Justice vis à vis welfare: how Austrian welfare economics should fit in the Austro-libertarian framework

The present work calls for a modification of Austrian welfare economics so that it should be given a distinct role within the Austro-libertarian paradigm. To this end, we clearly distinguish between justice and welfare. We proceed by trying to stick to the spirit of Austro-libertarianism as firmly as possible. In particular, we operate within the universe of voluntary (viz. rights-respecting) exchanges, having assumed the libertarian absolutist side-constraints view of rights. Furthermore, we take for granted the Austrian actual preference satisfaction view. The specific purpose of the present paper is two-fold. First, we subject Austrian rights-based welfare economics to critical scrutiny. Concluding that the above theory begs the question in favor of free market, we assert that to gauge market efficiency we need to resort to some descriptive criterion, which could reasonably be deemed constitutive of welfare. And it is only then that the question of market efficiency is an open question.


Introduction
What initially gave impetus to our investigations was the very phenomenon of Austro-libertarianism 1 as well as a rather intriguing fact that a person's being an Austrian economist makes it very likely that the person is also a political libertarian (see Boettke, 1995). However, this rather strong correlation is particularly puzzling in the light of Austrian programmatic Wertfreiheit. If their economics is to be genuinely wertfrei (in particular, free of any underlying principles of justice), 2 then we should naturally expect some Austrian economists to be socialists, some to be outright communists and still others classical liberals or libertarians for that matter. Consequently, we somehow sensed that this intimate link between political libertarianism and Austrian economics cannot be a mere contingency. Rather, we presumed that there should be some implicit conceptual connection between the two. That the above is indeed the case is most conspicuously evinced in the Rothbardian ([1956] 2011) reconstruction of welfare economics as well is in its subsequent defenses. 3 1 Austro-libertarianism as a doctrine is perhaps most explicitly referred to in Gordon's (2017aGordon's ( , 2017bGordon's ( , 2017c voluminous work, in Hoppe (2015) as well as in the recent work by Futerman and Block (2021). As Futerman and Block (2021, p. xvii) make clear, Austro-libertarianism is a combination of both "the libertarian philosophy"-the prong the authors label "the normative thread" and Austrian economics, dubbed by the said authors as "the positive thread". The reasons why-even in the absence of the label Austro-libertarianism-we can speak of the marriage of Austrian economics and the libertarian ethic are going to be provided later on in this section when we cite examples of Austrian economic findings coupled with the characteristically libertarian focus on property rights. Still more prosaically, one can think of Austro-libertarians as those Austrians who subscribe to the libertarian ethic of private property rights. 2 Certainly, not all branches of economics can be wertfrei. Any welfare economics worthy of its name hinges on some normative premise. After all, welfare economics is supposed to measure welfare and welfare must consist in some natural property or fact. And what constitutes welfare is precisely the normative question we mean. Given this, it would be unreasonable to demand that welfare economics should be free of any normativity. However, what we object to is for welfare economics to presuppose any theory of justice. And as we are about to see, it is the latter sort of normative entanglement that is damaging for the Rothbardian branch of welfare economics. 3 It must be granted that the Rothbardian branch of welfare economics does not exhaust all the sorts of welfare theories advanced within Austrian School of Economics at large. However, it would be desirable to assure the reader that the Rothbardian theory is widely shared by the most prominent Austro-libertarian scholars; see: e.g. Hoppe (1990), Gordon (1993), Herbener (1997Herbener ( , 2008, Hülsmann (1999). Still, it must be noted that some Austrian welfare theories do not follow Rothbard's footsteps. For example, Kirzner (1988aKirzner ( , 1988bKirzner ( , 1998, identifies social welfare enhancement with increasing coordination of economic agents' actions. Kirzner's criterion of welfare (i.e. coordination) contrasts with the Pareto-criterion employed by Rothbardians in that, as Kirzner (1998, p. 298) puts it, "[n]onetheless use of the Paretiancriterion does…presume a moral acceptance of the satisfaction of individual preferences as an important element of "goodness." Use of the coordination criterion involves no such moral commitment, at all, on anybody's part. Use of the coordination-criterion presumes that those advised by the economist are morally concerned that members of society undertake their actions in a way that does not inevitably spell disappointment and/or regret". In a word, Kirzner's welfare economics differs from Rothbard's in that the former is rather cautious about whether it is preference satisfaction that constitutes welfare, something Rothbard takes for granted with his employment of the Pareto-criterion and demonstrated preference. On the other hand, Kirznerian coordination clearly builds on the Hayekian (1945) defense of the open market, based on the dispersed character of knowledge. The idea is that, roughly speaking, since information in society is dispersed, central planning cannot match the efficiency of the free market. Or in other words, a central planner knows only an infinitesimally small fraction of what all the members of a society know in aggregate and therefore cannot coordinate their actions. Clearly then, there is an implicit premise in Hayek to the effect that it is coordination that translates into efficiency. Another author who jumps on Having said that, the entire Austro-libertarian edifice seems to be caught in a predicament. On the one hand, Rothbard and (much later) Herbener purport to provide a distinct economic argument for free market from efficiency. 4 And their pretenses notwithstanding, as we are about to see, their agenda fails as the idea of mutually beneficial exchanges they employ presupposes the free market. The above statement holds true since both authors find voluntariness of a given exchange sufficient to make it mutually beneficial, with voluntariness being defined solely in terms of the absence of right-violating acts or proposals. But if that is so, then it is the free market as a rights-respecting regime alone that necessitates mutually beneficial exchanges only. However, given this tight conceptual link between property rights and beneficial exchanges, whether the free market is efficient or not is not an open question any longer. The avowed economic case for the free market proves to be just an illusion. Instead, upon final analysis, the Rothbardian welfare economics merely reaffirms the author's natural-rights-based defense of the free market provided elsewhere (e.g. Rothbard, [1973Rothbard, [ ] 2006. In fact, the Rothbardian welfare economics is not just an isolated case wherein justice-related considerations are inextricably intertwined with Austrian economics. In fact, the general case for the existence of Austro-libertarianism can easily be made. To illustrate our point, let us just cite some of the most conspicuous realms of Austrian economics presupposing the libertarian ethic. For instance, Austrian entrepreneurship theory (e.g. Kirzner, 1973) rests on the assumption that entrepreneurs discover temporarily undervalued means (relative to the ends attainable therewith) while voluntarily (i.e. under the condition of respecting rights) interacting with other actors. Similarly, there are Austrians whose monopoly theory takes monopoly to be essentially a result of the government illegitimately bestowing trading 4 A word of explanation is due at this point. Certainly, efficiency as such "is the measure of how well we have achieved our goal" (Wight, 2015, p. 59). Therefore, in our present work, what we are going to mean by market efficiency is a measure of the extent to which the market contributes to people's welfare (whatever constitutes it). This tallies well with our overall agenda involving the distinction between two independent goods within Austro-libertarianism: justice and welfare, with the former being accounted for by the libertarian prong of the above doctrine, and the latter being an exclusive domain of Austrian welfare economics. the Hayekian tradition is Huerta de Soto (2009). This author stresses the temporal dimension of resource allocation. Hence, he is not preoccupied with mere allocative efficiency, that is distributing resources at a given time in such a way that no further mutually beneficial exchanges are possible, the so-called state of Pareto-optimality. Rather, this author emphasizes the fact that new information is generated over time by entrepreneurial efforts and thus economic possibilities (for mutually beneficial trades and for speculation) also expand over time. In short, for Huerta de Soto, it is this dynamic aspect-and not the problem of mere allocation of resources at a given time-that primarily matters. There is, of course, more to be said about different Austrian welfare theories but due to space constraints, we must, sadly, rest content with this necessarily sketchy overview. Suffice it to say that the present essay deals with the Rothbardian branch of Austrian welfare economics and our criticism is levelled precisely at this branch. On a final note, to avoid the tediousness of the prose, from now onwards whenever the phrase "Austrian welfare economics" is used, what will be thereby meant is the Rothbardian branch thereof unless specified otherwise.
