2.1. The standard model and its problems

In welfare economics and in undergraduate economic textbooks, competition standardly means perfect competition. A perfectly competitive market is defined as one that has many buyers and sellers, free entry and exit, perfect information, zero transaction costs and homogeneous goods (Mas-Colell et al. Reference Mas-Colell, Whinston and Green1995: ch. 10).Footnote ³ In perfect competition, no individual firm can influence the market price. Equilibrium price is equal to marginal cost, there is no economic profit and the allocation of resources is Pareto efficient.

This model is theoretically useful. But there are two reasons why it is unhelpful for thinking about justice in pricing. First, the model eliminates the very phenomenon it purports to explain. In a perfectly competitive equilibrium, there is nothing for competition to do. Every firm already knows the optimal price. There are no errors to correct, and no opportunities to discover. As Hayek (Reference Hayek1948: 92) observed, if the state of affairs assumed by the theory of perfect competition ever existed, it ‘would not only deprive of their scope all the activities which the verb “to compete” describes but would make them virtually impossible’ (see also Kirzner Reference Kirzner1973: ch. 2).

Second, because real markets virtually never satisfy the conditions of perfect competition, a normative standard built on that model will condemn almost everything that happens in those markets. Ignorance and transaction costs are ubiquitous, and this by itself ensures that real-world prices will systematically deviate from the perfectly competitive benchmark. The most sophisticated normative account based on the idea of perfect competition is Joseph Heath’s ‘market failures’ account. Heath argues that firms have an obligation to behave as though markets were perfectly competitive. But even Heath does not say that real-world prices are unjust. He grants that the concept of a just price is coherent, but holds that the market-clearing price is arbitrary from the standpoint of distribution, and so is to be defended on grounds of efficiency rather than justice (see Section 3). Heath recognizes that a standard built on perfect competition would make every real-world price presumptively unjust, and he avoids that absurdity only by abandoning the evaluation of individual token prices as objects of transactional justice. The argument of this paper is that we would do better to abandon the model of perfect competition instead.

2.2. Competition in the just price tradition

Some authors have suggested that it is anachronistic to read competitive conditions into the just price tradition. The Scholastics spoke of ‘common estimation’ and ‘current prices’, not of competitive markets, and we should not project a concept they lacked onto their work. Elegido (Reference Elegido2009: 32) makes the point sharply: ‘The members of the Salamanca School had never heard of a perfectly competitive market; in fact, nobody did for two hundred years after them’. This is correct as a claim about perfectly competitive markets in the technical neoclassical sense. But the deeper claim that the tradition was indifferent to competitive conditions is mistaken.

As de Roover (Reference de Roover1958: 425) has shown, ‘the whole discussion on the just price assumed the existence of competitive conditions’, even though the word concurrentium did not appear in Scholastic treatises until Luis de Molina used it in the late sixteenth century.Footnote ⁴ In other words, the Scholastics relied on the concept even if they lacked the modern terminology. Moreover, I will argue below, the kind of competition on which the Scholastic account depends is much closer to the dynamic, process-based view I propose in this paper than to the neoclassical model of perfect competition.

What, exactly, did the Scholastics condemn? First and foremost, the Scholastics opposed monopoly, defined far more broadly than the modern term suggests as ‘any pacts or rings formed to keep up or to depress prices above or below the competitive level’ (de Roover Reference de Roover1958: 426). Beyond that, medieval law had a whole vocabulary of market offenses that we have lost. To ‘forestall’, for instance, was to buy goods up before they ever reached the market. To ‘engross’ was to corner the supply. A ‘regrater’ bought goods only to sell them again, marked up, in the very same market (Baldwin Reference Baldwin1959: 29). The Scholastics also condemned the merchant who took advantage of buyers’ ignorance or desperation in order to charge different prices for the same good (de Roover Reference de Roover1958: 426). By the lights of later economics, not all of this holds up. Most of what the Scholastics dismissed as ‘speculation’ is now understood to be efficiency-enhancing, and price discrimination, sometimes even monopoly itself, can serve the competitive process rather than corrupt it.Footnote ⁵ But the general principle underlying the specific condemnations – that prices lose their justification when the conditions of open competition are absent – was sound.

What emerges from this picture is a clear, practically oriented understanding of the conditions required for prices to be just. The just price was the price set by common estimation in a market free from monopoly, artificial restriction of supply, fraud and discriminatory extraction. When competitive conditions fail, the doctrine of the just price mitigates the resulting problems by importing the competitive price as the governing norm (Friedman Reference Friedman1980).

Of course, the Scholastics’ understanding of competitive conditions was informal and pre-theoretical. They lacked a general account of why these conditions mattered, beyond the intuitions of natural law and the practical wisdom of the marketplace. But even if they did not recognize it, the conditions they identified are roughly those conditions under which the market process performs its morally important functions. Thus, what the Scholastics grasped through practical wisdom, the market process framework can explain and ground theoretically.

In saying this, I do not mean to claim that the Scholastics were market process theorists before their time. There are many reasonable competing reconstructions of the just price tradition, and reading that tradition exclusively through the lens of modern theory risks anachronism (Langholm Reference Langholm1998; Reyes Reference Reyes2025).Footnote ⁶ My point is simply that the anti-monopoly, open-entry aspect is a real and central strand of the tradition, and that it is best explained by the framework this paper develops. My account is a rational reconstruction of that strand. I am not offering it as a uniquely correct historical interpretation.

2.3. The market process view

The understanding of competition I want to defend was developed in the twentieth century by economists like Friedrich Hayek, Joseph Schumpeter, and especially Israel Kirzner (Hayek Reference Hayek1945, Reference Hayek1948; Kirzner Reference Kirzner1973, Reference Kirzner1997; Schumpeter Reference Schumpeter1942: ch. 7).Footnote ⁷ According to this ‘market process’ view, competition is best understood as a dynamic process of discovery, driven by entrepreneurial alertness to opportunities for profit.

