds, amongst whom it would make little sense to assert moral equality.
But avoiding these objections is of little value, unless we can also reconstruct the distinctive judgments that we above identified with competitive fairness. We do this by examining domination along two dimensions: amongst market participants; and between states and market participants.
Consider, first, what it means for one market participant to dominate another. Market participants obviously interact intensively, and through their actions will frequently restrict options open to their competitors. However, in competitive markets, there is a symmetry between competitors: I may restrict your options, but you can equally restrict mine. You might beat me to the sale today, but I’ll win tomorrow. Equally, while I may be able to attribute the foreclosing of a particular option to the actions of a specific competitor (“I didn’t get that contract because John offered a lower price”), in a competitive market the prevailing price – which ultimately defines the opportunity set of both buyers and sellers – is a function of the joint actions of all market participants. All competitors, on both sides of the market, are price-takers. The restrictions they face as a result of competition are thus experienced as analogous to a natural fact. None is dominated by any particular other. 52
Where, by contrast, one or more participants has power to unilaterally determine (or at least significantly alter) conditions in the market, this ceases to be the case. Now some participants become price-makers, while others remain price-takers. The former exercise power over the latter, and the latter are powerless to respond. Market conditions – and in consequence, participants’ capacity to flourish, or even economically to survive – are intentionally authored by the enterprise with market power. Smaller firms must tread carefully as a result, lest those conditions be turned against them. If they are, absent institutional protections, there is little they can do to respond; they are dominated. 53
This, then, is the first way non-domination might explain competitive fairness. If the cases we identify with competitive unfairness involve the exercise by one competitor of market power over the other, then a concern with domination would explain why we found these cases troubling. Returning to the examples in Part I.1 above, this is plausibly what is at stake in Dumping, Exclusivity and Scale. This is not to say that these are necessarily cases of domination – they are described in sufficiently general terms to leave this open. However, in each, conduct is at least suggestive of market power on the part of the ostensibly unfair competitor. Price differentiation, for example, may indicate that the seller enjoys sufficient power in the higher-priced market to make it rational to sell cheaper in another market, rather than risk reducing the prevailing market price there, and that his doing so reflects an effort to alter the market structure in the target market, to other competitors’ detriment. (The fact that the undersold competitor cannot simply divert their own production into the higher-price market further suggests that this is the case.) Where this is indeed the case, Dumping evidences domination, and is properly subject of complaint and institutional control. Similarly, Exclusivity, in so far as it reflects the market power of big breweries to lock out new competitors, is also plausibly a case of domination, and objectionable as such. These are not simply cases of market-mediated harm. Rather, they reflect harm through choice, by an agent that is neither disciplined by market nor regulation, nor subject to any remedy or response from the victim or others in their situation.
One caveat here: These, and the other examples discussed, are predominantly cases where there is – or at least may be – actual interference; yet domination is distinguished in part by its obtaining where an agent is exposed to the possibility of arbitrary interference, whether or not such interference takes place. 54 This might distinguish competitive fairness from non-domination: perhaps the former requires actual interference? However an alternative reading suggests that the particular instances of interference, and the complaints of unfairness they evoke, alert us to the risk of domination in particular relationships. If A does arbitrarily interfere with B, this demonstrates that A can arbitrarily interfere with B. The incident of arbitrary interference is evidence of the relationship of domination. It is perhaps unsurprising that actual interference, as opposed to domination, generates the loudest complaints: those who are dominated, after all, have every reason to keep their heads down. And the institutional remedies that are demanded in the face of these incidents in turn modify those relationships. The incident of arbitrary interference is the opportunity for highlighting, and remedying, domination.
