Austrian Welfare Economics?: A Misesian Response
Abstract
April 4, 2003
Several neo-Austrian economists have proposed various standards for a special welfare
economics to rival the welfare economics of the economics profession. This paper shows that
two of these (Tyler Cowen and Israel Kirzner) contain implicit and vague definitions of social
welfare that violate methodological individualistic principles. In addition, it argues that
neo-Austrians misconceive the new welfare economics largely because they misconceive the
role of equilibrium. Next, it shows that these neo-Austrian efforts are fundamentally different
from the recommendations of Ludwig von Mises. Mises advised that economists should
evaluate the logic of the arguments presented by others and should determine whether the
arguments are relevant to the goal that the others want to achieve. They should evaluate
arguments, not policies. Finally, it shows that Murray Rothbard's reconstruction of welfare
economics is not based on Mises or subjectivism and that to follow it would render the
fundamental task of economics, as Mises conceived it, impossible to carry out.
Austrian Welfare Economics?: A Misesian Response
Interest among neo-Austrian economists
in welfare economics is exemplified by four works:
Murray Rothbard's effort to reconstruct welfare economics (Rothbard 1956), Israel Kirzner's suggestion
that the norm of coordination be substituted for the Paretian-type norms of professional economics
(Kirzner 1992), Tyler Cowen's effort to construct a non-Paretian welfare function, which may be partly
based on Kirzner (Cowen 1991),
and Roy Cordato's "Austrian perspective" on welfare
economics.(Cordato 1992) Each of these, in one way or another, is critical of welfare economics as it has
developed in modern professional economics. Another common feature is their neglect of Ludwig von
Mises's position regarding the task of economics. Each of the above writers has sought a fundamental
means of rendering judgments about policies or institutions to rival the judgments or ideologies that they
perceive in non-Austrian professional economics. For Mises, however, economics is value free. It
employs logic and the principle of relevance to appraise the arguments of others.
The purpose of this paper is to use a Misesian perspective to assess three of these neo-Austrian
views -- Cowen, Kirzner, and Rothbard.
It will be argued that Kirzner and Cowen conceal what
professional economists would call social welfare judgments, albeit they are vague. In other words, the
authors do not succeed in presenting an alternative; they merely articulate vague goals like “innovation”
or “a steady improvement of mankind” instead of more distinct and specific judgments related to
individual wants and opportunities. The exercises are primarily a kind of verbal juggling. The reason may
be that the authors do not appreciate the true significance of the new welfare economics. In any case the
views stand in opposition to Mises’s value-free economics. Mises acknowledges that value judgments are
necessary to judge whether a policy is good or bad. To maintain the value-free status of economics, he
promotes an economics that evaluates not policies but arguments about the effects of policies. Thus he
advocates an economics that differs from the neo-Austrians described above.
Mises's distinct view of a value-free Austrian economics being concerned with evaluating
arguments about effects and not policy has been obscured by Rothbard, who promotes a rather different
welfare standard and who erroneously associates this standard with Mises. It is argued below that
Rothbard’s welfare standard is irrelevant to the issue of making or appraising arguments related to
market intervention. The paper also argues that Rothbard’s claim that his view is related to that of Mises
is based on a misrepresentation of Mises’s views.
The paper is organized as follows. Part 1 discusses Cowen's and Kirzner's proposed alternative
welfare standards. Part two reviews theoretical welfare economics with the intent of suggesting that some
of the neo-Austrian criticism of the new welfare economics is misconceived. Part three reviews Mises's
statements regarding the goal of economics. Part four discusses Rothbard’s contribution. Part five
presents a brief conclusion.
1. Examples of Neo-Austrian Welfare Economics
Cowen's Tripartite Standard
Cowen presents "a tripartite standard based upon discovery/innovation, complexity, and
provision of consumer goods."(Cowen 1991: 290-91) He describes each leg of his standard separately
and identifies what he calls its "desirable properties." Because he provides reasons why the properties are
desirable, we can surmise that his welfare criteria are more fundamental than he first suggests.
In the case of discovery/innovation, he gives three reasons for desirability: (1) that because
discovery is incentive-driven, we are led to focus on the incentive systems possessed by different
economies; (2) that empirical history indicates that public policies intended to increase resources devoted
to discovery beyond some optimal limit have failed; and (3) that balanced discovery is likely to lead to
the "healthiest all-round pattern of growth," which he equates to "long-run sustainability."(ibid.: 291-2)
Regarding the first reason, he does not tell us why he would like the analysis of policy to focus on
incentive systems.
Regarding the second reason, because of his use of the concept of an optimal limit of
resources devoted to discovery and his claim that public policies have failed, he implicitly asks the reader
to employ some unspecified social welfare criterion. What does he mean by "optimal," one must ask?
Since he is referring to public policy, it appears that he means "socially optimal." And how is one to
judge whether a policy has failed without a social welfare criterion? Regarding the third reason, he does
not tell his readers why he believes that a balanced, healthy, or sustainable all-round pattern of growth is
desirable. The growth of weeds is balanced, healthy, sustainable, and all-round. But is it desirable? It
appears that he has in mind the growth of "social welfare," although he does not say this.
His second property is complexity. He cites two reasons why this property is desirable. His first
begins with the following sentence: "Policies which increase complexity are identified with the
progressive development of an economic system, as empirical evidence indicates that successful
economies are generally characterized by an extraordinarily high degree of complexity."(ibid.: 293) The
criterion of desirability here seems to be "progressive development" and "success." He does not define
these terms. They suggest, as in the other cases, that he is implicitly asking the reader to employ some
other unspecified social welfare criterion. The second reason for desirability is "its close relationship
with innovation and discovery." As he sees it, complexity and innovation are mutually reinforcing. Since
innovation seems to imply some unspecified social welfare criterion, this reason seems to fit the pattern
of implying an unspecified social welfare criterion. The question is: how do we use this criterion to judge
a policy?
The third property is provision of consumer goods. Here he notes that "[s]ystems which do not
manage to procure consumer goods for citizens must be judged as failures, regardless of any other
virtues...(ibid.: 293) This suggests that his ultimate standard in all of the above cases is "a long-run
favorable impact upon the provision of consumer goods."(ibid.: 294) Now this seems similar to the
standard welfare criteria that employ the production possibilities apparatus. However, since Cowen does
not tell his readers how one is to determine what a consumer good is or why a system that does not
provide them must be judged as a failure, since he writes nothing about the distribution of such goods,
and since he does not explain how one might compare different collections of goods; the criterion is still
largely unspecified and therefore lacking usefulness as a criterion for judging policy.