The market process view takes as its point of departure the pervasive ignorance of market participants. As Kirzner notes, consumers typically have only a partial and incomplete grasp of ‘the kinds of commodities technologically possible with currently available resources and of the relative prices at which these commodities can in principle be produced’ (Kirzner Reference Kirzner1973: 18–19). Resource owners, for their part, are ignorant of the kinds of commodities consumers will buy. This market ignorance leads to widespread errors. Goods are priced too high or too low, resources are directed to inferior uses and mutual gains from trade are left unrealized.

The ‘pervasive ignorance’ of market participants actually takes two distinct forms. The first is Hayek’s knowledge of ‘the particular circumstances of time and place’ which exists only in dispersed form, scattered among millions of individuals and unavailable in its totality to any single mind (Hayek Reference Hayek1945: 521). The second is Kirzner’s ‘sheer ignorance’, in which agents are ignorant not only of the facts but of their ignorance itself (Kirzner Reference Kirzner1997: 62). These two conditions ground different functions of price. Dispersed knowledge grounds the informational function, by which prices aggregate and transmit what no individual knows in full. Sheer ignorance grounds the discovery function, where profit opportunities reward those alert to unrealized opportunities. I use the phrase ‘pervasive ignorance’ for the conjunction, and the account that follows draws on both.

Competition, on the view adopted here, is the process by which this pervasive ignorance is discovered and corrected. Entrepreneurs notice that a resource is underpriced here and overpriced there, or that an unmet need could be profitably served. They buy where prices are too low and sell where prices are too high. And in doing so, they push prices in the direction of better coordination. Entrepreneurial discovery leads to better alignment of buyers’ and sellers’ plans and less of the mutual ignorance that separated people from opportunities for gain.Footnote ⁸

In this process, prices play three important roles:

1. 1. Information: Prices communicate dispersed knowledge about relative scarcities, knowledge that exists only in distributed form among countless market participants and that, as Hayek (Reference Hayek1945) argued, no central authority could aggregate.

2. 2. Incentive: Prices create profit-and-loss signals that motivate entrepreneurial action, attract supply to where it is needed and encourage buyers to economize on what is scarce.

3. 3. Coordination: Prices reconcile the independently formed plans of buyers and sellers, bringing them into a pattern of mutual compatibility that no individual participant designed or foresaw.

The account of competition is strikingly different from the neoclassical one. On the market process view, there is no uniquely determinate ‘competitive price’ to which actual prices should conform. Prices are constantly changing both because circumstances change and because market participants are constantly discovering new information about those circumstances. What this process generates is a process that tends, through entrepreneurial discovery, toward better coordination. The question for a theory of the just price is therefore not ‘is this the price that would exist in a competitive market’, but rather ‘is this price serving its functional roles within the competitive process?’

This dynamism is the market’s most distinctive feature. The very stock of goods the process allocates is, in large part, discovered and created in the course of the process itself (Schumpeter Reference Schumpeter1942; Young Reference Young2022). A price within it is a standing offer in a contest still running, held in place only by the possibility that someone alert will find it mistaken and profit by correcting it. Openness to entry just is openness to discovery, a point Kirzner (Reference Kirzner1988: 181) extends to all market incomes, wages included.

3.1. The challenge

I have described the market process and the functional roles that prices play within it. But one might grant all of this and still deny that serving these roles makes a price just. The functions I have described sound like contributions to economic efficiency, not to justice. A version of this objection is developed by Heath (Reference Heath2018: 26). When two parties trade on the market, the buyer and seller each capture some share of the gains from trade. But the division between them, he claims, is ‘essentially arbitrary from the standpoint of distribution, since it will be determined entirely by the slope of the supply and the demand curves’. The most that can be said for the market-clearing price, on this view, is that it is efficient. But efficiency is not justice.

I take Heath’s argument to be the deepest challenge to the view defended in this paper. If it cannot be met, my theory reduces to the claim that efficient prices are useful. Whether efficiency is all there is, however, depends on one’s account of justice. If justice is a distribution to be achieved by desert or by need, then coordination will indeed probably fall short of it. A price could coordinate plans perfectly and still leave a theorist’s favored distribution unmet.

But there is a different tradition of thinking about justice, running through David Hume, Friedrich Hayek and David Schmidtz, according to which justice is in the first instance about the rules by which people with competing claims coordinate their use of scarce resources. The subject of justice, on this view, is the very kind of situation that market prices are meant to address. Justice is a function of rules governing situations involving a multitude of people with competing claims to scarce goods, and no shared ranking of whose purposes come first. In a large society, such competing claims cannot possibly be resolved on a case-by-case basis. Justice addresses them through rules that are general, stable and public, rules under which each person can form reliable expectations and lay plans of her own (Fuller Reference Fuller1964; Hayek Reference Hayek1976). On this conception of justice, which I adopt, the main question is then whether the competitive process is a coordinating rule of the right kind. I do not claim this is the only defensible conception of justice. But it is a serious one, well represented in the tradition, and taking it seriously allows us to resolve the trilemma with which this paper began.

My argument has two parts, cumulative rather than independent. Each of them contributes something that the other lacks. Before I get to them, however, I want to give an analogy to help explain the nature of the position that I am defending here. It might be helpful to compare the market process view of prices to reliabilist theories of justification in epistemology. For the reliabilist, a belief is justified when it is produced by a cognitive process that reliably tracks truth, even if the process is fallible and even if the believer knows nothing about it (Goldman Reference Goldman and Pappas1979). The market process account makes the parallel claim about prices. On the view defended here, a price is just when it is produced by a process that reliably performs morally significant coordinative functions. The coordinative functions play the role that truth plays for the reliabilist. They are the independent standard against which the process is assessed, and the reason its outputs carry authority.