In Dumping, Exclusivity and Scale, it is private market power that enables one competitor to arbitratily interfere with another, without being themselves constrained by the disciplines of the market. In others, however – Subsidy, Regulation – public power does the work. In each of these, the complaint is that choices made by states have improved the competitive positions of their own enterprises, at the expense of others. Of course, the exercise of public power alone cannot be enough to constitute either competitive unfairness or domination. States and laws are necessary enablers of both non-domination and functioning markets. 55 Democracy, constitutionalism, and checks and balances serve to prevent them becoming loci of domination. However, in our examples it is significant that the public power complained of is usually that of a state other than the complainant's. It is foreign subsidies, and under-regulated foreign producers, that are complained of. This matters because the political and constitutional protections on which we standardly rely to restrain state domination are usually available only to citizens / residents of that state. The state's impact on outsiders is a domain of untrammeled discretion, with those outsiders denied voice or influence. Domestic constitutionalism does little to prevent states dominating outsiders. 56
This need not mean that every choice of a foreign state that affects us, or limits our options, constitutes interference, nor that we are all dominated by each and every foreign state with power to affect us. However, it does require us to think about when and how that power might be restrained. At least three options stand out. The first denies that there is a significant difference between an agent's relationship with their own state and another state, implying a need for political representation and constitutional protection in cosmopolitan, all-affected terms. (See e.g. Abizadeh, 2008) 57 A second emphasizes the horizontal relationships between states, as well as the vertical relationship between state and person, relying on our own state's capacity for response to check the arbitrary power of others over us. 58 The third emphasizes the role of international institutions in checking individual states' power of interference with outsiders. (Pettit, 2014: 165–8) 59
But what exactly must those mechanisms – whether vertical, horizontal or hierarchical – do to prevent domination? It cannot be that every action a state takes which has the potential to affect outsiders must be restrained, or required to track the interests of those affected outsiders. To demand this would be to unduly restrict those states themselves, and thus interfere with the legitimate projects of their citizens, and the state's efforts to track those citizens’ interests as it exercises power over them. 60 Recall, not every foreclosing of options constitutes interference: it is restriction through intentional, or at least responsible, agency that is our concern. Living together means affecting one another: it is when we exercise power over one another that concerns of domination arise. Of course, applying intentionality criteria to states raises thorny questions about how a collective agent can intend anything. I won’t try to address these here. 61 However, we can at least plausibly hope to distinguish between public measures that make particular individuals, groups, institutions and activities their focus, and those that affect these only accidentally or incidentally. 62 It is only the former that require to be checked (although presumably subject to an appropriate de minimis threshold).
Faced with public power that crosses borders, then, we require mechanisms that can both distinguish interference from other market-mediated effects, and ensure that such interference, to the extent it is present, tracks the interests of those subject thereto. Where state policies affect others, without interfering with them, those others have no basis in non-domination for demanding that these track their interests. Where they do constitute interference, we need a further account of what it means to track subjects’ interests, in circumstances where these presumably include the state's own citizens and outsiders in diverse states and industries, whose interests are frequently directly at odds. Between symmetrical agents (e.g. participants in a competitive market), we can avoid this question, relying on their mutually checking each-other. In other cases, agents may not be symmetrical, (e.g. dominant enterprises in monopolistic markets) but we can require them to act as though they were. When faced with public power, however, symmetry is neither possible nor desirable: the value of states lies in large part in their being more powerful than other agents within their domain, so what is required is mechanisms and standards for restraining that power. 63 Those standards will define the proper uses to which such public power can be put. At the margin our concern for non-domination thus raises distinct questions of social and global economic justice, albeit with a different focus than liberal distributive justice. 64
Returning to our paradigm examples, this suggests they may – but need not – exhibit competitive unfairness. Not every subsidy that affects trade will be competitively unfair: it is only where subsidization constitutes arbitrary interference, and thus evidences domination, that it is properly characterized as unfair. Whether it constitutes interference depends on both whether it is somehow addressed to those affected outsiders, seeking to advantage subsidized enterprises over them in particular, and whether it has a meaningful impact on their opportunities. (i.e. whether it actually affects prices) Whether it tracks their interests depends on the purposes for which it is applied, and how far these are justifiable having regard to all whose interests are affected. Non-domination requires mechanisms of some kind to ensure that it does, and where it does not, we should properly regard it as competitively unfair. Similarly, variations in regulations from state to state may constitute interference, as one seeks to undercut competitors and conditions in another; or may simply be incidental side-effects of national choice; where they constitute interference, we must examine how far they track the interests of all those affected, examining the justifiability of such competitive regulation, which may look different depending the characteristics of the particular states or industries involved.. And competition with low wage economies may raise concerns for domination, where wage differences reflect the kind of unrestrained and intentional agency that non-domination tracks; or may simply reflect differing conditions in different countries, something closer to a brute fact confronting and conditioning the opportunities all relevant agents in both high- and low-wage settings. If the former, this view suggests, it may be competitively unfair; if the latter, then not.