The only sensible conclusion to reach about Cowen's proposed non-Paretian welfare economics
is that it ultimately, if partly implicitly, employs a notion of social welfare -- one that does not differ in
any obvious way from that used in modern welfare economics except perhaps by virtue of its lack of
clarity.
Kirzner on the Norm of Coordination
Kirzner (1992) recommends what he calls the "norm of coordination," which is based on his
concept of the entrepreneur. He associates entrepreneurship with the reduction of error.
Unlike Cowen, he does not in the article write or imply that the reduction of error,
entrepreneurship, or coordination, enhances "development," promotes "growth," makes the market
economy more "healthy," makes it more successful, or enables more consumer goods to be produced.
However, he does make the following rather colorful statement. He writes that his coordination norm
refers to the "process through which initially clashing, dis-coordinated activities are somehow being
hammered out in a manner such as to approach a more smoothly dovetailing pattern of activities."(1992:
191) He ends his paper with the conclusion that the recognition of genuine error can be seen as "opening
the door...for the possible reconstruction of normative economics along truly Austrian lines, i.e. in a
manner fully consistent with (a) subjectivism, (b) methodological individualism and (c) an emphasis on
dynamic processes."(ibid.: 192)
* Although the "door may be open," it is not evident to one who reads only this paper what the
inside holds in store. Why should one care about such things as the reduction of error or coordination?
The answer can be found in his earlier treatments of entrepreneurship and error, to which he refers
toward the end of the paper.(ibid.: 189) He “opens the door” to these earlier treatments with his reference
to chapter eight of his 1979 book. Consider the closing paragraph of that chapter:
Only an economics that recognizes how the profit motive -- by which we mean the lure of pure
entrepreneurial profits -- can harness entrepreneurial activity toward the systematic elimination of error can be
of service in pointing the way to those institutional structures necessary for the steady improvement of
mankind.(Kirzner 1979: 136, italics added)
In case there is any doubt what he means by the "steady improvement of mankind," consider his remarks
in the following chapter:
Since individuals obviously differ in their entrepreneurial alertness, it is clear that opportunities for social
improvement will tend to be exploited most fruitfully if institutional arrangements can be patterned so as to
translate such opportunities into opportunities that will be encountered by those whose entrepreneurial
alertness is the most acute, the most sensitive, and the most accurate."(ibid.: 149, italics added).
He does not define social improvement.
If Kirzner had gone on in his 1992 paper to repudiate his earlier statements about the
improvement of mankind and social improvement, we might have reason to believe that he had not fallen
into the same trap as Cowen. But he did not do this.
His use of these terms indicate that his conception
of the desirability of particular institutions is based on a similar non-specific notion of social welfare.
2. Welfare Economics and General Equilibrium
Some neo-Austrians appear to misunderstand the function of equilibrium in economics. They
begin by pointing out correctly that real market economy interaction cannot be captured in the
equilibrium model. The robots of the general equilibrium model are not actors. Their behavior is fully
specified, or programmed, by the economist. It follows, they reason correctly, that to use the
equilibrium model to represent a real market economy is an error. It is likewise erroneous to base
predictions about the effects of a public policy on comparative static analysis.
However, they go on
to reason that because modern welfare economics employs the general equilibrium model, it must be
misconceived. They do not consider the possibility that modern welfare economics (or at least some
modern welfare economists) uses the model for a different purpose than to predict the effects of
public policy. To see why such reasoning may be wrong, a review is necessary.
A Review of Modern Welfare Economics
There seems to be wide agreement that welfare economics began with the publication of A.C.
Pigou's The Economics of Welfare (1920).
In opposition to Pigou's assumption that individual
utilities can be compared, Nicolas Kaldor proposed the Pareto optimum principle as a more
acceptable means of judging whether a public policy is good.(Hicks 1975: 309) This was followed by
discussions of possible schemes for compensating assumed losers. After this, the "old welfare
economics" ended for all intents and purposes. It was replaced by the "new" welfare economics
(NWE), championed by Abram Bergson and Paul Samuelson.
These writers and others produced "a
systematic way of introducing from outside of economics various ethical norms (as embodied in what
is called a social welfare function)."
The original social welfare function was designed to allow an
individual to use his personally devised procedure for selecting among a number of alternative
positions that individuals themselves were assumed to regard as yielding maximum satisfaction (or
highest levels of indifference).(Samuelson 1956: 22)
Neo-Austrian economists generally believe that NWE assumes a hypothetical general
competitive equilibrium. In order to help us understand whether this is true and its relevance, we
must consider the details.
The Production Possibilities Frontier
NWE begins by conceiving of the technical possibilities for producing goods independently
of the satisfactions that individuals are assumed to receive from the goods. In doing so, it presumes to
know which items are perceived as goods. The making of such an assumption immediately restricts
the policy relevance of any theorems that NWE derives. Policy relevance is further restricted in the
same way by the assumptions that it knows (1) which items and behaviors are perceived as factors of
production and (2) the perceived technical possibilities of production. Finally relevance is restricted
by the assumption that the goods are produced in the most technically efficient ways.
NWE defines the “highest amounts of goods” in terms of a production possibilities frontier.
In a three-good case, for example, the frontier shows the greatest amount of a that can be produced,
given that specific amounts of b and c are already being produced in their most technically efficient
ways. To find the combinations of goods that lie on the production possibilities frontier, one can,
roughly speaking, use the following procedure. Begin with a given quantity of one of the goods, say
a. Then consider all of the combinations of b and c that can be produced with the remaining factors of
production, given that the factors are used in the most technically efficient ways. Next, select a
second amount of a and repeat the procedure. Continue to choose different amounts of a until all of
the possible quantities of a have been considered. Repeat the procedure for b and c.
This procedure is often represented by the deceptively simple notion that we should imagine
a collection of produced goods such that it is impossible to produce one more unit of one of the goods
without giving up a unit of some other good. What is important to realize, however, is that, in a case
of n goods, the combinations on the production possibilities frontier is extremely large, even to the
point of being unfathomable. Mathematicians pretend to make this number fathomable by
substituting simple expressions or using models with only two goods. However, no competent
mathematician is fooled by the simple expressions into thinking that there is a set of combinations
that is mentally comprehensible in the sense of a real human being able to conceive of all of the
separate combinations of different goods that lie on the production possibilities frontier in a real,
modern economy. It is comprehensible only within the realm of pure mathematics.