There are two points about this parallel that deserve emphasis. First, my account of just prices is externalist. The justice of a price no more depends on the seller’s intentions than the justification of a belief depends on the believer’s grasp of her own cognitive processes. Second, it remains to be shown is that the functions in question are morally significant, or, in other words, that coordination stands to justice as truth stands to justification. That is the work of the two arguments to which I now turn.

3.2. The adversarial system analogy

Heath sets forward an analogy between the competitive market and the adversarial legal system.Footnote ⁹ In an adversarial trial, the defense lawyer has an obligation to mount the strongest possible case for the defendant. This is true even when the defendant is probably guilty, and even when the lawyer’s personal sympathies lie elsewhere. Evaluated in isolation, this obligation looks morally suspect. Its justification comes from the institutional context in which it operates. The adversarial system produces just outcomes precisely through the vigorous, one-sided advocacy of lawyers who are not themselves aiming at those outcomes. As Heath (Reference Heath2014: 28) puts it, channeling Adam Smith ([1776] Reference Smith, Campbell and Skinner1981: IV.ii.9), the effective trial lawyer ‘promotes an end which is no part of his intention’.

There are three important things to note about this analogy. The first and third can be stated briefly. The second will require a somewhat more extended treatment. First, the norms governing participants in an adversarial system are genuine moral norms, not merely technical or prudential ones. The defense lawyer is morally obligated to mount a vigorous defense, whether she wishes to do so or not. This obligation comes from the moral importance of the system’s function, combined with the fact that this function is best served by differentiated roles with specific requirements. Thus far, Heath and I are in agreement. He, too, thinks that the norms governing competitive market behavior are moral norms, genuinely binding because of the moral importance of the functions they serve (Reference Heath2014: ch. 1).

Second, and more importantly for our purposes, the outcomes of a well-functioning adversarial process carry a presumption of justice. When a trial is conducted according to the rules, its outcome carries a defeasible claim to correctness. That claim can be overridden only by a showing of procedural failure or extraordinary injustice. For Rawls (Reference Rawls1971: 85–86), the criminal trial is thus an example of imperfect procedural justice. Whether the defendant committed the crime is an independent fact, but no procedure can track that fact perfectly. The verdict’s authority therefore rests instead on the reliability of the process.

The market process account makes a stronger claim: there is no independent fact, outside the market process, of what the just price is. Nothing stands to the just price as actual guilt stands to the just verdict. Guilt or innocence is a factual matter, so even a perfectly conducted trial can get the answer wrong. A price has nothing comparable to get wrong. This does not mean there is no independent standard governing the market process account. But it is standard is rooted in the coordinative functions of markets, not the particular prices those functions generate. Where the process is functioning under conditions of openness, the price it yields simply is the just price.

It is tempting to call this a case of pure procedural justice, one in which the fairness of the outcome is determined entirely by the fairness of the procedure that generates it (Rawls Reference Rawls1971: 86). But this would be a mistake. The market process confers a presumption of justice only because of the coordinative function it exists to serve. Prices that fail to serve that function because, for instance, entry into the market is foreclosed generate no presumption of justice. Coordination thus plays the same role in the process of price generation that truth plays for the process of belief formation on the reliabilist account described earlier. The market process account has an independent standard. What it lacks is an independent target. There is no independently correct figure that it approximates and might miss.

One final clarification before moving on to the third point. The claim I am making here is about the content of a procedure’s outcome. That outcome’s force, however, is another matter. A procedurally impeccable outcome can be outweighed by independent moral considerations. But this does not make those considerations into rival standards of correctness. Consider, for instance, a fair coin flip for the last seat on a lifeboat. There is no antecedent fact about who was entitled to the seat, so this is a clean case of pure procedural justice.Footnote ¹⁰ And yet it might still be wrong to enforce the result of the coin flip if doing so means that a child would drown. In this case, we would not say that the outcome of the flip was mistaken. Rather, it was overridden by a competing moral claim.

The presumption I defend here has this same structure. The competitive process settles what the just price is, but it leaves open whether charging it is what one ought to do all things considered. Whether a price is just is therefore one question, and whether one is permitted to charge it is another. This is why I speak of a pro tanto procedural presumption rather than of pure procedural justice in Rawls’s strict sense. Showing that the market process was not genuinely competitive would rebut the presumption, which would mean the price that seemed just was never really just to begin with. On the other hand, a countervailing duty such as the duty of charity could make charging a price wrong, without thereby entailing that the price is unjust.

The third and final point to note about the analogy is that in the adversarial system, the presumption of justice attaches to outcomes because of the system’s functional success. It’s not that the adversarial system merely happens to produce just outcomes while also happening to be efficient. Its functional design is precisely what makes its outcomes just. The adversarial structure, with its differentiated roles and procedural constraints, is the best available method for achieving the system’s morally important ends. It follows that when participants act within their roles and the system functions properly, the outcomes inherit the moral authority of the procedure that produced them.

The obvious objection to all this is that I have chosen a flattering analogy. Lots of rule-governed practices produce outcomes through procedurally impeccable competition without any generating any presumption that those outcomes are just. Football, for instance, has genuinely moral norms against intentional injury and cheating. When the better team wins under those norms, the result is legitimate, maybe even fair. But no one would call the final score just. Why think that the market resembles the courtroom rather than the playing field?

The answer lies in the nature of the conflict that each procedure settles. A game’s outcome is internal to the practice. Who won the championship has no moral significance outside the game except whatever significance people choose to attach to it. For practices like this, procedural legitimacy exhausts the normative question.