Footnote 3 (continued) 1 3 privileges (e.g. Rothbard, [1962] 2009). Moreover, Austrian business cycle theory 5 (e.g. Huerta de Soto, [1998] 2006) interestingly relies on normative considerations (viz. banks acting unrightfully lending depositors' money). Furthermore, Austrian definition of inflation (e.g. Hülsmann, 2008) also characteristically presupposes governmental violence, and is hence also normatively tinted (viz. being rights-based). Eventually, Rothbard ([1962] 2009, p. lvi) himself conceived of exchanges as inseparable from the notion of property rights, while claiming that "[e]xchange cannot be adequately analyzed until property rights are fully defined". Just to reiterate, it is in this sense that we can validly speak of Austro-libertarianism. To wit, there are scholars, such as the ones mentioned above, who combine the Austrian positive reasoning with the libertarian ethic of property rights.
However, we take no issue with the mere fact that many Austrian theories simply assume this ethic. 6 Quite the contrary, this very fact in and of itself is not damaging at all, for Austrian economics might simply take the libertarian theory of justice as a given and then proceed with purely positive considerations, while spelling out logical consequences of the former. But then again, the most sensitive spot of Austro-libertarianism is Austrian welfare economics. For Austrian welfare theory to be meaningful, it must relate the regime of property rights (the libertarian prong) to welfare, with the latter being identified by Austrians with actual preference satisfaction (e.g. Mises, [1949] 1998, pp. 94-97).
To make Austrian welfare economics informatively contribute to the study of welfare on the free market, we argue that we should clearly distinguish between two types of good. First, in order to fit nicely within Austro-libertarian framework, Austrian welfare economics should take the libertarian ethic of private property rights for granted. And secondly, it should focus on the actual preference satisfaction, which economists at large usually take to be constitutive of welfare (see : Hausman, 2012;Wight, 2015). And only then, we can meaningfully probe the problem of how the free market relates to actual preference satisfaction without begging any question in favor of the former. Furthermore, while keeping to the uncompromising spirit of the libertarian theory of justice (see Block, 2017;Hoppe, 1993;Nozick, [1974Nozick, [ ] 1999Rothbard, [1982] 2002), 7 we shall take justice as lexically prior to welfare, 8 with the former being a non-tradable good weighing infinitely more than welfare.
The present paper proceeds in the following order: in Sect. 2 we provide an overview of the general concepts of justice and welfare as well as probe the possible relations between the two. Having done so, we try to identify the specifically Austro-libertarian notions of justice and welfare. In Sect. 3, we subject Austrian welfare economics to critical scrutiny. In Sect. 4, we hint at some formal requirements that Austrian welfare economics should satisfy to coherently fit within Austro-libertarianism as well as to be able to informatively study market efficiency. Section 5 concludes.

Welfare, justice and the relation between the two
In this section, we are going to be concerned with two types of good that are of utmost importance to our essay; that is, justice and welfare. To this end, we shall provide a brief overview of various theories of welfare and justice as well as of different ways in which the relation between the two might be conceived of. Having thus mapped the general conceptual terrain, we shall place Austrian welfare economics and the libertarian theory of justice within the sketched framework, while making sure that Austrian economics is relegated to deal with justice-independent welfare and libertarianism-with justice construed independently of welfare considerations. 9 Without doubt, welfare economics is at the bottom a normative enterprise. Whatever positive apparatus it might possibly resort to, it must ultimately rest on some normative premise. That is, for any welfare economics to even get off the ground, there should be an attempt made to identify some fact or property that makes the economic actor or society better off or worse off. 10 Incidentally, the very problem of what constitutes welfare and what can thus serve as a standard for welfare comparisons makes welfare economics such an interesting and contentious discipline. One-and very general at the same time-important division of theories of wellbeing is mentioned in Hausman and McPherson (2006, p. 119): Theories of well-being can be classified as either "formal" or "substantive. A substantive theory of well-being says what things are intrinsically good for people. "Hedonism" is an example of a substantive theory of well-being; it says that well-being is happiness or pleasure. "Formal" theories of well-being specify how one finds out what things are intrinsically good for people, but they do not say what those things are. To maintain that welfare is the satisfaction of preferences is to offer a formal theory of well-being. This theory does not say what things are good for individuals. Instead it says how to find out: by seeing what people prefer. Formal theories may be compatible with substantive theories. For example, if happiness is the ultimate object of preference, then it could be true both that well-being is the satisfaction of preference and that well-being is happiness.
