Rothbardian Welfare Economics
Every economics undergraduate is taught that economics is a positive science. Introductory textbooks always take the space to emphasize that the economist qua economist can never establish ethical judgments. In his capacity as a social scientist investigating economic problems, he can only describe and explain the world as it is, never as it ought to be.
For example, basic economics teaches us that fixing the price of milk below the price that would have been established on the free market, will lead to a shortage. Likewise, setting wages above market clearing prices will bring about involuntary unemployment. These propositions say nothing about the desirability of their consequences. They are wertfrei.
But then a different question arises: can economics tell us the effect that a particular change will have on social welfare? Can it establish when "social utility" is maximized, and if so, how do we arrive there? If the answer to these questions is affirmative, it would seem that prima facie the economist could nevertheless make political-ethical statements without violating the value-free character of his science.
The branch of economics that deals with these questions is known as welfare economics. It is an arcane subfield that historically went through rough patches, ups and downs, premature deaths, and several reincarnations. In what follows, we will briefly outline the evolution of welfare economics up to the time of Rothbard's contribution, discuss its implications, and review the critiques charged against it.
The classical economists like Smith, Ricardo, and Mill had a primitive, pre-subjective conception of welfare. They argued that one should adopt those public policies that aim to maximize physical output (usually referred to as "Bushels of Corn"). Based upon their positive analysis, the way to achieve this goal was straightforward: promote the scope of the division of labor to the greatest possible extent and accumulate the largest amount of capital.1
While the classical economists were mostly correct in their analysis of the effects of the division of labor and capital accumulation, they erred in thinking that the physical results alone could furnish them with justification for their policies. For the subjectivist, neoclassical economist the fault in their thinking is obvious: welfare is not dependent upon the objective, physical amount of consumer goods in society; it is rather a function of people's preferences and their ability to satisfy them.
It is quite possible that more consumer goods, all other things being equal, can mean a higher level of welfare, but these goods must come at a cost. If the cost to produce them, e.g., forgone leisure, is of higher value than the needs that these goods will satisfy, then more goods would lead to less welfare.
Moreover, following the reasoning of the classical economists, one would think that it is possible to increase social welfare by forcing people to work more than they wish to, or by transferring wealth from poor individuals with high time preference to wealthy individuals with lower time preference, thus increasing the total amount of capital.
The Rejection of Old Welfare Theory
With the arrival of the marginal revolution and its emphasis upon the subjective, individualistic conception of welfare, Classical Welfare Economics was rejected. Instead, neoclassical economists, led by Pigou, Edgeworth, and Marshall, utilized the new value theory and the Law of Diminishing Marginal Utility (LDMU) to develop what came to be known as Old Welfare economics. The main argument was that since everyone has a diminishing utility of money, the marginal utility of income of a rich person is smaller than that of a poor person. Therefore, a transfer of income from the rich to the poor, given that it would not overly hamper production, will increase "total utility," and is thus economically legitimized.
If this argument sounds familiar, it is no accident. Politicians and economists appeal to this line of reasoning even to this day. Recently, Paul Krugman used it to support the attempt by U.S representative Alexandria Ocasio-Cortez to introduce a higher top marginal income tax rate.
This argument might seem convincing at first glance, so it is worthwhile to spell out precisely the assumptions it involves. First, the Old Welfare theorists assumed that all people have the same capacity for satisfaction. They admitted that this was no more than a metaphysical presupposition, but it was a reasonable starting point and mostly harmless. Second, and of much greater importance, was the assumption of cardinal utility. Following the pioneers of the marginal revolution, Jevons and Walras, the Old Welfare theorists understood utility as a quantifiable physiological magnitude that has an existence beyond the choice of the individual.2 They thought that this quantity lends itself to mathematical operations and aggregation. Under this assumption, Interpersonal Utility Comparisons (IUC) are permissible, and therefore, it makes sense to conclude that "total utility" has increased, even if some people are worse off.
The endeavors of the Old Welfare theorists came to an abrupt end with Lionel Robbins's demonstration of the futility of IUC. Robbins showed that these economists erred in extending the LDMU beyond its proper field, for the LDMU applies only to the economizing individual in employing means to satisfy his ends. We can speak of an individual ranking goods on his value scale and the diminishing utility of additional units. We can explain an exchange by referring to the opposing ranking of the goods on the individuals' value scales. It is an entirely different thing, however, to compare this value rank between individuals. Furthermore, since there is no objective unit of measurement for utility, it is impermissible to speak of quantitative differences in satisfaction. Thus, Robbins concluded, the argument advanced by the Old Welfare theorists is nothing but an ethical judgment, and as such, must be excluded from the economic science.