The analysis in cases involving public power is thus somewhat more complicated than amongst market participants. Between market participants, we can hope to restrain the application, if not the existence, of unequal power. Where states enter the picture, unequal power is unavoidable, and valuable. We can seek mechanisms to check its arbitrary use, but this requires engaging not simply with the fact of power imbalances, but also with the proper uses of that power (and who gets to decide, and how). Disagreements on the latter will likely feed back into disagreements about what counts as competitive unfairness. Nonetheless, by emphasizing relationships of power and the capacity for arbitrary interference, non-domination provides a more constructive framework for that disagreement than simply invoking the language of “fair competition”.
Conclusion
Non-domination, then, suggests a way of thinking about competitive unfairness that can reconstruct many of our intuitive judgments (and existing laws), while also suggesting criteria for evaluating novel and contested cases. It avoids many of the objections that I noted against alternative explanations, while offering an explanation of what competitive fairness is about, and why it might matter.
Understanding competitive fairness in terms of non-domination might lead us to clarify or revise our intuitive moral judgments about particular practices. This may in turn have implications for how we assess, interpret, and potentially revise, key legal disciplines to which competitive fairness seems relevant. Appropriate lines of future inquiry might thus include examining the WTO rules on anti-dumping and subsidies, bilateral treaties on investment protection, and domestic rules on anti-trust, restraint of trade and intellectual property, from this perspective. The institutional challenge lies in identifying both appropriate standards of market-affecting conduct, and mechanisms to restrain public and private power in accordance with those standards.
That, however, cannot be the end of the story. I have focused on competitive unfairness in relative isolation, seeking to identify the pro tanto wrong that complaints of competitive fairness express. But economic interactions do not take place in isolation; and if our concern for competitive fairness reflects a more general concern with non-domination (as it presumably must), then it need not follow that policing particular instances of competitive unfairness is the best way to address it. We may find other issues of market-mediated domination are more pressing. We may need to balance concerns for domination with other market-relevant values including efficiency, welfare, equality, rights. And that balance may look quite different across different settings: some domination in the market may be remediable at reasonable cost to these other values, while some must be accepted as the price of realizing efficient allocations, or poverty alleviation, or some other goal.
Even focussing narrowly on non-domination, competitive fairness is one concern amongst many, and perhaps not the most pressing. Complaints of competitive fairness take the role of the market for granted, but often the existence of markets, or the marketization of particular goods, will expose many to domination. So non-domination implies limits on the scope of markets, as well as conduct rules within them. (Anderson, 2017; Dagger, 2006) Equally, if we are to focus on domination within markets, it is not enough to identify cases where power is exercised by some over others. As noted above, we must also ask what are the appropriate purposes for which that power may be exercised. We can assume that the private profit of another market actor will not usually constitute an adequate justification. But the protection of marginalized or less advantaged groups, through for example the promotion of industries on which they rely for employment, might. If those industries fail, then former employees may be particularly exposed to domination. Equally, if we consider the ways that poorer countries are exposed to domination by the rich world, we should recognize the economic advancement of poorer countries as a justification for interference with competitors. Imposing environmental and labour conditions on imports is clearly interference with the options of exporting industries and states; but it may also be the best way to address domination suffered by those working in, or living near, those industries. And if it is only arbitrary interference that is to be constrained, then the various institutions through which power is checked must be tailored to draw these distinctions, and subjected to appropriate controls themselves, to ensure the guardians against domination do not themselves become its agents. Making sense of all the various ways markets, regulation and domination might interact is beyond the scope of this paper. However, if the argument advanced above holds, and non-domination is indeed the best way to understand competitive fairness, then this is a challenge faced by those who would invoke competitive fairness, whether as a complaint or a justification.
Footnotes
Acknowledgements
I am grateful for helpful comments received from participants at the following events where earlier versions of this paper were presented: IVR UK Branch Annual Meeting 2016 (Leeds) and 2017 (Sheffield); Queens University Belfast School of Law Research Seminar 2018; ECPR Joint Sessions 2019 (Mons) Workshop on “Sovereignty Justice and International Law”; Dublin Political Theory Workshop 2019 (University College Dublin). Alex Green commented on an earlier unpublished paper from which many of the questions in this paper emerged. Brian Flanagan and Simon Palmer read and commented on my discussion of sporting fairness, helping me to understand how sport works. Detailed and helpful comments from two anonymous reviewers have significantly improved the paper. All errors are of course my own.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.