As far as substantive theories of welfare go, the literature dedicated to them is enormous (e.g.; Crisp, 2006;Parfit, 1984Parfit, , 2011aParfit, , 2011bScanlon, 1998;Sumner, 1996). These days, some of the most hotly debated theories of welfare are hedonism, desire-based theory and objective list theories (e.g. Lauinger, 2017;Woodard, 2013). 11 However, economists are normally committed either to actual preference satisfaction view (a formal theory of welfare which regards people's actual preferences as proxies for what is good for them) or to the view identifying welfare with the satisfaction of needs, with the latter being subscribed to by economists bent on utilitarian calculations (Hausman & McPherson, 2006, p. 111). 11 On hedonism, what contributes to our well-being are pleasant mental states, however broadly conceived. Hedonism differs from desire-based theories in that on latter theories, a desire may be satisfied and thus allegedly render the actor better off without the actor's necessarily having an accompanying feeling of satisfaction. In other words, according to desire-based theories, what would matter for one's welfare enhancement is the sole fact that the world turned out as desired. Eventually, on objective-list theories, what constitutes one's well-being is the achievement of certain goals, such as, say, knowledge, friendship etc., regardless of whether or not achieving them translates into a pleasant mental state or not and whether they are desired or not. For an excellent and comprehensive exposition of the theories of well-being, see Parfit (1984). For the classical and simultaneously uncompromising defense of the pleasant-mental-state conception of well-being, see Brandt (1979).
Footnote 10 (continued) Hausman 2012, pp. 88-100). The usual reason why actual choices are only an imperfect (and therefore only a proxy) evidence of what is good for economic actors is that economic actors normally hold (to a lesser or greater degree) false beliefs. In an extreme case, imagine an actor who desires to quench his thirst and believes that the bottle nearby contains water and so grabs it and drinks its content. However, the bottle contains poison. Certainly, given his false belief, we cannot infer that picking up poisonous drink was (instrumentally) good for the actor as the outcome was much worse than what the actor desired; viz., he died rather than quenched his thirst. However, Austrian economists are usually concerned with benefitting in expectation. For this reason, even in the above-presented extreme scenario, we can always maintain that picking up the bottle with poison benefitted the actor in expectation. After all, the actor believed that this action would satisfy his most pressing need.
Austrian economics, as already hinted at, instantiates a formal theory of welfare. That is, it studies economic actors' choices, then infers their preferences therefrom, and finally it treats those preferences as evidential of what is good for the actors in question (see: e.g. Mises, [1949] 1998, pp. 94-97). The reason why in Austrian economics choices must be treated as only evidential (or proxies) of what is good for respective choosers is that the Austrian view has it that at the moment of choice they benefit only in expectation. That is, it might as well turn out that ex post they in fact lose due to their false beliefs. Just to provide one more example, suppose person A chooses to buy a book on economics since he desires to increase his knowledge on the subject and believes that this book will enable him to do so. However, the book turns out to be badly written and the agents ends up learning very little from it. In conclusion, buying a book was not-with the benefit of the hindsight-instrumentally good for the actor involved. After all, buying the book did not contribute much to the satisfaction of his desire. However, given more or less correct beliefs, particular choices are a pretty reliable guide to what is good for the economic agents.
Still, given our commitment to Austro-libertarianism, welfare (however broadly conceived) does not exhaust our normative realm. It is the libertarian prong of Austro-libertarianism that forces us to pay due attention to considerations related to justice. It is not only a libertarian theory of justice but theories of justice in general that serve-in one way or another-to distribute rights. Since the present essay is mainly preoccupied with both justice and welfare as they are understood within the Austro-libertarian framework, it is crucial to ask how the two actually relate to one another. Specifically, the question is: do libertarian principles of justice assign rights independently of welfare considerations or are the former a function of the latter? As for the former option, as Hausman and McPherson (2006, p. 172) put it: "The philosophical libertarian's commitment to liberty is in principle independent of any welfare consequences. But libertarians would like it to be the case that protecting freedom also makes people better-off". 12 Indeed, in libertarian theories of justice, the just distribution of (property) rights does not appeal to welfare considerations at all. Rather, a just distribution of rights is explained by the so-called entitlement theory (Nozick, [1974(Nozick, [ ] 1999, which provides two sorts of principles of justice of interest to us: (1) justice in original appropriation and (2) justice in transfer. Neither of them ties the distribution of rights to welfare. The so-called principles of justice in original appropriation explain how one may acquire ownership of a thus far unowned resource. In this regard, the two competing theories are (a) the labor theory (e.g. Block, 2008;Rothbard, 1969) and (b) the first possession theory (e.g. Epstein, 1979;Hoppe, 2015;Kinsella, 2008). The former accounts for the original appropriation in terms of mixing one's labor with a resource, whereas the latter takes first possession to count as a fact which earns the first possessor a property title in a possessed thing. As for principles of justice in transfer, libertarians recognize consent and title transfer theory of contracts. Then again, in neither of them, welfare figures as an important factor.
What is also worth mentioning is that normally for libertarians, rights count as absolute side constraints. To wit, their violation cannot be justified at all on the grounds that welfare was thereby increased. That is to say, the absolutism of libertarian rights implies that they weigh infinitely more than welfare. (e.g. Block, 2017;Nozick, [1974Nozick, [ ] 1999Rothbard, [1982Rothbard, [ ] 2002. What runs counter to typically libertarian rights absolutism are those theories that maintain that "the problem of right articulation can be formulated as one of selecting rules that maximize good consequences" (Hausman & McPherson, 2006, p. 167). Such theories are broadly labelled as consequentialist ones. Under those theories, good outcomes (or welfare for that matter) are logically prior to rights (and correlatively: to duties). To put it more precisely, it is good outcomes (or welfaremaximizing states of affairs) that mut be first identified. Only then, can one assign rights and-correlatively-duties. Still in other words, specific distributions of rights (and duties) depend on good consequences or on welfare-maximization, if you will. 13 To wit, a moral agent ought to bring about (if he can) such a state of affairs that is best. And conversely, a sub-optimal state of affairs is such that one has a duty not to bring about. Note the contrast: for libertarians, the facts that would invest people with rights are not related to good outcomes at all, with what counts as a good outcome depending on a theory of welfare. Rather, the said facts are amoral. These are either about homesteading (mixing one's labor with a resource) or about the first possession. What follows is that under libertarianism it is logically possible to have a right to do something that would not bring about the best possible outcome-a logical impossibility on the grounds of consequentialism.
Another very important strand of thought modeling rights (or, more precisely, legal institutions) along the lines of welfare-maximization is the economic analysis of law. This tradition is mainly associated with the pioneering works by Coase (1960) and then Posner (1973). Also, the insight that the common law is efficient is pervasive in Friedman (2000). Then again, libertarianism stands in sharp contrast to the Coasean and Posnerian view. 14 To reiterate, the original rights assignment according to the libertarian theory of justice in acquisition is a function of two facts: (a) homesteading and (b) first possession, with welfare-related considerations being immaterial in this respect.