Markets are different in two ways. They allocate scarce resources among people with competing claims on them, and that allocation raises questions of justice whatever institution performs it. And whereas a game leaves the world as it found it, markets create the wealth on which people’s life prospects depend (Young Reference Young2022). The greater a practice’s external stakes, the less plausible it is that procedural legitimacy exhausts its normative assessment. Like legal systems, markets settle questions that are antecedently questions of justice.Footnote ¹¹

Heath would resist this conclusion. His view is that the norms governing market behavior are moral norms, but that the prices those norms produce are to be defended as efficient rather than just, since their distributive upshot is arbitrary (Heath Reference Heath2018: 26–27). I think this position is unstable. If the norms governing the competitive process are genuinely moral, then the prices that emerge from compliance with these norms carry moral significance.

Heath might reply that this proves too much. Simply because a process is governed by moral norms, it does not follow that whatever outputs that process generates can be called ‘just’. A factory governed by morally appropriate safety norms doesn’t produce just products. But these cases are disanalogous. Safety norms impose constraints upon the production process. The norms governing the competitive market, in contrast, constitute the process whose outputs we are evaluating. When those norms are violated, the process is simply something different. It is not a competitive process at all, and its outputs therefore do not carry the authority of such a process.

The analogy between markets and trials received further support from consideration of the way they respond to the problem of background injustice. It is an obvious fact that many criminal trials take place in a context of significant injustice. Some defendants arrive in court with inadequate counsel, and inequalities of wealth and education have a significant influence on legal outcomes. Not all inequalities are unjust, but even if many of them are, we do not conclude from this that verdicts lack presumptive standing. Background inequities call for remedies of their own, such as indigent defense or ineffective-assistance doctrine. This same logic applies to markets. Background injustice in holdings and opportunities is real. It calls for remedies of its own, which I take up in Section 5. But just as in the legal case, it is not a reason to deny that the competitive process confers a presumption of justice on its outputs.

3.3. Coordination as a standard of justice

The previous section drew on the adversarial system analogy to argue that the competitive process confers a presumption of justice on the prices it produces. This section will set out the normative standard by which the market process should be evaluated.

Kirzner (Reference Kirzner1973: ch. 6) argues that the correct standard is coordination, which he defines as the mutual compatibility of the independently formed plans of market participants. Markets coordinate when buyers and sellers can carry out their intended transactions, and when the decisions people make based on price signals are consistent with the decisions others are making. This is different from the more traditional evaluative standard of Pareto optimality, which Kirzner regards as incoherent when applied to a world of dispersed knowledge. Because information is fundamentally dispersed, the question of optimal allocation is simply unanswerable. No mind holds the data against which an allocation could be checked (Hayek Reference Hayek1945). Worse, the ‘optimal allocation’ is not even a fact that exists in advance of market activity. It is constituted, not merely revealed, by the very process one is trying to evaluate (Buchanan and Vanberg Reference Buchanan and Vanberg1991).

Coordination might seem like a purely economic concept. That is, it might seem like a matter of mere efficiency, not justice. But it is worth thinking about what a lack of coordination means in practice. Discoordination means that people’s plans will be frustrated. It means that they will make decisions that they would not have made if they had known more. And it means that resources will be wasted, and that human needs that could have been met will go unmet. In a world of scarcity, none of these are morally neutral outcomes.

It is possible to say more than this, and to argue that coordination is actually a central part of what justice is for. David Schmidtz (Reference Schmidtz2023), for instance, has argued that the fundamental task of justice is traffic management. Justice tells us whose ‘turn’ it is so that people with different destinations can share the road without needing to agree on whose destination is more important.Footnote ¹² Property rights and market conventions help people to know what to expect from each other, and they do so without requiring anyone to rank the relative importance of different people’s projects (Schmidtz Reference Schmidtz2023: 19). It is precisely this understanding that the market process account applies to prices. Prices manage the traffic of a complex economy. They settle whose turn it is to use scarce resources, and they do so through a process that does not require any central authority to know (let alone rank) the values of the plans they coordinate.

Of course, not all coordination is a matter of justice. Coordinating which side of the road to drive on matters morally, but any convention will do. The question settled by the rule isn’t one that anybody has any pre-existing competing claim to. But prices coordinate the use of resources to which people do have competing moral claims. Settling who gets to use what, when others want or need or deserve it too, is exactly the kind of problem theories of justice exist to address.

One might still object that this proves too much. If justice simply means coordinating competing claims, then central allocation by a dictator would seem to count as justice too. But this objection relies on an ambiguity in the word ‘coordination’. Recall that for Kirzner, coordination refers to the mutual compatibility of independently formed plans. This is something that central planning simply does not do at all. A well-functioning command economy has exactly one plan. The dictator’s plan overrides individuals’ plans; it does not coordinate them. If justice is coordination, then the dictator’s plan simply does not qualify.

3.4. The market process account

The arguments of the previous two sections jointly support this paper’s thesis, that the competitive market process, when it is operating, confers a genuine though defeasible presumption of justice on the prices it produces.Footnote ¹³

We can state this claim more carefully in the following way. A price is just when it is the product of, and serves the functions of, an open competitive process. In practice, this requires three things. First, the market must be open to entry. ‘Openness’ here refers to the barriers between sellers and the opportunities signaled by prices. Openness can be violated when entry into markets is closed by law, as it is in the cases of exclusive licenses and monopoly grants. But entry can also be closed by prohibitively high economic costs. And it can be closed by restrictive circumstances, as when the parties are so isolated or the need is so urgent that no rival could arrive in time. Markets are open when supply responses are not closed off in any of these ways. Importantly, openness is a matter of possible entry, not actual entry. What matters is that outsiders are capable of taking advantage of the opportunity if they choose to, thereby exposing the price to potential correction.Footnote ¹⁴

Beyond openness, the second requirement is that the price must be of a kind that performs the market’s coordinative roles. Those roles consist of reflecting information about scarcity, creating incentives for supply responses and demand adjustment and reconciling the plans of buyers and sellers.Footnote ¹⁵ And the third requirement is that the process must not be corrupted by fraud, coercion, collusion, or the artificial restriction of supply.