Barred from making IUC, economists were forced to adopt the so-called Unanimity or Pareto Rule. The Pareto Rule, first developed by Italian economist Vilfredo Pareto in 1906, maintains that we can speak of increasing social utility only when one individual is better off without causing another to be worse off. When a change brings this about, it is called Pareto-Superior. When there are no more Pareto-Superior moves left, the situation is called Pareto Optimality; otherwise, the situation is Pareto Inferior.3
The Pareto Rule stood as the test which all statements concerning social welfare must pass if they are to remain value-free. If two individuals partake in a trade or an individual acts without harming anyone else, then the economist would be permitted in deducing that social welfare has indeed increased. However, if a group of individuals gains at the expense of another group, as happens with all state interventions, then the economist cannot conclude anything meaningful about social welfare.
It is this constraint that the New Welfare Economics of the mid-twentieth century sought to evade and instead made an economic case for state intervention. Two different paths were taken to this end: the first, associated with Harvard University, trivialized the Pareto Rule by incorporating it into a general equilibrium framework. The second, emerging from London School of Economics, circumvented the rule with the help of the Compensation Principle. The first path led to the development of the social welfare function (SWF) and the concept of market failure, the second to the Kaldor-Hicks Compensation Criterion.4
The social welfare function was first pioneered by American economist Abram Bergson5 and later developed by Paul Samuelson.6 This approach trivialized the Pareto Rule by adopting the Pareto Optimality version of it, thus focusing on the static, end-state market outcome. By establishing several efficiency conditions, the SWF would derive an Optimal Pareto equilibrium which maximized social welfare. Failure to achieve this maximum could be then used to justify state intervention.7
This procedure is analogous to finding the optimal consumer goods bundle for an individual's utility function subject to a budget constraint. Only that instead of an individual's indifference curve and a budget constraint, a social indifference curve is maximized at the point of tangency with a so-called Utility Possibility Frontier, similar to a Production Possibility Frontier.
Right from the start, the SWF went through harsh criticism by fellow neoclassical economists, mostly because it failed to eliminate cardinal utility and IUC from its analysis. However, the final nail in the coffin came with Kenneth Arrow's famous "Impossibility Theorem."8 Arrow demonstrated that it is impossible to construct a SWF that simultaneously satisfies several elementary conditions. Hence, there is no method of aggregating individual preferences which leads to a consistent social preferences scale. The SWF had to be abandoned.
The market failure approach had a much more fruitful existence. This relies on the so-called first and second Fundamental Welfare Theorems. The first welfare theorem states that under the assumption of perfect competition,9 the market will inevitably arrive at a Pareto Optimal equilibrium. The second theorem states that allowing for an initial transfer of income between individuals and then letting the market run its course, the market would still reach a Pareto Optimum.10
The appeal of this approach lies in the fact that economists could then point to cases in the real world where the market fails to achieve this result. Throughout the latter half of the twentieth century, hundreds of papers were published on this topic, each trying to find new types of market failures, and calling for state intervention to fix them. Among the cases of market failure that have survived in the literature up to this day are public goods, asymmetric information, natural monopoly, and externalities.
Alongside the Harvard approach, LSE economists John Hicks and Nicholas Kaldor developed the "Compensation Criterion." This criterion states that we may speak of increasing social welfare when the winners can hypothetically compensate the losers and remain better off. That this compensation need not actually take place is irrelevant. Economists could utilize this criterion to recommend certain policies without involving value judgments yet remain within the Pareto Rule.11
A classic example frequently used to illustrate the superiority of the Compensation Criterion over the Pareto Rule was that of the repeal of the Corn Law in the nineteenth century. Economists have pointed out that it would be impossible for the economist qua economist to support this kind of measure. Despite obviously making everyone better off in the long run, abolishing the tariff would have hurt the short-run interests of the landlords. Hence, Kaldor and Hicks argued, the Compensation Principle avoids this pitfall by only asking that the gains be hypothetically distributed among the losers.
Rothbard Enters the Scene
This was the state of welfare economics when Murray Rothbard arrived on the scene with his seminal article "Toward a Reconstruction of Utility and Welfare Economics." There was at the time but a few dying theories scattered around without any viable future in sight. Rothbard's solution for the revival of the field was simple yet profound. It consisted of placing the Pareto Rule within the constraint of demonstrated preference. The notion of demonstrated preference is straightforward: the economist can only know of the individual's preferences through their actions. Every action implies a choice; when an individual chooses A over B, he demonstrates that he prefers A. The economist cannot deduce from this how much does the individual prefers A, for this is purely subjective and ordinal and is not revealed to him through action. Moreover, under demonstrated preference, it is impermissible to construct hypothetical value scales that contradict the preferences which individuals themselves make known in action.