There is one more interpretation of the relation between justice and welfare naturally stemming from the claim that "(…) rights (and again especially property rights) are often taken as starting points in economic analysis. For example, analyses of the distributional properties of economic arrangements often begin from a given allocation of property rights" (Hausman & McPherson, 2006, p. 165). This in turn might prompt conceiving of welfare as presupposing justice. In other words, welfare-enhancing or welfare-diminishing exchanges would be a justice-bound sphere. That is, no exchanges could even count as welfare-enhancing or welfarediminishing unless they were just in the first place. This view was subscribed to by e.g. Dworkin (1990). In his Justice and the Good Life, Dworkin construes justice as a condition of welfare. As we understand it, whatever it is that constitutes welfare would have no normative importance unless the demands of justice are first satisfied. If we were to conceive in this manner of the relation between justice and welfare within Austro-libertarianism, we would be committed to saying that a preference-satisfying or preference-frustrating exchanges are powerless to render a given economic actor better off or worse off, respectively, unless property rights were respected in the meantime. More technically, as long as the demands of justice are not satisfied (in our case, property rights are violated), then actual preference satisfaction (or frustration) is irrelevant to the overall good of a given outcome. Still in other words, when rights are infringed upon, whether some actual preferences are satisfied or indeed frustrated cannot make any difference to the overall good of an outcome. By contrast, what we take to be an overall good of an outcome is always a function of both (1) respecting or not respecting rights and (2) actual preference satisfaction or frustration, with the proviso that rights weigh infinitely more than welfare. Hence, in our model, welfare (or, more precisely, whatever constitutes welfare) matters independently of whether justice was given its due or not with the caveat that unjust outcomes with a very high degree of actual preference satisfaction are still infinitely worse on balance than just outcomes with a low degree of actual preference satisfaction. The inquisitive reader might wonder if there is any sufficient reason to prefer our view on the relation between justice and welfare to Dworkin's. What we can offer at this point is just an intuition. Suppose A is a very affluent person living in a beautiful house. However, he is stuck with an old useless sofa at the back of his garden. Person A genuinely consider the sofa an economic bad. Therefore, he would be even ready to pay for the removal of this ugly bulky item. Obviously, he would rather sell it at a positive price. Alas, there is no market for old pieces of furniture in his area. 15 Moreover, even if there were a market price for such items, A ex hypothesi would not bother to sell it. After all, he is very rich and the opportunity cost of selling the sofa would be too high. For the sake of argument, we can also stipulate that even if the opportunity cost would not be too high, A can be blissfully unaware of the existence of a market price for this sort of items. 16 Therefore, whichever scenario obtains; that is, whether A finds selling the sofa unworthy of all the effort necessary or he is simply unaware of a possibility of getting rid of it at a positive price, one day he notices that a thief enters his premises and starts 15 An incisive anonymous reviewer rightly pointed out that apart from use value, resources have an exchange value. Therefore, in presenting the present variation on the original thought experiment, we will assume away the possibility of the sofa having a market value. However, we believe that for reasons indicated in the main text, even in the presence of a market price for such old sofas, our forthcoming point would stand. 16 In this scenario, even if A were to later find out that he could have sold the sofa at a positive price, he would still be benefiting in expectation if somebody were to steal the sofa from him, given his present belief that he cannot sell the item but can only pay for its removal. stealing the sofa. To his delight, A realizes that he is just having the economic bad 17 removed for free. Moreover, it seems that both parties benefit from this illegitimate transfer: the uncompelled thief, after all, engaged in the act of theft, whereas A omitted 18 to prevent the theft. The latter might have stepped in but did not, thus demonstrating that he deemed allowing the thief to do his job preferable to anything else he saw as a possibility. And crucially, by assuming that A had a property right in the sofa and did not explicitly consent to the thief's taking possession of it, the libertarian principle of justice in transfer would condemn this act (this exchange) as unjust. Still, if the reader shares our intuition that despite the actual injustice in question, the world was rendered better at least in one other respect 19 ; viz. welfare increased, then our view that welfare matters independently of justice is given enough plausibility.
So, in the end, Dworkin's (1990) view subtly but importantly differs from the absolutist side-constraints view of rights we adopt herein. Our perusal of the relation between justice and welfare is that of a lexical priority of the former over the latter, with both and justice and welfare contributing independently to the overall good of an outcome, whereas on Dworkin's view, welfare fully depends on justice in such a way that only when postulates of justice are satisfied can welfare (or rather, whatever constitutes it) exert its normative force.
To summarize, the adoption of our libertarian prong of Austro-libertarianism commits us to the lexical priority of justice over welfare. On the other hand, Austrian welfare economics is relegated to study whatever constitutes welfare independently of whether standards of libertarian justice were met or not. The fact that we are about to study in(efficiencies) given the free-market regime does not then imply the justice-bound view of welfare. Quite the contrary, as stressed out many a time before, justice and welfare are two distinct and incommensurable goods, with the former weighing infinitely more than the latter. The reason we shall focus on the question of (in)efficiency on the free market is that we take our libertarian prong for granted. Our agenda is to provide a conceptual framework that would allow for coherent moral assessments within Austro-libertarianism. But before we formally carve out the proper area of study for Austrian welfare economics within the Austro-libertarian doctrine, we shall demonstrate why the received Rothbardian welfare economics (and the Herbenerian defense thereof) fails. 17 Note that we stipulate that A treats the sofa as an economic bad in order to comply with the Austrian principle of the impossibility of interpersonal utility comparisons. It seems that our stipulation guarantees that both parties benefit (their preferences are satisfied) due to A's theft. After all, the outcome in terms of actual preference satisfaction is identical to the one which would obtain were the current thief to consensually get the sofa for free. For other economists allowing for interpersonal utility comparisons, the assumption that A values the sofa negatively would be too strong. It would be sufficient if we intuitively conclude that the thief's act of stealing benefits him more than it harms A-the interpretation our thought experiment is designed to invoke anyway. 18 Note that on Austrian grounds, omissions count as actions. Mises ([1949] 1998, p. 13) had it that "[a]ction is not only doing but no less omitting to do what possibly could be done." Therefore, we can unproblematically conclude that A demonstrated his preference. Granted, he was omitting to prevent the theft but he acted nonetheless. 19 However, we should bear in mind that the world after the theft compared to the possible world in which no theft would occur is (infinitely) worse.