These three conditions are not ad hoc moral requirements appended to the functional story. They are, instead, the explanation of what it means for a price to be part of a functioning competitive process. And each of them can be satisfied to a greater or lesser degree. The presumption of justice conferred on prices is stronger or weaker accordingly. In the extreme, a sufficiently severe failure along any one or more dimensions would defeat the presumption entirely.

One difficulty for the market process account involves specifying what it means for a price to be ‘part’ of a competitive process. How can we tell whether a particular price is part of such a process, or something falling outside of it? The answer cannot be a merely causal one. A price could be doing everything it ought to be doing in terms of signaling new supply without any new supply actually showing up. But the answer cannot be merely behavioral, either. A seller can comply perfectly with the rules of bilateral exchange while standing entirely outside the competitive process, as illustrated in the case of Drowning.

What the market process account needs, then, is a criterion with two components, one structural and one functional. The structural component is that the conditions for open competition must be in place. The functional component is that the price must be of a kind that characteristically serves the coordinative functions under those conditions. That is, it must be a price that responds to market-wide scarcity, and one that would be charged to anyone similarly situated. A price that is based on an individual buyer’s desperation, one that would evaporate the moment a rival appeared, would fail to satisfy the functional component.

The logic here is familiar from functional analysis in other areas of philosophy. According to Cummins’ (Reference Cummins1975) causal-role account of function, a component’s function is the contribution it makes to a containing system’s capacities when the system is operating. Functions, on this account, are determined by their role rather than by their history, or even by their current activity. A heart has the function of pumping blood even if no one designed it to do so, and even if it has temporarily been placed on surgical bypass. Prices work the same way. A particular price that fails to attract new suppliers still has its coordinative function, as long as the structural conditions obtain and the price is of the coordinative kind. Similarly, a reliably formed belief remains justified even when it happens to be false. In contrast, a price extracted in a bilateral encounter, with no market around it, has no coordinative role at all. This is why the price charged in Drowning carries no presumption of justice, even if the rescuer perfectly follows the rules of bilateral exchange. Prices derive their function from the competitive process of which they are a part. And in Drowning, there simply isn’t any competitive process in play.

Here again, the parallel with reliabilism is instructive. Reliabilism faces a well-known difficulty known as the generality problem. The problem is that every token belief is produced by a process that instantiates many types, not just one (Conee and Feldman Reference Conee and Feldman1998). But since reliability is a property of types, how do we know which type to look at to determine whether a belief is justified? The standard answer is to fix the relevant type by reference to the features of the process that were causally operative in producing the belief.

We can use this same approach to address a similar problem for prices. Any given price in the market is a token instantiating many different types. The sellers’ act of charging $12 for a bag of ice in Ice Gouging could be described, alternatively, as charging ‘the market price for emergency supplies’, or ‘a price charged during a shortage’, or as ‘a price extracted from a desperate buyer’. How do we know which one to choose? The answer is that the relevant type is fixed by the causally operative features of the setting. Is entry to the market open, such that the price could potentially attract new supply? Is the price responding to market-wide scarcity, or to a single person’s desperation? In Drowning, the price is extractive because it is produced not by market conditions but by the buyer’s necessity and the absence of alternatives. Ice Gouging is more complicated, because the price in that situation is plausibly responding to two features that point in different directions. Buyers are under real hardship, but there also is a genuine market-wide shortage. Nevertheless, the same tests allow us to arrive at a clear verdict. The $12 price would be charged to anyone similarly situated and would not evaporate the moment a rival seller arrived. We can conclude, therefore, that the price is produced by the condition of the market, not a particular buyer’s desperation.

To recap, then, a price is the product of and serves the functions of an open competitive process when three conditions hold: entry is open, the price is of the coordinative kind and the process is free of fraud, coercion, collusion and artificial restriction. When these conditions are met, prices inherit the moral authority of the process that generated them. They are just in the same way and for the same reason that a verdict reached through a fair trial is just. In other words, they are just because the process that produced them is the kind of process whose outputs carry moral weight. When the conditions are not met, this presumption does not hold. A high price charged in such circumstances might still be defensible on other grounds, but it cannot claim the authority that the competitive process confers.Footnote ¹⁶

This does not mean the account has nothing to offer in the way of guidance when the conditions fail. The just price tradition, as I noted in Section 2, was developed largely to address precisely failures of that sort. What the market process account adds to this tradition is the identification of which conditions have failed and what follows from each kind of failure. When the market is not open, sellers cannot rely on the competitive process to justify the price. In such circumstances they face a heightened obligation to price in a way that approximates what the competitive process would deliver, a price reflecting genuine costs and scarcities rather than the exploitation of market power (Guzman and Munger Reference Guzman and Munger2020).Footnote ¹⁷

The graded operation of the three conditions finds support in the common law doctrine of unconscionability.Footnote ¹⁸ That doctrine does not employ a bright-line test for distinguishing between permissible and impermissible transactions. Instead, it employs a kind of balancing act. The greater the procedural unfairness (lack of meaningful choice, information asymmetry), the less substantive unfairness is required to invalidate a contract, and vice versa. The market process account shares this structure. The more severely the competitive conditions are compromised, the less a price can claim the presumption of justice, and the heavier the seller’s obligation to justify the price on other grounds.

As I have noted, the justice of a competitive price is only a pro tanto reason to charge it. Even when a price is generated under the right conditions, that reason can be outweighed by sufficiently weighty countervailing moral considerations. The most important of these is a duty of charity. What exactly the duty of charity demands, and when, are questions for the theory of beneficence, not the theory of just price. My argument assumes only that such a duty exists in some form and can, in the gravest circumstances, override the case for charging a just price. For this, it is enough to specify two features that any plausible account of the duty will share. The first is that it must be bounded. A duty of charity demanding enough to displace market pricing generally would erode the very coordination on which a society’s capacity to meet needs depends, and would thus be self-defeating (see Schmidtz Reference Schmidtz2023, chs. 5–6). The second is that the duty leaves the justice of the price untouched.Footnote ¹⁹ The price is just by virtue of its coordinative functions. The duty of charity bears only on whether, all things considered, the seller ought to charge it.