As Rothbard went on to show, the implications of confining welfare economics within demonstrated preference were far-reaching. First, the whole focus on conditions of Pareto Optimality was fallacious. Since the never-never land of perfect competition could not materialize in the real world, it is thus futile to analyze changes in social welfare under these ideal conditions. On the contrary, preference can only be demonstrated in real-world markets, where actors are neither price takers nor omniscient. Hence, complaints about producers selling products at above marginal costs, asymmetric information, and natural monopolies, are irrelevant.
Instead, welfare economics teaches us that by participating in the free market, both parties to every exchange demonstrate that they expect to benefit themselves. In other words, every voluntary exchange increases utility in the ex-ante sense. According to Rothbard, then, at every point in time, by going from one Pareto-Superior move to another, the free market maximizes social welfare.
Furthermore, restricting statements about social welfare to demonstrated preference illustrates the absurdity of the Compensation Criterion. If no actual compensation from the gainers to the losers takes place, then the economist cannot conclude anything meaningful about social welfare. He cannot know whether people actually prefer the new situation to the old one. However, if the compensation really did happen, given that it was made voluntarily, then the Compensation Criterion merely collapses into the old and familiar Pareto Rule.
In the aftermath of Rothbard's article, several economists have criticized his thesis. First, it was argued that Rothbard's approach still fails to solve the problem of escaping from the status quo, as in the example of the repeal of the Corn Law. The landlords, it is said, preferred having the tariff in place, as demonstrated by their vocal opposition to Parliament's decision. However, it is wrong to assert that this opposition shows that the landlord's welfare was reduced after the change. As far as the economist knows, the landlords could be lying or just playing a game. What the economist can observe, however, is that after the laws were repealed, the landlords reentered voluntary contracts, demonstrating they were benefiting from the free market. Therefore, in the new state of affairs, all parties mutually gained; as opposed to the old state of affairs, where the welfare of the rest of the population was reduced due to the tariff.
In his controversial article, economist Bryan Caplan objected that the same argument above could be pointed against Rothbard's thesis. Regarding the claim that the emotions of a third party cannot invalidate the conclusion that each voluntary exchange increases social welfare, since (as Caplan quotes Rothbard) "even if he publishes a pamphlet denouncing these exchanges, we have no ironclad proof that this is not a joke or a deliberate lie," Caplan writes:
"Rothbard could have taken this principle further. When two people sign a contract, do they actually demonstrate their preference for the terms of the contract? Perhaps they merely demonstrate their preference for signing their name on the piece of paper in front of them. There is no "ironclad proof" that the signing of one's name on a piece of paper is not a joke, or an effort to improve one's penmanship."
However, signing a contract is not just penmanship or a game; it is rather, as Walter Block points out, a binding activity that transfers ownership of goods from one person to another. It, therefore, obligates the person to act according to the terms of the contract. Thus, a person entering a legal agreement cannot be said to engage in play.
In addition, economist Roy Cardato has argued that Rothbard is mistaken in focusing on utility ex-ante alone. In reality, people may have false expectations of the future and thus can and do lose utility in the ex-post: "Rothbard's welfare economics... ignores the fact that preferences are expressed sequentially through time, as part of a general set of goal-oriented activities."
Although Cardato is correct in his observation, he is mistaken in thinking that this poses a problem to Rothbard. First, as economists, we cannot infer anything about ex-post utility from demonstrated preference. This knowledge is barred to us. It is, therefore, irrelevant for all welfare economics, not just Rothbard's. Second, as Austrian economics teaches us, the unhampered market is the best institution for mitigating errors and thus maximizing ex-post utility. The market has a built-in mechanism for eliminating unsuccessful entrepreneurs and unwanted products. It is certainly possible that a consumer may be unsatisfied with a single purchase, but it is hard to imagine that the seller of the product would stay in business for long.