The attempt to debunk Austrian welfare economics
Before we scrutinize Rothbard's apparently economic argument for the free market, it must be first granted that Rothbard ([1973Rothbard ([ ] 2006 in another work endorses the free-market economy independently of a putatively happy coincidence that this kind of economy is the most productive: It so happens that the free-market economy, and the specialization and division of labor it implies, is by far the most productive form of economy known to man, and has been responsible for industrialization and for the modern economy on which civilization has been built. This is a fortunate utilitarian result of the free market, but it is not, to the libertarian, the prime reason for his support of this system. That prime reason is moral and is rooted in the naturalrights defense of private property we have developed above. Even if a society of despotism and systematic invasion of rights could be shown to be more productive than what Adam Smith called 'the system of natural liberty', the libertarian would support this system. Fortunately, as in so many other areas, the utilitarian and the moral, natural rights and general prosperity, go hand in hand.
And yet, in his Reconstruction of Welfare Economics, he sets himself an additional task of arguing for the free market from the concept of welfare. 20 In our view, the Rothbardian project fails and it does so on two counts. First and foremost, we believe that once we disentangle the logical relations standing behind the concepts of free market, voluntariness and being beneficial, it turns out that the entire reasoning by Rothbard is an exercise in begging the question. More specifically, we maintain that the conclusion that all market exchanges are beneficial merely restates his non-definitional premise; that is, that all voluntary exchanges are beneficial. General remarks aside, let us critically scrutinize how the Rothbardian deductive reasoning unfolds in detail.
After putatively repudiating all hitherto proposed theories of welfare economics, Rothbard ([1956] 2011, p. 320) launches his new positive account thereof by saying: Let us now consider exchanges on the free market. Such an exchange is voluntarily undertaken by both parties. Therefore, the very fact that an exchange takes place demonstrates that both parties benefit (or more strictly, expect to benefit) from the exchange. The fact that both parties chose the exchange demonstrates that they benefit. The free market is the name for the array of all the voluntary exchanges that take place in the world. Since every exchange demonstrates a unanimity of benefit for both parties concerned, we must conclude that the free market benefits all its participants.
Having quoted Rothbard at length, let us establish why we contend that the above reasoning is question-begging. For the sake of convenience, let us represent his argument in a syllogistic form: 1. All voluntary exchanges are beneficial (or at the very least, the parties thereto expect to benefit) 2. All market exchanges are voluntary exchanges 3. Therefore: 4. All market exchanges are beneficial (or at the very least, the parties thereto expect to benefit).
Also, crucially, Rothbard adds that "[t]he free market is the name for the array of all the voluntary exchanges that take place in the world." But then, premise 2 is a definition. Certainly, we take no issue with this definition. It might be simply understood as a terminological stipulation. So far, so good. What is more, we do grant the validity to the above reconstructed argument made by Rothbard. Still, we believe that the above argument suffers from a dialectical fallacy of beginning the question. For it seems to us that the only audience who would not find the above argument question-begging is the one which is unaware of the way Rothbard uses the phrase 'free market'. In other words, if a person is already convinced that all voluntary exchanges are beneficial (viz., he would accept premise 1), he would find the conclusion informative, for he would learn that free market is equivalent with all voluntary exchanges. This is closely paralleled to the example given by Walter Sinnott-Armstrong (1999, p. 176 ---------------------Mary lives in the Buckeye State. The second premise is equivalent to the conclusion, since Ohio is the Buckeye State. So Kate's argument is weakly circular. But Kate's use of the argument is still informative for Larry, who did not know that Ohio is the Buckeye State. By the same token, the Rothbardian conclusion ("all market exchanges are beneficial") is equivalent to his premise 1 (all voluntary exchanges are beneficial), since all market exchanges are voluntary exchanges. 21 So to reiterate, the above argument does not beg the question only against the audience which is not informed about the Rothbardian terminology (the use of the phrase 'free market'). However, and more interestingly, any audience skeptically inclined towards its substantive conclusion (3) 22 cannot be persuaded by the argument's only substantive premise (1) for (1) only restates (3). This in turn meets the criterion for an argument to beg the question. By analogy, anybody who knows that Ohio is the Buckeye State but casts doubt upon the conclusion of (A3) argument, that very argument cannot work, for its second premise ("Mary lives in Ohio") merely restates "Mary lives in the Buckeye state" and the latter is precisely this proposition that the person in question doubts.
Shifting back to the Rothbardian argument, let us note that it suffers from the fallacy of begging the question due to its inadequate account of voluntariness. As a brief aside, let us consider how to remedy the Rothbardian argument. On the face of it, premise 1 (viz., all voluntary exchanges are beneficial) seems very plausible. Indeed, premise 1 can be given a fairly straightforward reading which would amount to just a neat way of saying that a voluntary exchange is indicative (or evidential) of welfare-enhancement (see : Hausman, 2012, pp. 88-103;Scanlon, 1998, pp. 115-116). 23 But under this interpretation of voluntariness, Rothbard's second premise ("all market exchanges are voluntary exchanges") could be only contingently true instead of being a mere definition. In other words, it would remain an open question whether market exchanges are voluntary given a market-independent account of voluntariness. And hence, most crucially, Rothbard's conclusion that all market exchanges are beneficial would be also open to rebuttal. 24 Given the said inadequacy (if not the sheer absence) of the Rothbardian account of voluntariness, 25 we should perhaps look further to find some ruminations on how free market exchanges relate to voluntary exchanges. Luckily, this task is performed by Herbener (2008, p. 61), who is rather explicit on the above relation: Voluntary and involuntary interactions are defined in economics to recognize the distinction between cases in which it is possible to deduce that a person is better off from an interaction with another person and cases in which it is possible to deduce that he is worse off. Each person comes to an exchange with his naturally-owned property. A voluntary exchange occurs when neither trader uses or threatens violence against the property of the other. If the two persons trade the ownership of property without aggressive violence, then the exchange is voluntary. Given their natural ownership of property, each person chooses an alternative he prefers more than the non-interaction alternative. Both traders benefit. If one person violently aggresses against the property of the other person, then the exchange is involuntary. Given their natural ownership of property, the aggressor chooses an alternative that he prefers more than the non-interaction alternative and the victim is forced to choose an alternative that he prefers less than the non-interaction alternative. The aggressor benefits and the victim loses.