This two-part structure might seem like a retreat from the trilemma with which this paper began. It is not. A theory of just verdicts is not incomplete because pardons exist, and a theory of just prices is not incomplete because charity can override it. The market process account answers the trilemma’s question with a single criterion: the functional structure of the competitive process. Charity supplies no rival standard of just prices. What charity requires varies with the buyer’s peril and means, so it fixes no prices of its own. The just price remains the baseline from which charity may require a seller to depart. And as we will see below in Section 4, appeals to charity do little of the work of the market process account. Every condemnation there, from Drowning to Mining Town, stems from the procedural standard itself. The duty of charity never determines which prices are just, only what may be asked of the seller.

This distinction matters more than it might seem. A rule remade for every hardship would fail to serve the coordinative function of justice. Hard cases make bad law. Moreover, there are different remedies for the two sorts of failure the distinction identifies. An unjust price calls for correcting the price by suppression or rectification. A just price that someone cannot afford, in contrast, calls for individualized aid, not for interfering with a price that is doing the very work a just price should. To call it unjust would mistake both the wrong and its cure.

4.1. Drowning

P charges Q $10,000 for a nearly costless rescue. On the market process account, this price is unjust. The reason is to be found in the membership criterion set out in Section 3.

Both components of that criterion fail in this case. The structural conditions fail because the price cannot encourage new entry. It is simply not possible for any rival to arrive in time. And functionally, the price is responding to the wrong kind of fact. The $10,000 price is a reflection of Q’s desperation, not of any market-wide scarcity. It is a price to the drowning man, not a price to the market in general. Significantly, the market process account arrives at this conclusion without needing to appeal to any moralized test of exploitation. The problem is a structural one, rooted in the absence of a competitive market process.

The membership criterion also tells us what sorts of circumstances would lead to a different verdict. Imagine a coastal economy with a standing market in rescue services, complete with posted rates and competing operators. A high price in that market would be a price to the market, exposed to entry and performing the signaling work that summons rescue capacity to dangerous waters. In this scenario, the posted rate is what the market asks of any swimmer in those waters. And that rate is held in check by the prospect of rival operators, at least as long as such rivals have the possibility of entering the market effectively. Whether a market is genuinely open depends on the time-scale of the need. A rescue market whose nearest competitor is a day away is, to a swimmer with minutes to live, no more open than no market at all. In such a case, the presumption fails.

4.2. Ice gouging

Four men buy ice for $1.70 and sell it for $12 in a hurricane-affected area. Here, the market process account yields a different verdict.

In this situation, the competitive process is operating. Indeed, the sellers themselves are part of that process. They noticed the opportunity, loaded a truck and brought fresh supplies where they were needed. And as long as roads are passable and rivals can arrive within the relevant horizon, nothing prevents other entrepreneurs from doing the same. In this case, the $12 price is performing each of its coordinative roles. It communicates that ice is scarce and valuable, it rations the existing supply, and it creates the profit incentive that drew these sellers to the area and will draw others. There is no deception or coercion. And the buyers, while facing hardship, are not in the position of Drowning’s Q. They face a temporary shortage in a market that is already generating a competitive response.

The market process account thus holds this price to be just. It might still be wrong to charge it to particular buyers who are facing the gravest circumstances and who cannot afford the market price. But the price is performing its function, and therefore carries the presumption of justice.

The market process account thus supports the conclusion reached by many economists (and some philosophers) about price gouging, while also explaining why cases like this are different from Drowning. The difference is not merely that the buyers in Ice Gouging in less desperate. Once again, the explanation is structural. In Ice Gouging, there is a competitive process at work. In Drowning, there is not.Footnote ²⁰

The account also explains why many people still feel so uneasy about cases of price gouging. It is easy to see a high price that profits one party at the other’s expense. What is harder to see is the competitive process that is already getting the next truck full of ice on the road. Our moral intuitions respond to the apparent zero-sum conflict between buyer and seller and miss the positive-sum process unfolding behind it.

4.3. Locke’s Venditio

The market process account’s most illuminating application is to the puzzles in Locke’s remarkable essay Venditio. The essay begins with a clear, simple statement of Locke’s theory of the just price: the just price is simply ‘the market price at the place where he sells’ (Locke Reference Locke and Goldie1997: 340). The remainder of the essay, however, demonstrates that this theory is not nearly as simply as it appears. And perhaps not even internally coherent. On the one hand, Locke holds that a Danzig merchant may sell grain at 20s per bushel to famine-struck Dunkirk while his partner sells at 5s in nearby Ostend. This is not unjust, because ‘he sells at the market rate at the place where he is, but sells there no dearer to Thomas than he would to Richard’ (Locke Reference Locke and Goldie1997: 342). On the other hand, Locke insists that it would be unjust to charge a desperate individual in Ostend a higher price than he charges his non-desperate neighbors. ‘He that makes use of another’s ignorance, fancy, or necessity to sell … dearer to him than to another man at the same time, cheats him’ (Locke Reference Locke and Goldie1997: 341).

These two positions seem to conflict. If it is wrong to exploit one person’s desperation, how can it be right to exploit an entire city’s? As Mack (Reference Mack2024) has shown, Locke’s conclusions across the essay’s various examples often appear ‘implausible in themselves or inconsistent with one another’. Locke applies a ‘necessity condition’ to condemn the sale of a horse at an elevated price to a desperate buyer, but then repudiates that same condition when defending the corn merchant’s high prices at Dunkirk. I do not claim that Locke’s essay is fully consistent, and I treat the market process account as a rational reconstruction of Locke’s best insights rather than an interpretation of his actual views.Footnote ²¹ But the account can explain why the Dunkirk and Ostend positions are more consistent than they appear, and in doing so, it offers a principled reconstruction of the puzzle that has dogged the Venditio literature.