F.A. Hayek (1945) has pointed out that no single individual could possess all the knowledge
needed by the numerous market participants, at all levels of the structure of production, to produce
the goods efficiently and in the efficient quantities. It would seem to follow that the production
possibilities frontier is an imaginary construction that could have no counterpart in reality. This is
true enough. The point to be made here, however, is that no one who reflects on the issue, including
the welfare economists, would think otherwise, although they might not have the same reservations as
Hayek about central planning.
The Utilities Possibilities Frontier
Starting with this incomprehensible combination of technically-efficiently-produced goods,
NWE assumes that the individuals have preferences for each separate combination of the goods.
Given these preferences, and for each of the unfathomable number of combinations of goods, we are
asked to conceive of all of the possible ownership possibilities. Let us take an example of three goods
that we shall assume lies on the production possibilities frontier: 10 units of a, 39 units of b, and 62
units of c. Further assume that there are three individuals: Mr. Chen, Mr. Lee, and Mrs. Wang. We
might begin by imagining that Chen owns one unit of a, 35 units of b, and 21 units of c. The rest of
the a, b, and c are distributed in some way between Lee and Wang. We are asked to imagine that the
three people engage in costless bargaining and that they reach a distribution of the three goods where
it is not possible for any redistribution to benefit one person without harming the others. There are
many possible distributions that satisfy this condition since the exact final distribution of goods that
would result from such bargaining is not completely determinate.
However, the possibilities are
limited. For example, Chen would never agree to a series of bargains that leaves him with, say, the
same amount of a and b, but less c than he started with. The possible distributions that could result
from bargaining are described by saying that they “lie on the contract frontier.” Note that for each
initial distribution, there is a very large number of combinations that lie on the contract frontier. Now
we must derive a contract frontier for all of the other possible initial distributions of a, b, and c.
Of course, the real world contains many individuals and many goods. Thus if we tried to
construct a contract frontier for all of the possible initial distributions of technically-efficiently-produced goods in the real world, we would have to build an image of an unfathomable number of
combinations.
Once this unfathomable number of combinations is imagined, one can imagine narrowing the
number down since some of the combinations would be inferior on the scales of preference of
everyone to other combinations. The number of combinations that can be omitted is...well,
unfathomable. The number of combinations that remain is less unfathomable than the original
number. However, since an unfathomably unfathomable number of combinations had to be initially
derived before it could be narrowed, the problem of computation is no less severe. The final,
narrowed-down set of combinations of combinations is what NWE calls a utilities possibilities
frontier.
Relevance
Clearly the idea of constructing a utilities possibilities frontier takes us far away from the
realm of reality. No ordinary human being -- not even a mathematical economist -- could think that in
building such an image, she is building a model of reality. What, then, is she doing? One answer is
that she is trying to inform us about using individual preferences to make judgments about the
outcome of production and exchange activities in a market economy. Even if we did not need to
worry about how goods get produced -- i.e., even if we only had to worry about distributing goods
that were already produced according to the best known (from the standpoint of omniscience)
methods of production -- we could never figure out a final distribution that would be “best.” The best
we could do would be to use our mathematical symbols to represent an unfathomably unfathomable
number of possible best combinations.
It is true that some of the new welfare economists casually assert that we could tell which is
best if we had a social welfare function. Perhaps for mathematicians who enjoy representing
unfathomable possibilities with simple expressions, this assertion is not uninteresting. However, for
anyone who understands the entire reasoning of modern welfare economics, the assertion is clearly
irrelevant to policy. The term “social welfare function” must, at best, be regarded as the embodiment
of the notion that an individualist who claims that some public policy is “good” must base his claim
on an image that is unfathomable.
Beyond this, one might ask whether it is necessary to build a mathematical model of such
complexity just to support the conclusion that it is impossible to make public policy judgments on the
basis of methodological individualism. If there had been no prior history of welfare economics, the
answer would seem to be no. However, the Marshall-Pigou tradition of making judgments about
public policy on the basis of partial equilibrium analysis naturally attracted the mathematical
wizardry of economists like Samuelson, who build a large part of his reputation by mathematically
generalizing various partial equilibrium notions that were in vogue among his contemporaries.
This
should be a useful point to recognize for Austrians, since their tradition of tracing the effects of a
policy, step by step, by means of the deceptively simple concept of opportunity cost, implies a level
of complexity that is substantially greater than that of the unfathomably unfathomable mathematical
models.
The Neo-Austrian Critique is Misplaced
The Neo-Austrian view that there is a major flaw in modern welfare economics seems to be
based on the fact that the general competitive equilibrium model has been used in applied policy
analyses as a model of an economy. The policy analyst compares an image of a general competitive
equilibrium in the absence of the policy with a general competitive equilibrium in its presence. He
describes the effects of the policy in terms of the difference. However, the model was not intended
for this use. To the masters in this field its purpose was to show that even if one knows the exact
effects of a particular policy, he could not evaluate it without making a welfare judgment (i.e.,
without introducing a social welfare function). This conclusion is not sensitive to the fact that the
NWE imaginary construction of an economy is unrealistic.
A virtue of the NWE imaginary construction is that it represents a high degree of complexity
while at the same time being built on the assumption of methodological individualism. Of course, the
individuals in the model are not human actors. They are automatons whose behavior or behavioral
algorithms are specified in the model. However, there are many of them and they each have roles as
consumers and factor-suppliers or producers. Thus, the model enables us to assess the practicality of
using methodological individualism as a basis for using welfare judgments to judge the effects of a
public policy
Social Welfare Judgments and Neo-Austrian Economics
Cowen and Kirzner use a different procedure for making judgments than that of NWE.
However, since they want to evaluate the effects of public policy and since they apparently want to
employ methodological individualism, they cannot avoid the difficulty of introducing social welfare
judgments from the outside. Unfortunately, (1) by failing to point out that, in all real cases of public
policy, some individuals would gain and others would lose according to their own standards and (2)
by failing to appreciate the complexity of production and distribution; they give the impression to
mainstream economists that Austrian economics is naive and even ideological. The latter impression
– that Austrian economics is ideological -- is directly opposite the view of Mises, to whose ideas this
paper now turns.