As we can easily see, Herbener perceives voluntariness of an exchange as a function of one and only one variable. In short, an exchange is voluntary when there is no aggression or a threat thereof on the part of either party to an exchange. And conversely, if aggression or a threat thereof occurs, then the resultant exchange is involuntary. Furthermore, Herbener contends that whether parties benefit is a function of whether the exchange was voluntary in the first place. After all, to reiterate, he says that "[i]f the two persons trade the ownership of property without aggressive violence, then the exchange is voluntary. Given their natural ownership of property, each person chooses an alternative he prefers more than the non-interaction alternative. Both traders benefit." So, in the end, his chain of reasoning is as follows: (1) voluntariness/involuntariness of an exchange is a function of only the absence/ presence of aggression and/or threat thereof, respectively. (2) whether the parties to an exchange benefit/do not benefit is a function of only voluntariness/involuntariness of that exchange, respectively. So, ultimately.
(3) whether parties to an exchange benefit/do not benefit is a function of only the absence/presence of aggression and/or threat thereof, respectively.
But note, in final analysis (as (3) has it), whether parties to an exchange benefit or not does not depend on whether the parties exchange voluntarily or involuntarily. Rather, positively speaking, the prediction is that both parties to an exchange benefit from it if it involves no right violation and that the exchange fails to be mutually beneficial if it does involve a right violation. 26 Somehow voluntariness of an exchange was taken out of the picture without any loss in explanatory or predictive power of the theory under consideration. If that is so, Herbener errs claiming that Rothbard makes a distinct economic argument for a free-market regime. If our above analysis is correct, Rothbard (rather implicitly) and Herbener (quite explicitly) merely restate the libertarian ethic of property rights they both adhere to. After all, introducing the rights-based concept of voluntariness cannot by itself render the argument economic in its nature. Just to reiterate, to discover that parties benefit from market exchanges is to have a market-independent account of voluntariness. Falling short of 26 Note that the rights-based concept of voluntariness abounds in subject-matter literature. One of the most outstanding proponents of the moralized view of coercion and voluntariness is Wertheimer (1987). In turn, the defender of two baselines (one moral, and the other statistical) for assessing coerciveness of proposals was Nozick (1969). Olsaretti (1998), on the other hand, levelled a very illuminating criticism at the rights-based concept of voluntariness. There is no room to adjudicate between various philosophical accounts of voluntariness. We rather take the rights-based concept of voluntariness for granted but demonstrate that this concept by itself cannot provide a synthetic argument for free-market efficiency as it already presupposes free-market arrangements.
this requirement, what Herbener with his explicit rights-based account of voluntariness at most establishes is, rather trivially, that market exchanges are just-a far cry from settling the issue of whether market exchanges are also welfare-enhancing.
To conclude this critical section, let us invoke a certain thought experiment aimed at illustrating that, as opposed to what seems to follow from Rothbard and his subsequent defenders' claims, the relation between rights-based voluntariness and welfare maximization is not that of equivalence. As can be inferred from the above quote from Herbener, Rothbardians would believe that every rights-respecting exchange is mutually beneficial and that every beneficial exchange implies respecting rights. By contrast, we would hold that some voluntary (in the sense of respecting rights) exchanges are not welfare-maximizing. 27 To illuminate this claim, let us resort to the following intuition.
Suppose an economic actor faces two following proposals: (1) Blackmail case: A blackmailer comes to you saying: "I will reveal your secret to the public unless you pay me $ 1000." Given the newly constrained options, you prefer to accede to the blackmailer's demand, you pay $ 1000 and there are no further consequences.
Given the newly constrained options, you rationally choose your life, thereby giving up $ 1000. The gunman disappears with your money and there are no further consequences. (Incidentally, for the original thought experiment and the criticism levelled at it, see Wysocki andMegger (2019, 2020), Megger (2021) and Wiśniewski (2019), respectively). Now, our two proposals are allegedly so designed that they appear to differ in only one respect. Allegedly, they differ as far as their respective moral legitimacy is concerned. Libertarians maintain that moral legitimacy of proposals in turn depend on the independent moral legitimacy of the threat element (see Block, 2017;Block & Gordon, 1985). The threat element is what the threatening party proposes if the victim refuses to accede to the threatener's proposal. So, in case (1) the threat element is to reveal your secret to the public, whereas in (2) it is to shoot you. And now, since it would be independently (in the absence of the proposal) legitimate in a free society to reveal your secret to the public (with any possible prior voluntarily incurred contractual duties not to reveal it assumed away), the very proposal (1) is also legitimate; that is, it does not constitute an act of aggression. By the same token, in case (2), since shooting you would be independently illegitimate (again, assuming away the possibility that you voluntarily gave up your self-ownership rights), proposal (2) is also illegitimate.
On the other hand, let us try to describe our two cases abstracting from their normative aspect. If we consider both exchanges in ex post terms, their outcomes seem to 1 3 be identical. Threatening parties are paid for not doing what they threaten to do. Once they are paid, they are gone with no further consequences ensuing. If one is ready to grant that these two outcomes are economically identical and that their only distinctive feature is that the payment to a blackmailer (case 1) was voluntary (due to a nonright-violating character of the blackmailer's proposal), and therefore valid, while the payment to a gunman (case 2) was involuntary (due to a right-violating character of the gunman's proposal) and therefore invalid, then one must conclude that the question of whether a given economic actor benefits or not does not (or at least not only) depend on the question whether a proposal threatens a right-violation or not.
