'
&
$
%
NEW PERSPECTIVES
ON
POLITICAL ECONOMY
Volume 4 Number 1
ARTICLES
2008
Colombatto
and Melnik:
Productive and Non-Productive Entrepreneurship and the
Interaction Between Founders and Funders
Beaulier
and Hall: Collective Consumption Externalities and Charitable Giving
Kvasnička: Rothbard’s Welfare Theory: A Critique
Herbener: In Defense of Rothbardian Welfare Economics
Mokoro: Book Review
Christopher Coyne: After War: The Political Economy of
Exporting Democracy
Editorial Board:
Josef Šíma Department of Institutional Economics,
University of Economics, Prague
Dan Šťastný Department of Economics, University of
Economics, Prague
Advisory Board:
Enrico Colombatto University of Turin
Richard Ebeling Foundation for Economic Education
Robert Higgs The Independent Review
Jesus Huerta de Soto University of Madrid
Jörg Guido Hülsmann Universite d’Anger
Language Editor: Shelly Gussis
ISSN: 1801-0938
ISSN 1801-0938
New Perspectives on Political Economy
Volume 4, Number 1, 2008, pp. 1 – 21
Productive and Non-Productive Entrepreneurship and the
Interaction Between Founders and Funders
Enrico Colombatto and Arie Melnik
JEL Classiﬁcation: L24, M13
Abstract: By focusing on the economics of productive and unproductive entrepreneurship
this paper explains venture capitalism in diﬀerent institutional contexts, deﬁned
in terms of property rights and regulation. It is shown that venture capitalism
is just one among the various possibilities of transforming good ideas into success
stories; and that tends to be discarded when transaction costs are high. Although
venture capitalism is likely to be positively correlated with growth, it is not a necessary
condition. Still, it can provide indications as for the nature of the prevailing
institutional framework and possibly signal unexploited opportunities.
Enrico Colombatto is Professor of Economics at the University of Turin (Italy), Director of ICER
(Turin) and Fellow at the Collegio Carlo Alberto, enrico.colombatto@unito.it.
Arie Melnik is Professor of Economics at the University of Haifa (Israel) and a Senior Fellow at ICER,
amelnik@econ.haifa.ac.il.
2 New Perspectives on Political Economy
1 A Tale of Two Entrepreneurs
This paper puts forward an explanation of the features and dynamics of venture capitalism
based on what is known as the Austrian approach to entrepreneurship. In
particular, sections 1 and 2 recall the concept of entrepreneurship and show its implications
for the traditional economics of venture capital. Section 3 introduces the
notion of simultaneous entrepreneurship and leads to the central part of the paper
(section 4), where the funder-founder relationship is reconsidered and a new and
broader vision of the economics of venture capital (VC) is suggested. It turns out
that VC is mutually rewarding for the investor and for the receiving ﬁrm only when
certain conditions apply. These ﬁndings are further clariﬁed in section 5, where the
interaction between venture capital and the institutional context is examined; and in
section 6, where some implications for growth are investigated.
1.1 On the Austrian Approach to Entrepreneurship
Following the Austrians, we deﬁne entrepreneurship as the propensity of each individual
to improve his wellbeing by exploiting his knowledge, resources, talents or
sheer luck so as to produce goods, services and ideas and transform them into profitable
opportunities.¹ Of course, individuals are not all equally endowed with entrepreneurial
talents. Moreover, such talents show up in diﬀerent ways as people
display diﬀerent propensities – say – to accept risk, to engage in one industry rather
than in another, to work alone or in teams. However, it is fair to assume that the distribution
of talents does not depend on race, geographic location or political regime.
The same also applies to entrepreneurship, which does not follow the institutional
framework, either – since the desire to improve one’s own wellbeing is always there.
Nonetheless, institutions and political regimes do aﬀect the rules of the game according
to which such eﬀorts take place and the chances to obtain success. That is,
individuals do react to the institutional environment where they operate. And ap¹
See for instance Boettke and Coyne (2003). According to this deﬁnition, consumers are not entrepreneurs.
They do use their knowledge and skills in order to enhance their well-being. But do
not seek to proﬁt from the sale of products or services. Honest politicians are not entrepreneurs
either, since they are supposed to make rules in the public interest, not in their own.
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 3
ply their entrepreneurial talents in diﬀerent ways and to diﬀerent degrees, following
their own inclinations, emotions, ideologies, expectations.