3. Mises on the Role of the Economist
The various efforts of neo-Austrian economists to devise an Austrian welfare economics
contrast sharply with the view of Mises. In Mises's view, the aim of economics is to evaluate the logic
and relevance of policy proposals made by individuals who are concerned with the satisfaction of
material wants. In other words, it is to address the arguments of others rather than to make arguments
oneself. It is worth tracing out Mises's reasoning.
"The main objective of praxeology and economics," Mises declared, "is to substitute
consistent correct ideologies for the contradictory tenets of popular eclecticism."(Mises 1966: 185)
This does not mean that the goal of economics is to build its own ideology. According to Mises,
economics is a set of logical deductions. The deductions are about "what is," given that economists
only build images that are relevant to the social facts one aims to describe. Ideology goes beyond the
study of things as they are. It is "about the ought, i.e., about the ultimate ends which man should aim
at in his earthly concerns."(ibid.: 178) Ideology can be distinguished from the broader concept of a
world view, which is
an interpretation of all things, and as a precept for action, an opinion concerning the best means for
removing uneasiness as much as possible...Religion, metaphysics, and philosophy aim at providing a
world view. They interpret the universe and they advise men how to act."(ibid.: 178)
Ideologies may be of many types. Mises distinguished two classes: (1) asceticism and (2)
non-ascetic ideologies. Economics is irrelevant to asceticism. It is relevant only to non-ascetic
ideologies since these "must pay heed to the fact that society is the great means for the attainment of
earthly ends."(ibid.: 179, italics added) The term "society," or social cooperation, has a special
meaning. It refers to individuals' recognition that they can gain through specialization and the
division of labor.
In effect, Mises asserts that anyone who aims, through giving advice on public
policy, to attain earthly ends must recognize the gains of specialization and the division of labor.
Because the other ideologies are concerned with "earthly ends;" if we want to evaluate them
from a logical point of view, we must use economics (ibid.: 182-4). "It is the task of economics to
expose erroneous doctrines in the field of social action."(ibid.: 93)
Thus, Mises is writing that the economist's task is to evaluate the arguments put forth by
those who we might call non-ascetic ideologists, i.e., by the ideologists who are concerned with the
attainment of earthly ends. For these, "no appeal to any religious or metaphysical dogmas and creeds
can invalidate the theorems and theories concerning social cooperation as developed by logically
correct praxeological reasoning."(ibid.: 180)
How does the argument that economics is value free square with the popular view that Mises
was a indefatigable advocate of laissez faire? In discussing the procedure of building an image of the
pure market economy, Mises writes
It is true that economists have drawn from their investigations the conclusion that the goals
which most people, practically even all people, are intent on attaining by toiling and working and by
economic policy can best be realized where the free market system is not impeded by government decrees.
But this is not a preconceived judgment stemming from an insufficient occupation with the operation of
government interference with business. It is, on the contrary, the result of a careful unbiased scrutiny of all
aspects of intervention.(ibid.: 238)
In other words, Mises argues that by analyzing arguments for intervention, the economist reaches the
conclusion that “most people, practically even all people,” will best realize their goals in market
systems where there is no government intervention. The economist's conclusions about the market
economy stem not from a value judgment but from the careful application of economic theory to the
study of intervention, the evaluation of arguments favoring intervention, and an understanding of the
effects that individuals expect intervention to achieve.
Although it may appear different to the uninitiated reader of Human Action, Mises for the
most part stuck to a policy of evaluating the arguments of others rather than presenting specific
arguments himself.
After presenting the imaginary construction of the market economy, he
discussed problems suggested by non-ascetic ideologists that were either evident in the literature or
that might be thought up in order to try to support a particular intervention in the market economy. To
enable his book to have the greatest effect along these lines, he could have organized it in a different
way. After presenting the theory of the pure market economy and the theory of money and credit, he
could have systematically dealt with a series of specific arguments. He could have begun each
subsequent section by stating a specific policy argument and then evaluating it on the basis of logic
and relevance. Although there are some chapters in which he takes this approach,
he is not
sufficiently systematic for the approach to stand out.
Have any noted economists followed Mises? The closest example this author could find was
Lionel Robbins. In Robbins's view, the significance of economic science (i.e., of the logical
propositions of the science) is that it "enables us to choose with full awareness of the implications of
what we are choosing."(Robbins 1984: 152) Because individuals do not ordinarily have full
awareness, Robbins argued, the economist adopts the role of an educator. Note the similarity between
(1) Mises's view that economists should try to show whether the policies proposed by intellectuals
who follow particular ideologies can achieve the goals implicit in those ideologies and (2) Robbins's
view that economists should try to teach the effects of policies to individuals who otherwise would
not be aware of those effects. The difference between the two is subtle. Whereas Mises targets the
arguments of ideologists that favor or oppose particular policies, Robbins targets prospective
unwitting beneficiaries and losers of policies who may have never formulated an argument.
Logic, Relevance, and Value Freedom
To claim that economics is value free in its dealing with policy arguments does not mean that
economists can avoid making judgments. They must indeed decide which images of economic
interaction are relevant in determining whether a policy will achieve its intended goal. Pure
praxeological constructions are merely exercises in logic. They are independent of any particular
manifestation in everyday life. However, when we apply these purely logical constructions to
describe real economic interaction, we must make additional assumptions. Mises points out in his
section on the "Procedure of Economics" that "the end of science is to know reality." "Therefore,
praxeology restricts its inquiries to the study of acting under those conditions and presuppositions
which are given in reality."(ibid.: 65) As a result, in the presentation of its results, economics adopts
"a form in which aprioristic theory and the interpretation of historical phenomena are
intertwined."(ibid.: 66)
Now a policy argument requires a description or prediction of what will happen if a
particular set of conditions prevail. The economist must build images of economic interaction under
the assumptions (1) that the policy is adopted and (2) that it is not adopted. It follows that to use
economics to analyze an argument about policy, one must make judgments about which images are
relevant. To do this, he must make judgments about the particular conditions under which individuals
act in reality. But these judgments are of an entirely different nature than the judgments described in
welfare economics.
Mises and Welfare Economics
In Chapter 35 of Human Action, Mises writes that welfare economics amounts to an implicit
denial of economics, as it was handed down by the old liberals and modified by the best literary
neoclassical economists.