However, it might be objected that at the very least, in blackmail case a blackmailee buys a negative right against a blackmailer. From then onwards, the former holds a right against the latter, which correlates with the fact that the latter is now under a duty not to reveal the former's secret to the public (see Hohfeld, 1913). In effect, the blackmailer gave up his liberty to reveal your secret for money. And you in turn earned a negative right against the blackmailer, whereas in case (2) no such (seemingly beneficial) rights redistribution takes place. We grant such an objection has some force. Still, what we are concerned with at this point is a non-normative description of an ex post situation. Our very thought experiment is designed to imply that the question of whether we benefit (in the sense of our welfare being enhanced) or not is only contingently related to whether rights are respected or not. More specifically, as we already suggested, we contend that some voluntary, viz. rights-respecting, exchanges are not beneficial. And thus, our considered judgement is that regardless of the fact that in blackmail case (1) we earn the said right, it still might be as bad (in terms of welfare) as case (2), in which no right was earned at all. What is more, (1) may be ex post even worse (only as far as welfare goes), 28 for there still might be a plausible ex post expectation that a blackmailer might blackmail me again despite the fact I have already earned a right that he does not blackmail me ever again. After all, there is no contradiction between the fact that the blackmailer 28 To fully appreciate the fact that as far as welfare alone is concerned paying the gunman may as well turn out to be more beneficial than paying the blackmailer, let us construct a slight variation on the gunman's proposal (2*) such that now the proposal (i.e. "Money or your life") is morally legitimate rather than illegitimate. This move would render the gunman's proposal morally on a par with the blackmail proposal, which would in turn allow us to compare the two solely in terms of their respective impact on their recipients' welfare. To this end, suppose the recipient of the gunman's proposal killed the gunman's family member before. Suppose further that there is no institution around able to enforce the gunman's right to, at the very least, compensation and so the gunman's proposal serves as the only legitimate way to enforce his right. To remove the reader's uneasiness due to the fact that a right to compensation is enforced by such drastic means (i.e. "or I will kill you), we can also stipulate that the gunman indeed tries to force the proposee to pay him the compensation but that the gunman is bluffing at the same time; that is, he does not intend to kill the proposee in case of the latter's non-compliance. Moreover, it should be borne in mind that, ex hypothesi, the proposee killed the gunman's family member in the first place. Given all this, so constructed, 2* (i.e. legitimate gunman's proposal) may differ from the blackmail proposal only with respect to its impact on the proposee's welfare, as their moral character of the two proposals is now identical by assumption. And, interestingly, it might as well be the case that paying the gunman proves to be more beneficial both ex ante and ex post than paying the blackmailer. And if so, since 2* is assumed to be morally legitimate, then there is no obstacle to treat paying the legitimate gunman as better overall than paying the blackmailer. After all, with both of these proposals being presumed to be morally legitimate, it is only their differential influence on the recipients' welfare that allows us to decide which of the two proposals yields a better result overall.
can blackmail me but he may not do so. If this intuition is correct, this would again show that voluntariness alone cannot render an exchange beneficial in a welfareenhancing sense, everything else equal. 29 The same point applies to an ex ante analysis of our two cases contrasted above. After all, is it not possible that even if I concluded a contract with a blackmailer, I might still reasonably expect a breach and even though I did not conclude any contract with a gunman, I do not expect him to come back and threaten me again? If so, this might show that in expectation I can benefit more by paying the gunman than by paying that blackmailer. But then again, if that is so, then benefitting is by no means as intimately related to rights as Rothbard and Herbener believe it to be. Crucially, if our audience shares our intuition, the two cases cited above fly in the face of the Herbenerian contention that a beneficial character of the exchange proves to be a function of one aspect thereof, that is its voluntariness, with voluntariness, in turn, depending only on whether prior rights of the parties to the exchange get violated or not. 30 Having levelled the criticism of Austrian welfare economics, we are about to commence the positive part our paper. Section 4 presents some formal requirements Austrian welfare economics should satisfy in order to fit within Austro-libertarian paradigm and to be able to non-question-beggingly study market (in)efficiency.

An exclusive domain of study for Austrian welfare economics within the Austro-libertarian framework
It is now high time to carve out a proper area of study for Austrian welfare economics so that it should be compatible with the idea of justice assumed by the libertarian prong of Austro-libertarianism. Note that the way we conceive of the relation between justice and welfare guarantees that the above condition is met. After all, we understand welfare as contributing independently of justice to the overall good of an outcome. Our view of conceiving of welfare as a sphere fully independent of justice considerations has this welcome advantage that thus modelled welfare economics would be in principle applicable to any political regime. However, given our 29 There is an easy-albeit more artificial-way to construct two scenarios that do indeed differ only in their moral legitimacy. Let us stipulate (1*) case (illegitimate blackmail). Suppose I did pay off the blackmailer in (1), thus concluding a contract whereby I earn a right that the blackmailer does not blackmail me ever again and he incurs a correlative duty of the same content. Alas, he approaches me again with the same proposal: "I will reveal your secret to the public unless you pay me $ 1000." Unwillingly, I pay him off again. However, this time I pay him involuntarily since the proposal violated my newly earned right. Still, we cannot see why (1) and (1*) should differ economically. 30 We anticipate that some Austrians would at this point retort that we cannot resort to some other intuitive notion of benefitting without begging the question against them. And this point would be correct.
On the other hand, the very purpose of our appeal to the above intuition is that if our reader (intuitively) finds both cases economically identical despite their moral differences, then perhaps there is something prima facie wanting about the Rothbardian view linking benefitting with the absence of aggression. After all, is there any decisive reason to treat the assumption that benefitting is about rights-respecting as unshakable? Instead, why not apply modus tollens, thereby arguing that this assumption is to be jettisoned? commitment to Austro-libertarianism, we are forced to say that no increment in welfare is able to compensate for any (even infinitesimally small) decrease in justice.
In other words, our overall normative framework provides that the primary good to be maximized is justice, with justice being non-tradeable for welfare. Coming back to our sofa thought experiment, our Austro-libertarian account predicts that even though the owner of the sofa got his economic bad removed for free by the thief, this outcome is still infinitely worse on balance than any one in which rights would have been respected. And the reason for this prediction is that in the scenario just cited justice was compromised. If, instead, the sofa owner had not had the sofa stolen, he would have been still stuck with the economic bad. That is, in this situation, although his well-being would be compromised, the overall outcome would be still better compared to the previous one. And what makes it the case is that in the latter scenario justice was served. Now, given the fact that within the Austro-libertarian paradigm justice lexically dominates welfare, Austrian welfare economics would serve to study welfareenhancing and welfare-diminishing exchanges on the free market. 31 In other words, since libertarian rights act as a side constraint on maximizing welfare, the corresponding welfare economics would necessarily operate in the universe of voluntary (i.e. rights-respecting) exchanges. Remember, just as the budget line limits the height of an indifference curve, so do libertarian (absolute) rights limit the welfare maximization. However, just as it is at least logically possible to end up at a lower indifference curve somewhere below the budget line, so it is at least logically possible to have rights-respecting but welfare-diminishing exchanges.