Why is the Dunkirk price just? In short, because the high price in Dunkirk is the competitive process at work. The high price at Dunkirk reflects genuine scarcity. It signals the desperate need for grain, creates an incentive for merchants to brave the risks of sailing to a famine-struck port and rations the available supply.Footnote ²² Moreover, the price emerges under conditions of openness. As long as ships can reach the port, other merchants can and presumably will respond to the same price signal. Finally, the merchant sells ‘no dearer to Thomas than he would to Richard’. In other words, the price is not targeted at any individual’s particular necessity but reflects the general conditions of the market.

Why, in contrast, is the individual exploitation in Ostend unjust? The answer, again, is rooted in the competitive process. In this case, that process is simply not operating at the level of the individual transaction. A seller who charges one buyer a higher price than another, simply because of that buyer’s ignorance or necessity, is not responding to a market signal. The price isn’t signaling scarcity, and it’s not incentivizing any new supply. The high price serves no coordinative function. It is pure extraction. The transaction is more like Drowning than like a competitive market.

The market process account thus explains the difference between Locke’s two cases. If the wrong of individual price discrimination were simply ‘differential treatment’, then selling at a high price to all of Dunkirk should be more objectionable, since more people are affected. My account explains why it is not. What changes between the cases is the functional character of the price.Footnote ²³

Finally, it is worth noting that Locke’s account has the same two-part structure as the account defended here. After defending the corn merchant’s 20s per bushel as no injustice ‘against the common rule of traffic’, he adds that the merchant who ‘carry[s] it away unless they will give him more than they are able … offends against the common rule of charity’ (Locke Reference Locke and Goldie1997: 342). This common rule of charity does not show that the Dunkirk price was unjust. Rather, it takes priority over the seller’s entitlement to charge it.Footnote ²⁴

4.4. Monopoly

Almost every theory in the just price tradition treats monopoly as essentially non-competitive and monopoly prices as presumptively unjust. On the market process account the verdict is more nuanced. My account identifies market power as a continuum rather than a binary and holds that the presumption of justice weakens as market power increases and the competitive process becomes less effective.

Consider two types of monopoly. The first type is a resource monopoly. Nozick’s (Reference Nozick1974: 180) example of the sole watering hole in a desert is a vivid example. In this case, the high price is a reflection of a genuine scarcity. Water in a desert would command a high price even in an open market. And a price that rations a scarce good is performing one of the price system’s legitimate functions. The problem with monopoly in this case is that it allows the owner to restrict output and raise prices based on nothing more than buyers’ desperation. This means that there is a portion of the price that reflects only the absence of alternatives, and that would evaporate if a rival source appeared. To that extent, the price carries with it no presumption of justice.

Things are different in the case of what we might call innovation monopoly. A firm that has developed a genuinely new product enjoys a temporary period of market dominance, perhaps protected by a patent. In this case, however, the higher price that firm is able to charge is performing an important function. It rewards the discovery of something that did not exist before and provides the incentive on which future such discoveries depend (Kirzner Reference Kirzner1973: ch. 3). It is true that the patent closes the market to entry by new providers. But it matters that the market it closes is one that did not exist until the innovation created it. It also matters that the closure is merely temporary, within a system whose point is to keep discovery open over time. To the extent that the patent system preserves openness over time, ensuring that the monopoly is temporary and that competitors can eventually enter, the high price can be just.Footnote ²⁵

4.5. Wages

A wage is simply the price of labor, and the market process account applies to them in the same way it applies to any other price. That application, however, puts my account into direct conflict with Joseph Heath’s central contention that ‘markets are structurally unable to deliver “just” wages, according to any everyday-moral understanding of what justice requires in cooperative interactions’ (Reference Heath2018: 4). According to Heath, wages perform the socially useful function of channeling labor to its best employment. But Heath insists that one can only defend that function by abandoning the idea that any particular wage is just. I think that Heath is correct about the function of prices, but wrong in his conclusion that wages cannot be just. As I argued in Section 3, the wage system’s allocative function is not a reason to withhold evaluations of justice but the ground of them. The rest of this section applies that view to some contested cases.Footnote ²⁶

Consider sweatshop wages, the case that has generated the most sustained debate about wage justice (Ferguson and Ostmann Reference Ferguson and Ostmann2018; Flanigan Reference Flanigan2018; Powell and Zwolinski Reference Powell and Zwolinski2012; Zwolinski Reference Zwolinski2007; and, on the connection between competition and wage fairness specifically, Richard Reference Richard2023). In the paradigmatic case, a multinational corporation outsources production to a factory in a developing country. The working conditions are harsh and the wages are low, at least by the standards of the corporation’s home country. Those wages often exceed, however, what workers could earn in locally available alternatives. Are sweatshop wages just?

The answer depends in large part on the extent to which the labor market is genuinely open, an empirical question that varies by place and by sector. When it is, and when multiple firms are able to compete for workers, and workers are able to choose among employers, the wage reflects genuine scarcity conditions and performs each of its coordinative functions. It signals the relative abundance of labor and the scarcity of capital, thereby creating incentives for further investment in the region. It attracts employers who might otherwise locate elsewhere. And it coordinates the plans of workers and employers without requiring any central authority to determine what labor is ‘really worth’.