He begins with a statement that, in essence, tells how the doctrines of the
socialist and interventionist propagandists have failed due to their bad economics. Rather than admit
the failures, however, the propagandists have introduced the “welfare principle.”
Propagandists must finally admit that the market economy....[f]rom day to day....increases the quantity
and improves the quality of products. It has brought about unprecedented wealth....But, objects the
champion of interventionism, it is deficient from what he calls the social point of view. It has not wiped
out poverty and destitution(Mises 1966, p. 833)
Instead of the allowing decisions about how to act to be based on the “profit principle” (i.e., to be
made by entrepreneurship) these propagandists want them to be made on the basis of the "welfare
principle." However, if we try to add content to this welfare principle by employing some concept of
utility that focuses on material well being, we must acknowledge the deductions made by economists
about the market economy's role in promoting that well being. In the market economy, each person
must "adjust his conduct to the requirements of social cooperation and look upon his fellow men’s
success as an indispensable condition of his own. From this point of view one may describe the
objective of social cooperation as the realization of the greatest happiness of the greatest number." In
other words, we already know that the “immense majority of nonascetic people,” will better satisfy
their wants by means of the market economy than by means of some alternative system. Thus, the
next step is to study the profit principle under the conditions of the market economy.
Mises discusses what we have in this paper called applied policy analysis using the welfare
economics terminology. He writes:
The plight of Western civilization consists precisely in the fact that serious people can resort to
such syllogistic artifices without encountering sharp rebuke. There are only two explanations open. Either
these self-styled welfare economists are themselves not aware of the logical inadmissibility of their
reasoning; or they have chosen this mode of arguing purposely in order to find shelter for their fallacies
behind a work which is intended beforehand to disarm all opponents. In each case their own acts condemn
them.
There is no need to add anything to the disquisitions of the preceding chapters concerning the
effects of all varieties of interventionism. The ponderous volumes of welfare economics have not brought
forth any arguments that could invalidate our conclusions. The only task that remains is to examine the
critical part of the welfare propagandists’ work, their indictment of the market economy.
All this passionate talk of the welfare school ultimately boils down to three points. Capitalism is
bad, they say because there is poverty, inequality of incomes and wealth, and insecurity(ibid., p. 833-835,
emphasis added).
In short, Mises dismisses welfare economics based on utilities of individuals as either (1) an exercise
in which the welfare economists are unaware of the mistakes in their reasoning or (2) an effort to
mask their interventionist propagandizing. Referring to the first possibility, the error is due to their
implicit denial that there are mutual benefits of exchange and the market system. To ask what
economic policy a government should adopt by employing the framework of welfare economics is
like asking what policy a lifeguard should adopt to help an olympic swimmer avoid drowning, given
the assumption that the swimmer cannot swim.
We can summarize Mises's view as follows. Mises asserted that every socialist,
interventionist, and classical liberal aimed to assure the satisfaction of material wants (or earthly
ends) and to achieve progress (as they defined it). He argued further that interaction under the
conditions of the market economy accomplishes these goals. For this reason, the study of social
cooperation in the market economy is the starting point of economics.
In his characterization of welfare economics and the "welfare principle," Mises's target is not
NWE but applied policy analysis that employs similar terminology. He says that the interventionist
who employs a broad social welfare function while disregarding the benefits of laissez faire to the
people does not comprehend the teachings of economics about the market economy. From the
viewpoint of Mises’s praxeological economics, the applied policy analysis of some welfare
economists is likely to depreciate the traditional teachings of economics, which began by recognizing
the benefits of a division of labor in the market economy.
4. Rothbard’s Welfare Economics
Rothbard showed an appreciation for developments in the new welfare economics that was
absent in the papers by Cowen and Kirzner. Giving due credit to NWE, he acknowledged that
methodological individualism could not realistically be used to develop a welfare standard which
could be applied in policy analysis. Not satisfied with this, however, he sought to “reconstruct”
welfare economics on the basis of a welfare standard of his own making.(Rothbard 1956: 248-9) The
discussion in this section aims to show not only that his new standard is incapable of being used to
evaluate policy but also that his paper contained serious misinterpretations of Mises. It also describes
works of other neo-Austrian economists, which have falsely associated Mises with Rothbard’s effort,
thereby disregarding or diminishing Mises’s value-free economics.
Rothbard’s “Reconstruction”
Rothbard's “reconstruction of utility and welfare economics” is based on the premise that
economists should assume that whatever they observe people doing reflects their preferences, one-to-one. To Rothbard, this means that when they observe coercion, they must deduce that it cannot be
beneficial, since coercion must be presumed to harm someone. When they observe exchange, they
must deduce that it is beneficial because the parties to the exchange are benefitted.(Rothbard 1976:
252) Rothbard extends this to claim that when economists observe people participating in the free
market; they must deduce that the people have benefitted from their participation.(ibid.: 250) He
writes “(1) that the free market always increases social utility; and (2) that no act of government can
ever increase social utility."(ibid.: 252-3)
At first glance, Rothbard’s welfare standard may seem rather simple and unobjectionable.
However, closer examination of the nature of a modern market economy reveals that it is also
irrelevant. There are three characteristics of a modern market economy that Rothbard seems to
disregard. First, without the coercion of government, there could be no market economy in the
modern sense. Second, private property rights can never be complete in a modern market economy.
Although individuals may prefer a system with complete private property rights (and the uses of
coercion it implies), their recognition of the costs or impossibility of producing such a system may
lead them to favor a system with incomplete property rights. Third, the creation of new property
rights -- which requires the use of coercion in new ways -- is a characteristic of any realistic image of
the market economy. In other words, to maintain a market economy in the light of economic change
requires a continuing alteration in the uses of government coercion.(Gunning 2000b)
Given these facts, one who aims to evaluate a policy under realistic conditions according to
any welfare standard must build images of interaction in which private property rights are not, or
cannot be, enforced or in which they are too costly to be defined. To invoke a welfare standard that
rules out coercion is to preclude the evaluation of policy altogether, at least in the world as we know
it. The following discussion elaborates on this point.
Coercion is a Necessary Prerequisite for a Modern Market Economy
One of the conditions, or prerequisites, of the market economy is private property rights.