At this point, it is worth noting yet again that such a project would be informulable on the grounds of the Rothbardian branch of Austrian welfare economics. For Rothbardians take market exchanges (via the rights-based concept of voluntariness) as sufficient for welfare-enhancement. But if so, then for Rothbard and his followers, market exchanges that would be at the same time welfare-diminishing are impossible, something that should raise a suspicion in the first place. By contrast, the intuitive appeal of our position is that it allows for welfare-diminishing exchanges on the market. We do not claim that such exchanges are indeed exemplified on the market but that they are at least logically possible. And it is precisely for this reason that welfare economics has some distinct work to do. That is, on our account, quite unlike in Rothbard's, the ideal of welfare does not in fact collapse into the one of justice. For instance, note that on our account the blackmail proposal was regarded as being economically on a par with the gunman's proposal. That is, two morally distinct proposals (one legitimate, and the other not, respectively) did not apparently 31 Remember, given the side-constraint character of libertarian rights, pursuing utility at the cost of violating rights would be banned on independent moral grounds. As we stressed time and time again, it is the libertarian prong that constraints welfare-related considerations. To wit, welfare may be permissibly maximized but not beyond the limits set by inviolable libertarian rights. In fact, we can think of side constraints as budget constraints. Just as we can maximize utility up to the budget line so we may maximize utility as long as we do not violate libertarian rights. Whereas in the former case the constraint is economic (the amount of money we have), in the latter case the constraint is normative (i.e. in your pursuit of utility, you may not cross the line settled by absolute libertarian rights). differ in terms of their respective effects on utility. And this shows, rather plausibly, that welfare is separate from justice. Now, what is left to specify is how thus carved welfare economics would be supposed to proceed. We maintain that assessments made by welfare economics should be based on two premises. First and foremost, it should be explicit about its normative premise; that is, it should identify some property (or a fact) constitutive of welfare. 32 Note that this task is not something positive economics can be relegated to do. Instead, once the purely normative works is done (i.e. what constitutes wellbeing gets established), the purely positive part of economics would enter, providing us with the means to maximize that which matters.
Our position towards welfare economics is thus structurally well aligned with that of Kirzner's (1998, p. 291): Economic science cannot, it appears, be deployed for social-normative purposes, only because the criteria we wish to apply in our evaluations cannot be coherently defined except as pure (non-scientific) judgments of value. Positive science may not, without appealing to non-scientific judgments of value, be able to pronounce heart disease desirable or undesirable. But these judgments of value are not needed to identify heart disease (and thus to be able scientifically to identify its causes).
By the same token, it does not take a value judgement to determine whether some preferences are satisfied or frustrated. Moreover, purely positive economics is in principle able to equip us with the guidelines of how to maximize preference satisfaction or even-however bizarrely-how to maximize preference frustration. And yet, it takes normative economics; that is, it takes a judgement of value to say that it is indeed the maximization of preference satisfaction that is desirable.
So then again, welfare economics should rest on both (a) a normative premise and (b) technicalities (the part of positive economics demonstrating how to causally get from means to desired ends). And the fact that the assessments (or policy recommendations) of welfare economics are informed by the said two premises makes the very discipline so contentious. For consider we take actual preference satisfaction as constitutive of welfare and we know how to satisfy the preferences in question (we know which actions to take to this end) but our interlocutor disagrees with our conclusion that such and such actions (or exchanges) would be welfare-enhancing. There are as many as three logical possibilities as to the premises he may deny. First, he may accept our positive (technicalities-related) part but deny that actual preference satisfaction constitutes welfare. Alternatively, he may concur that actual preference satisfaction is indeed constitutive of welfare but we may diverge as far technicalities are concerned. Or indeed, he may deny both premises at the same time, in which case he may reach the same conclusion as we do only by accident. In short, two persons reaching different policy recommendation does not imply that they necessarily subscribe to different moralities (or that either person's positive economics is necessarily deficient).
There is just one loose end to tie up. If actual preference satisfaction is to be this property that constitutes welfare, at least some sketchy remarks pertinent to the former are due. Unfortunately, our agenda was meant to operate on a rather formal level. Thus, specifying what would count as an actual preference-satisfying or preference-frustrating exchange is beyond the scope of the present research, for it would necessarily involve us in investigating the problem of the relevant baseline. 33 However, if we were to venture a hypothesis as to which exchanges translate into welfare-enhancing ones and which into welfare-diminishing ones, we would construe this distinction more or less along the lines of Nozickian ([1974] 1999, pp. 84-87) productive and unproductive exchanges, respectively. 34 However, building on the Nozickian account, while trying to overcome the problems haunting it, would take a separate paper.

Conclusion
That aim of the present paper was primarily critical. We attempted to demonstrate that the Rothbardian variant of Austrian welfare economics cannot really prove its points. Its avowed programme of establishing that voluntary (i.e. rights-respecting) exchanges are mutually beneficial and-conversely-that involuntary (i.e. coerced) exchanges are non-beneficial fails. We established that both Rothbard and Herbener beg the question in favor of the free market by conceptually tying beneficial exchanges to market exchanges (via the concept of rights-based voluntariness). On their grounds, market exchanges simply necessitate mutually beneficial exchanges. In our turn, just to avoid begging the question against Rothbard we did not just bluntly deny his premise that all voluntary exchanges are mutually beneficial but instead we presented a thought experiment that might give us a sufficient reason to believe that, in terms of welfare, a voluntary exchange might be as bad as an involuntary one, which was meant to show that voluntariness (or justice for that matter) of an exchange should be conceived of independently of the question whether the exchange is mutually beneficial or not.
As far as the positive part of the present work goes, first, we made some (both historical and conceptual) case for Austro-libertarianism as a serious and already well-entrenched research programme. So, instead of trying to free Austrian economics of any normative assumptions, we conformed to the current practice of Austrian economists presupposing the libertarian ethic of property rights. Trying to fit a slightly remodeled Austrian welfare economics into the fully-fledged Austro-libertarian doctrine, we realized that we needed to distinguish between two goods: justice and welfare. And we did so in accordance with the characteristically libertarian construal of rights as absolute side constraints, with justice being lexically prior to welfare. However, in the light of a thought experiment at the end of Sect. 2, we putatively established that welfare still independently contributes to the good of an overall outcome.
Moreover, we concluded that it is only when the draw the above-mentioned distinction between justice and welfare that Austrian welfare economics is given a distinct job to do. From then onwards, the entire Austro-libertarian normative framework can make coherent moral assessments: the libertarian prong evaluates the justice of a given exchange according to the libertarian principles of justice, whereas Austrian welfare economics studies welfare-enhancing or welfarediminishing exchanges, whatever the political regime. However, the adoption of Austro-libertarianism commits us to maximizing, first and foremost, justice and only secondarily-welfare. That is, a just exchange is always better than an unjust one regardless of welfare-related considerations. On the other hand, given that two given exchanges are equally just, the dimension of welfare makes all the difference to overall assessment of the two exchanges in question. And last but not least, only when welfare considerations are effectively divorced from justice considerations can we meaningfully (viz. non-question-beggingly) study the problem of market efficiency, that is to what extent does the market regime contribute to the actual preference satisfaction (or to whatever constitutes welfare).
1 3 are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.