Lastly, Cardato misunderstood Rothbard's objective. Rothbard did not try to create a whole ethical or philosophical basis for the free market with his welfare economics. This could be found in his natural law theory of property rights. Instead, he only set out to show how demonstrated preference can be utilized to salvage this field from pure ruin, providing us with a framework upon which ethical claims could be based.12
Now just one last obstacle remained for Rothbard. The First Welfare Theorem, if it can fit into Rothbard's analysis, would not apply to the unreachable state of perfect competition. Instead, it would apply to real-world markets, stating that the free market ensures the highest level of welfare possible, compared to other actual institutional systems. However, the Second Welfare Theorem was left untouched by Rothbard. Economists could still advocate for a single lump-sum transfer of income to achieve their preferred state of equality, and then let the market run its course.13
It was left to Rothbard's protégé, Hans Hoppe, to find the solution. Hoppe pointed out that the New Welfare economists were entangled in a logical contradiction. On the one hand, they accepted the social welfare consequences of voluntary exchange from the individuals' point of view. Thus implicitly embracing the principle of self-ownership. But on the hand, they refused to accept its logical consequence — the Lockean principle of homesteading and acquisition.
If welfare economics must start from the undeniable fact of self-ownership, then it must apply the Pareto Rule both to the use of property and its acquisition. As Hoppe wrote:
"a person's original appropriation of unowned resources, as demonstrated by this very action, increases his utility (at least ex-ante). At the same time, it makes no one worse off, because in appropriating them he takes nothing away from others. For obviously, others could have homesteaded these resources, too, if only they had perceived them as scarce. But they did not actually do so, which demonstrates that they attached no value to them whatsoever, and hence they cannot be said to have lost any utility on account of this act. Proceeding from this Pareto-optimal basis, then, any further act of production, utilizing homesteaded resources, is equally Pareto optimal on demonstrated preference grounds… And finally, every voluntary exchange starting from this basis must also be regarded as a Pareto-optimal change, because it can only take place if both parties expect to benefit from it."
Rothbard, probably against the wish of his fellow economists, succeeded in reconstructing welfare economic by confining it to the Pareto Rule and demonstrated preference. He showed that the free market, i.e., the network of voluntary interactions between individuals, always produces the greatest degree of social welfare possible. Government intervention, on the other hand, could never be justified in terms of welfare. And while Rothbard’s contribution probably ranks among his least-known achievements, it is truly a tour de force and another tribute to his great originality and talent as an economist.
Originally published at AustroLibertarian.com | Subscribe to Austro-Libertarian Magazine here
- 1. Hla Myint, Theories of Welfare Economics (London, U.K.: Longmans, Green and Co. 1948), p. 12.
- 2. "But to the preceding generation of economists, interindividual comparisons of utility were made almost without question; to a man like Edgeworth, steeped as he was in the Utilitarian tradition, individual utility — nay social utility — was as real as his morning jam. And with Marshall the apostrophe in consumers' surplus was always after the s." Paul Samuelson, Foundations of Economic Analysis (Cambridge, Mass.: Harvard University Press, 1947), p. 225.
- 3. Vilfredo Pareto, Manual of Political Economy (New York: Augustus M. Kelley,  1971).
- 4. Jeffrey Herbener, "The Pareto Rule and Welfare Economics," Review of Austrian Economics 10 (1997): 86.
- 5. Abram Bergson, "A Reformulation of Certain Aspects of Welfare Economics," Quarterly Journal of Economics 70, no. 2 (February 1938): 310–34.
- 6. Paul Samuelson, Foundations of Economic Analysis (Cambridge, Mass.: Harvard University Press, 1947), pp. 219-229.
- 7. The procedure is analogous to finding the optimal consumer goods bundle for an individual's utility function subject to a budget constraint. Only that instead of an individual's indifference curve and a budget constraint, a social indifference curve is maximized at the point of tangency with a so-called Utility Possibility Frontier, similar to a Production Possibility Frontier.
- 8. Kenneth Arrow, Social Choice and Individual Values, 2nd ed. (New York: John Wiley and Sons. [I951] 1963).
- 9. The model of perfect competition consists of the following three assumptions: all actors are price taker, i.e., cannot influence the market price, no transaction costs, and homogenous product.
- 10. Mark Blaug, “The Fundamental Theorems of Welfare Economics, Historically Considered,” History of Political Economy 39, no. 2 (2007): 185–207.
- 11. John Hicks, "The Foundations of Welfare Economics," Economic Journal 49, no. 196 (December 1939): 696–712.
- 12. David Gordon, "Toward a Deconstruction of Utility and Welfare Economics," Review of Austrian Economics 6, no. 2 (1993): 103–4.
- 13. Jeffrey Herbener, “Hoppe in One Lesson, Illustrated in Welfare Economics,” Property, Freedom, Society (Auburn, Ala.: Mises Institute, 2016) pp. 301–08.