Based on experience and perhaps knowledge of human nature, we are able to reason that a modern
market economy requires a monopoly over superior force to insure the enforcement of private
property rights. The costs of enforcing such rights may be partly borne by private parties. However, if
the only enforcement was private, the prospects of both being a victim and a successful perpetrator of
theft, robbery and extortion would, in turn, provide incentives for those who did not possess superior
force to employ resources to produce it. The ensuing military competition would not only be
wasteful, it would reduce the prospective benefits of producing durable goods and resources, thereby
limiting material wealth and progress. A monopoly over superior force could deter this. This
monopoly is ordinarily what we have in mind when we refer to a government, or state. We assume
that this entity possesses such power. Yet Rothbard conceives of a welfare function that would rule
out government coercion. It follows that Rothbard’s image of a market economy without the
accompanying use of government coercion is unrealistic.
Incomplete Property Rights
Let us suppose that the enforcement of property rights somehow occurs without the use of
government coercion or that the use of coercion for this purpose alone is somehow regarded as
acceptable according to Rothbard’s standard. Then there is a second problem due to the impossibility
of completely enforcing private property rights. Experience and perhaps intuition inform us that to
define and enforce all private property rights at a minimal cost is impossible. It is not economically
feasible to enforce private property rights in all goods and factors. The operational question is which
property rights should be established and enforced. Rothbard’s exchange-coercion welfare standard
renders such a question irrelevant. Consider a common property resource, like clean air. The fact that
people live in an unhealthy, smoggy environment, by Rothbard's measure, would demonstrate that
they did not want a rearrangement of property rights that would enable them to live in a healthy,
cleaner environment. Realistically, however, clean air is a common property resource for which
private property rights are either too costly or impossible to fully define and enforce. The important
policy question is whether coercion should be used to try to enforce property rights in this resource.
Yet according to Rothbard’s standard, we would have to accept whatever private enforcement
happens to emerge and to reject any alternative enforcement.
Rothbard’s Response: A Misinterpretation of Mises
Since Rothbard actually deals with an issue of this type and since he refers to Mises in doing
so, one is obliged to analyze Rothbard’s treatment. We shall see that he misinterprets or
misrepresents Mises’s discussion.
Both Rothbard and Mises discuss the common property resource under the “external
diseconomy” rubric. First consider Mises. Writing mainly about the case of pollution, Mises notes
two causes of this external diseconomy: (1) lawmakers’ intentions to grant special exemptions
(presumably from common law) to polluting firms or (2) an unintentional effect of the traditional
wording of laws.(Mises 1966: 655-6) Mises goes on to say that when individuals and firms perceive
external costs,
the economic calculation established by them is manifestly defective and their results deceptive. But this
is not the outcome of alleged deficiencies inherent in the system of private ownership of the means of
production. It is on the contrary a consequence of loopholes left in this system. It could be removed by a
reform of the laws concerning liability for damages inflicted and by rescinding the institutional barriers
preventing the full operation of private ownership.(ibid.: 658)
Rothbard transforms this into the following: “These ‘problems’ [noise, smoke, fishing, etc.]
are due to insufficient defense of private property against invasion. Rather than being the result of
loopholes, they become, in Rothbard's discussion, the results of invasions of property. Moreover,
these invasions are ruled out of the free market by definition.”(Rothbard 1956: 259n)
The first point to note is that Rothbard inaccurately reports Mises’s views. He is correct in
implying that the problems are not a defect of the free market. They are at best a characteristic of the
free market under certain conditions. Mises cites one condition -- the traditional wording of laws.
However, it is incorrect to say that Mises regarded such actions as invasions which are ruled out of
the free market by definition. It is true that they are ruled out by the definition of the pure,
unhampered market economy. In this imaginary construct, there is a complete set of rights to control
all actions that have externally harmful effects.
However, when Mises turned to the issue of the
common property resource, he did not assume a pure market economy. It is clear from his discussion
of the history of such situations that he envisioned the possibility that individuals would reach a stage
when they considered the harmful actions taken under the conditions of a common property resource
wasteful.(ibid.: 657) From the standpoint of the definition of the pure market economy, external
diseconomies may be treated as invasions of property; but from the standpoint of real people
operating in real market economies where laws are “deficient,” -- that is, in the environments where
free markets really exist -- external diseconomies are properly considered, according to Mises, either
as the consequence of special privilege or of the traditional wording of laws.
It should be evident that whereas Rothbard takes an ethical approach to the problems
associated with external diseconomy, Mises takes a non-ethical, realistic approach -- i.e., an approach
that recognizes the reality of an incomplete property system. Since Rothbard was at best misleading
about Mises's approach, we are led to ask whether there is any merit to Rothbard’s position,
independently of his claim to be following Mises. It seems evident that there is not. To say that the
common property resource situation is always an invasion of private property rights is to assume that
private property rights in all resources already exist. For example, in the case of pollution, Rothbard
must be assuming that, prior to the emergence of the external diseconomy, the rights to avail oneself
of clean air are already established. No new delineation of rights is necessary. On what grounds can
Rothbard make this assumption? Certainly not on the grounds of reality.
Creation of New Property Rights
Even if property rights at some particular time were complete, it is the nature of a market
economy for individuals to want to avail themselves of the potential gains associated with creating
new rights. Every contract entails the creation of new rights. Such new rights must often be enforced
by the government monopoly over coercion. Even if these new rights need no government to enforce
them, someone may claim that they interfere with existing rights and, consequently, that they violate
those rights. Who is to adjudicate such claims? And who is to enforce the judgments? It is difficult to
imagine adjudication and enforcement of such new rights without a government. And it is difficulty
to imagine a progressing market economy without a continuing process of rights creation,
adjudication, of claims, and government enforcement.
Mises on Demonstrated Preference: Another Rothbard Misinterpretation
Rothbard began his 1956 paper with the mistaken assertion that his notion of demonstrated
preference was derived from Mises. The assertion was based on a misconception of Mises's views on
the scale of values. To understand this error, consider Rothbard’s two examples of demonstrated
preference. (1) "[I]f a man chooses to spend an hour at a concert rather than a movie, we deduce that
the former was preferred, or ranked higher on his value scale."(Rothbard 1956: 225) (2) "[I]f a man
spends five dollars on a shirt we deduce that he preferred purchasing the shirt to any other use he
could have found for the money."(ibid.) These two examples are fundamentally different. In the first,
the reader is implicitly asked to put himself in the shoes of the chooser. If you choose, Rothbard
implies, you must demonstrate your preference for the chosen alternative. In the second, the reader is
implicitly asked to observe behavior. If you observe someone’s behavior, Rothbard asserts, you must
agree that he had a preference to perform it instead of some alternative. A moment’s reflection
indicates that while making a choice necessarily implies preference, the existence of observable
behavior does not necessarily imply either choice or preference. A person could behave in a
particular way on spontaneous impulse, as a result of operant conditioning, through reflex, or out of
habit. Choice implies preference in the sense that the preference is a causal factor in the choice. In the
first example, the assumption that the individual chooses implies preference. In the second example,
behavior may or may not imply preference. To support a claim that it does, one must judge that the
behavior was chosen.