On the market process account, such a wage is just. And it is worth noting that the account reaches this conclusion without having to rely on controversial moral or economic theories. The argument here does not turn on the ‘nonworseness claim’ that it cannot be worse for employers to provide some benefit to employees by paying them a low wage than it is for them to provide no benefit by not hiring them at all. Nor does it depend on the neoclassical test that wages will reliably equal the worker’s marginal revenue product. Rather, the market process account holds the wage to be just because the competitive process is operating and the price is performing its morally significant functions. In this respect, my account concurs with Richard’s (Reference Richard2023) claim that defenders of sweatshop wages do not actually need the nonworseness claim at all. The competitive character of the labor market already does the normative work.

Things are different in Richard Mining Town example, in which a mining company is the sole employer in an isolated village. The example involves a company that pays subsistence wages to its employees while earning large profits. Because of its geographic isolation, it is relatively immune to competition. Here, the market process account reaches a different verdict. In Mining Town, the labor market is simply not functioning. There is no competitive process to bid wages toward their coordinative level, and there is no discovery mechanism through which entrepreneurs might notice the profit opportunity and respond. The wage merely reflects the bargaining power of a single buyer facing sellers with no alternative. This is pure extraction, and the wage therefore carries no presumption of justice.Footnote ²⁷

Mining Town is, of course, a hypothetical example. But it points to a fact of real-world significance, namely, that labor markets fail the openness conditions more easily than markets for other sorts of commodity do (Manning Reference Manning2003). Labor is both less mobile and less divisible than physical goods. There are often severe information asymmetries between employers and employees. And workers are dependent on employers for survival in a way that makes the extraction-from-necessity condition more likely to be triggered. None of this undermines the market process account, or its application to wages. A wage is still a price, and there is still a strong case for regarding wages as just in competitive markets as presumptively just. But it does mean that the account’s conditions are systematically more demanding for wages than for most other prices, and that whatever presumptions it generates deserve correspondingly closer scrutiny.

A more theoretically challenging problem for the market process account involves cases where a wage set in a competitive labor market nevertheless leaves workers unable to meet their basic needs. The market process account addresses the challenge through its pro tanto structure. The competitive wage is just, but its justice is not the last word on what to do. The shortfall calls out for redress. Whether that redress is best met by the employer, by public provision, or otherwise, however, is an all-things-considered question that my account leaves open.Footnote ²⁸

5.1. Why individuals, not just institutions?

It is possible to grant that market prices play have an important social function, but to deny that this has anything to do with the obligations of individual transactors. Aren’t those functions a matter of institutional design such as antitrust regulation or consumer protection law?

The short answer is ‘yes, and the market process account does not say otherwise. The account I have defended does not ask individuals to promote market values, as though each seller bore responsibility for the health of the competitive process. That, indeed, is the job of institutions. The market process account simply identifies when a price can claim the justification of the competitive process, and it says what is required when it cannot. The additional duties that are imposed on buyers will be ones that pertain to her own transaction, such as pricing in a way that approximates what the process would deliver or providing additional transparency where information has been corrupted.

Institutional design and transactional duty are both appropriate remedies, but they are remedies to different kinds of problem. The former maintains the conditions of competition. The other forbids taking advantage of their absence.

5.2. Does the account smuggle in a theory of rights?

Richard (Reference Richard2023) objects that accounts of fair pricing that draw on Austrian economics reconcile fairness and efficiency only by smuggling in a libertarian theory of rights. He develops the charge against an Austrian reconstruction of Wertheimer’s disequilibrium-price account. The distinction between the mere absence of competition and a defect in competition, he argues, can be drawn only against a background structure of independently defined property rights. This means that the non-exploitative counterfactual rests on the libertarian premise that whatever follows from a voluntary transfer of claim rights is just (Reference Richard2023: 344–45).

Richard’s objection is aimed at the literature defending sweatshop wages, including work of my own, which has claimed the advantage of neutrality between competing theories of fairness (Powell and Zwolinski Reference Powell and Zwolinski2012; Zwolinski Reference Zwolinski2007). The market process account, however, makes no such claim. It claims that the competitive process confers justice on prices because of the functional roles those prices play. That is a substantive normative commitment, but it is not a libertarian one. It does not depend on any prior theory of property rights, or self-ownership, or the absence of positive obligations. The account presupposes that some system of exchangeable property is in operation, since without one there is no rivalrous bidding from which prices can emerge. But presupposing that some system of property operates is not the same as presupposing a distinctively libertarian theory of property, or any particular theory of property for that matter.

5.3. What about externalities?

Even prices in an open market can go wrong by failing to take account of costs that fall on third parties. Pollution is the obvious case. What does the market process account have to say about prices such as these? The answer is that externalities involve a failure of prices’ informational function. When prices fail to register costs to third parties, they thereby misstate the relevant scarcities and signal that resources are cheaper than they actually are. This leads to mis-coordination and conflict. Such prices fail the second condition of the account. Standard legal remedies like liability rules and Pigouvian taxes can be thought of as attempts to restore to prices their proper function. Whether they are ultimately efficacious in this aim is, of course, an open empirical question.

5.4. What about background justice?

The market process account says that prices are just if they are the product of, and serve the function of, a competitive market. But competitive markets can exist and operate in deeply unjust contexts – unfair distributions of wealth, inadequate provision of social services, and so on. Can the prices that emerge in such contexts truly be said to be just?

Nothing in the account I have offered should be read to deny either the existence of background injustice or its moral significance. The question, however, is whether this renders the idea of transactional justice meaningless. I deny that it does (see also Zwolinski Reference Zwolinski2012). Even in non-ideal circumstances, there is still a difference between prices that serve a coordinative function and prices that are pure extraction. The market process account tells us what that difference is, and thereby provides a standard for distinguishing between more and less defensible prices without requiring us to first resolve every question of distributive justice.

Accounts that focus on background justice, such as Melkevik’s (Reference Melkevik2019) insistence on fair institutional preconditions or Steiner’s (Reference Steiner1984) theory of exploitation as rooted in antecedent injustice in holdings, address important and complementary questions. The market process account is a supplement to such theories, not a replacement.