Mises surely did not believe that one can deduce preference from behavior. His entire logic
of action and economics was built on the a priori assumption that all behavior is chosen and therefore
preferred. He did not suggest that economists study or observe behavior and “deduce” that the
behavior was preferred. He confined economics to the study of action, which he assumed to be
always chosen.
The praxeological economist assumes all behavior to be chosen. It is true that
economics entails the application of praxeology to real situations. Serious economics is not simply
the construction of logical images. The images must be relevant to everyday life.(Mises 1966: 65)
Accordingly, to evaluate an argument for or against market intervention, the economist must include
behavior that is not chosen. However, he does not regard the existence of such behavior as
demonstrating that the behaver has specific preferences to perform it.
Rothbard seems to have based his interpretation on some passages in Human Action where
Mises dealt with ideas of rationality and consistency of preferences. Mises was considering the
argument that if a is preferred to b and b is preferred to c, then logically a must be preferred to c. In
his section entitled “The Temporal Relations Between Actions,” Mises pointed out that this argument
is wrong because “two acts of an individual can never be synchronous.”(ibid.: Mises 1966: 103) In
other words, this argument is logically correct but praxeologically irrelevant.
Rothbard interpreted Mises's statements about preference as support for his notion of
demonstrated preference.(Rothbard 1956: 229) However, Mises was not writing about demonstrated
preference; he was merely pointing out the error involved in not accounting for the fact that action
takes time and that two actions, or choices, cannot occur at the same time. Thus an individual could
never, in the same action, face a choice between a and b and also between b and c. The fact that one
of the choices must occur in an action at a later time means that one’s preference may change.
In spite of the clear difference between Rothbard and Mises, Rothbard concluded his
discussion by saying that the "clearest and most thorough formulation of the concept [of
demonstrated preference] has been in the works of Professor Mises."(Rothbard 1956: 229) This
assertion is not supported by Mises text. Perhaps because of Rothbard's status in neo-Austrian
economics, a number of neo-Austrian writers have been led to think that Rothbard’s work on welfare
economics is based on Mises, apparently without exploring the original Misesian text.
Regrettably,
this is not the only serious misinterpretation of others' work in Rothbard’s paper.
Herbener’s Defense
Herbener has recently presented a spirited effort to resurrect Rothbard’s reconstruction. One
of Herbener’s main arguments -- that policy judgments under NWE are based on the assumption of
general equilibrium (Herbener 1997: 93) -- has been discussed earlier in this paper. One can agree
with Herbener that the use of the general competitive equilibrium model to make policy predictions is
irrelevant because it disregards the market process. However, a critique on these grounds does not
invalidate its use to demonstrate the impracticality of methodological individualism as a basis for
making policy evaluations.
Herbener's second argument is that, under the assumption of private property rights and
freedom from coercion, each successive situation will represent an improvement corresponding to the
notion of Pareto superiority.(ibid.: 98) In other words, he asserts that every exchange represents an
improvement for the exchanging parties. What Herbener seems to regard as his telling defense of this
argument comes in a discussion of Cordato’s critique of Rothbard. He considers the case of a person
who, having demonstrated a preference for participating in the free market, comes to demonstrate
regret over the loss-making actions he performed in that context because he lacked entrepreneurial
insight.(ibid.: 101) Herbener concludes: “That such losses occur in no measure diminishes the
conclusion that the free market renders the greatest satisfaction of individual preferences compared to
any other institutional arrangement for social interactions.”(ibid.) To Herbener, the individual’s
behavior of participating in the free market demonstrates his preference, even though he now regrets
his actions performed in the free market context.
Although this argument may be sufficient to deal with Cordato’s criticism, it does not get to
the heart of the matter. There are two problems. First, a person may prefer a system that protects her
from the consequences of her errors. If her only options are to participate in a free market system or
not to have such a system, she would prefer to participate rather than to rely on self sufficiency.
However, she may prefer a system in which she has some measure of taxpayer-provided protection.
Second, the concept of a free market is not unambiguous. As we have noted, a complete property
system is not only unachievably but, due the enforcement costs, also likely to be undesirable to those
who participate in it. Because of this, the practicality, or relevance, of Herbener's argument, like
Rothbard's, is questionable. The fact that people participate in a market economy that has a particular
set of private property rights does not imply that they prefer that economy over one with a slightly
different set.
5. Conclusion
The task of the economist, as understood by Mises, is to determine (1) whether the logic of a
particular argument for some policy is correct and (2) whether the policy is relevant to the goal that
the policy proposer says he wants the policy to achieve. Such a task can be pursued from a
completely value free stance. However, the various welfare economics proposals of neo-Austrian
economics would demote this task. And for what purpose? We have seen that two proposals (Cowen
and Kirzner) contain implicit, though vague, references to social welfare. We have also seen that
Rothbard’s proposed welfare economics denies economists the capacity to evaluate policy arguments.
The conclusion is that the neo-Austrian efforts to develop an Austrian welfare economics seem to be
an unfortunate waste that cannot but reflect badly on those neo-Austrians who are genuinely involved
in the task of evaluating policy arguments. Moreover, neo-Austrian criticisms of the new welfare
economics are misplaced.
References
Bator, Francis M. (1957) "The Simple Analytics of Welfare Maximization.” American Economic
Review. 22-59
Bergson, Abram. (1938) "A Reformulation of Certain Aspects of Welfare Economics." Quarterly
Journal of Economics. 52: 310-34.
Boulding, Kenneth E. (1952) "Welfare Economics." In Bernard F. Haley (ed.) A Survey of
Contemporary Economics. Homewood, Illinois: Irwin.
Cordato, Roy E. (1992) Welfare Economics and Externalities in an Open Ended Universe: A Modern
Austrian Perspective. Boston: Kluwer Academic Publishers Group.
Cowen, Tyler. (1991) "What a Non-Paretian Welfare Economics Would Have to Look Like." In Don
Lavoie (ed.). Economics and Hermeneutics. London: Routledge.
Cowen, Tyler. (1994) "Austrian Welfare Economics." In Peter J. Boettke (ed.) The Elgar Companion
to Austrian Economics. Brookfield, Vermont: Edward Elgar Publishing Company.
de V. Graaff, J. (1957) Theoretical Welfare Economics. Cambridge University Press.
Gordon, David. (1993) “Toward a Deconstruction of Utility and Welfare Economics.” Review of
Austrian Economics. 6 (2): 99-112.
Gunning, J. Patrick (1989a). "Mises on the Evenly Rotating Economy." Review of Austrian
Economics. Vol. 3: 123-35.
Gunning, J. P (1989b) "Professor Caldwell on Ludwig von Mises's Methodology." Review of
Austrian Economics 3: 163-76.
Gunning, J. Patrick. (1991) "Praxeology, Economics, and Ethical Philosophy." In Ebeling, Richard
M. (ed.). Austrian Economics: Perspectives on the Past and Prospects for the Future. Hillsdale,
Michigan: Hillsdale College Press.
Gunning, J. Patrick.(2000a) "Roy Cordato's 'Austrian' Critique of Coase on Social Cost." Review of
Austrian Economics. 13 (2): 175-192.
Gunning, J. Patrick.(2000b) "The Property System in Austrian Economics." Review of Austrian
Economics. 13 (2): 209-220.
Gunning, J. Patrick. (2003a). "How to Be a Value-Free Advocate of Laissez Faire: Ludwig von
Mises's Solution." American Journal of Economics and Sociology. (Forthcoming)
Gunning, J. Patrick (2003b) "Did Mises Err? Was He a Utilitarian?: Reply to Block." American
Journal of Economics and Sociology. (Forthcoming)
Hayek, F. A. (1945) "The Use of Knowledge in Society." American Economic Review. 35 (4)
Hayek, F. A.(1952) The Counterrevolution in Science. Glencoe, Ill.: The Free Press.
Herbener, Jeffrey M. (1988) “Austrian Methodology: The Preferred Tax Type.” Review of Austrian
Economics. 2: 97-111.
Herbener, Jeffrey M. (1997) “The Pareto Rule and Welfare Economics.” Review of Austrian
Economics. 10 (1): 79-106.
Hicks, J. R. (1975) "The Scope and Status of Welfare Economics." Oxford Economic Papers. 307-26.
Kirzner, Israel. (1976) "Philosophical and Ethical Implications of Austrian Economics." In E. Dolan
(ed.) The Foundations of Modern Austrian Economics. Kansas City: Sheed and Ward.
Kirzner, I. (1979). Perception, Opportunity, and Profit. Chicago: University of Chicago Press.
Kirzner, Israel (1987). "Austrian School of Economics." in J. Eatwell, M. Milgate, and P. Newman
(ed.) The New Palgrave: a Dictionary of Economics. London: Macmillan.
Kirzner, Israel M. (1992) "Welfare Economics: A Modern Austrian Perspective" in Israel M. Kirzner
(ed.) The Meaning of the Market Process: Essays in the Development of Modern Austrian
Economics. London and New York: Routledge. 180-192.
Kirzner, Israel. (1994) "Value-Freedom." In Peter J. Boettke (ed). The Elgar Companion to Austrian
Economics. Brookfield, Vermont: Edward Elgar Publishing Company.
Menger, Carl (1981). Principles of Economics. Translated by James Dingwall and Bert Hoselitz. New
York: New York University Press. (Originally published in German in 1871)
Osterfeld, David. (1987) "'Social Utility' and Government Transfers of Wealth: an Austrian
Perspective."
Pigou, A. C. (1938) The Economics of Welfare. London: Macmillan (4th edition).
Prychitko, David. (1993) "Formalism in Austrian-School Welfare Economics: Another Pretense of
Knowledge." Critical Review. 7 (4): 567-592.
Robbins, Lionel. (1953) "Robertson on Utility and Scope." Economica: 99-111. Reprinted in Robbins
(1974) Economic Science and Political Economy: Selected Articles. Edited by Susan Howson.
London: Macmillan.
Robbins, Lionel. (1984) An Essay on the Nature and Significance of Economic Science. (Third
edition) London: Macmillan.
Rothbard, Murray N. (1956) "Toward a Reconstruction of Utility and Welfare Economics" In Mary
Sennholz (ed.) On Freedom and Free Enterprise. New Jersey: D. Van Nostrand. Reprinted in
Rothbard (1997) The Logic of Action I: Method Money, and the Austrian School: Cheltenham UK:
Edward Elgar.
Rothbard, Murray N. (1976) "Praxeology, Value Judgments, and Public Policy." In E. Dolan (ed.)
The Foundations of Modern Austrian Economics. Kansas City: Sheed and Ward.
Salerno, Joseph T. (1990). "Mises as Social Rationalist." Review of Austrian Economics. 4: 26-54.
Samuelson, Paul. (1947) Foundations of Economic Analysis. Cambridge: Harvard University Press.
Samuelson, Paul A. (1952) "Comment (on Welfare Economics)" In Bernard F. Haley (ed.) A Survey
of Contemporary Economics. Homewood, Illinois: Irwin.
Samuelson, Paul A. (1954) "The Pure Theory of Public Expenditure," Review of Economics and
Statistics. 36: 387-9.
Samuelson, Paul A. (1956) "Social Indifference Curves." Quarterly Journal of Economics. 70.
Schumpeter, J. A. (1961) The Theory of Economic Development. Cambridge: Harvard University
Press. Originally published in German in 1911.
von Mises, Ludwig (1966). Human Action: A Treatise on Economics. Chicago: Henry Regnery
Company.
Wieser, Friedrich von. (1927) Social Economics. New York: Adelphia Co. (first published in German
in 1914)
Gunning’s Address
J. Patrick Gunning
Professor of Economics/ College of Business
Feng Chia University
100 Wenhwa Rd, Taichung
Taiwan, R.O.C.
Please send feedback
Email: gunning@fcu.edu.tw
Go to Pat Gunning's Pages
Mirror Site of Pat Gunning's Pages