In many cases entrepreneurial commitments beneﬁt both the economic actor and
his counterparts. Obvious examples are product and process innovations, but also insights
in the realms of marketing and organization.² If so, overall economic growth
takes place. In other cases all the beneﬁts are internalized and thus restricted to
the actor. It might also happen that property rights are violated, e.g. by means of
rent-seeking or criminal behavior. Then, entrepreneurship actually produces unjustiﬁed
welfare losses for at least some economic actors. In fact, the dynamic eﬀects
of competition always lead to welfare gains for some (the successful producers and
the buyers) and to welfare losses for others (the losing producers). From an Austrian
standpoint, this process is ‘just’ when it complies with the freedom-to-choose principle
and does not violate freedom from coercion. As such, it can obviously generate
undesired negative indirect eﬀects (i.e. undesired by some). It is ‘unjustiﬁed’ otherwise.
Of course, fear of unjustiﬁed conﬁscation reduces the incentive to create wealth
and thus inhibits growth.
1.2 Baumol’s Productive and Unproductive Entrepreneurs
Consistent with this broad framework, some years ago Baumol (1990) relied on the
institutional tradition and identiﬁed two types of societies. One generates incentives
that induce entrepreneurs to be productive and use their abilities in order to ultimately
satisfy consumers’ wants. In this context innovative action enhances overall
economic growth. Instead, the alternative institutional framework encourages unproductive
behaviours, whereby entrepreneurs ﬁnd it rewarding to engage in rentseeking
or even in outright destructive operations (violence). When so, their skills
are directed to making a proﬁt (or avoiding a loss) without creating new wealth. In
fact, they often absorb resources and/or provoke ineﬃciencies.
² The pure imitation of a product or of a process is not enough to qualify a producer as an entrepreneur.
But if this producer manufactures a well known product in a way that others had not ﬁgured out or
made use of before, or ﬁnds new methods to reach demand, then he is indeed a productive entrepreneur.
However, to the extent he does so by preventing others from reaching potential buyers – e.g.
thanks to his eﬀorts to obtain regulation or tariﬀ barriers – he becomes an unproductive entrepreneur
(see sections 2 and 3 on the various categories of entrepreneurship).
4 New Perspectives on Political Economy
Of course, the distinction between productive entrepreneurs and rent-seekers was
already prominent in the literature on the economics of privileged interest groups,
which has indeed grown enormously since Gordon Tullock’s seminal work in the
late 1960s. However, Baumol’s contribution is important in that it posits that entrepreneurs
change the nature of their activity – productive or unproductive – from
one period to another, as a consequence of the reward structure in the economy. In
Baumol’s view this provides useful guidelines to analyze growth episodes in history.
For instance, low growth during the classical period was explained by its institutional
incentive structure, which encouraged entrepreneurs to concentrate on rent-seeking,
protect their privileges and disregard productive ventures. The opposite was true
during the Industrial Revolutions.
2 Baumol’s Founders and Traditional Venture Capitalists
By developing the concepts mentioned in the previous section, we take advantage
of Baumol’s distinction between productive and unproductive entrepreneurship, we
introduce the idea of simultaneous interaction between these forms of entrepreneurship
and then use it to shed light on the relations between two categories of entrepreneurs:
founders and funders. Founders engage in the production of goods and
services in order to make a proﬁt; while funders intervene by supplying ﬁnancial re-
sources.
In this paper the term ‘funders’ actually refers to a speciﬁc category of investors –
venture capitalists (VCs).³ If one accepts Baumol’s original framework with no further
qualiﬁcations the role of venture capitalists is straightforward. Within a productiveentrepreneurial
context, VCs acquire control in relatively small ﬁrms with a potential
for growth, and eventually end their participation by going public (IPOs), direct selling
or liquidating, depending on the success and features of the company. When
doing well, VCs’ proﬁts consist of the remuneration of their ability to spot a potential
winner and to take steps to transform it into a real victor.⁴ Instead, when the
³ VCs will be deﬁned in greater detail in section 4.
⁴ Tykvová (2007) provides an exhaustive survey of the existing literature on the theoretical questions relating
founders and funders, which are understandably concentrated on designing optimal contracts
within a dynamic principal-agent context.
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 5
rent-seeking scenario applies the name of the game is creating or protecting rents,
not ﬁnancing new ideas and economic activities; and traditional VC funders are often
out of the picture.⁵ Indeed, when young companies present a potential for high
proﬁts because of their promising rent-seeking prospects, their founders devote most
of their energies to maintaining and expanding their connections, rather than to hitting
upon new equity and thereby losing control. That explains why unproductive entrepreneurs
are more likely to raise funds by asking friendly bankers to supply credit
or by going public, than by opening to venture capitalists; indeed, in Western Europe
there are many examples of companies that enjoy state protection (e.g. trade barriers
or exclusive licenses, if not outright monopoly power), make fat proﬁts and succeed
in attracting both bankers and shareholders looking for limited exposure to competitive
pressure and prospects of safe dividends.⁶ In fact, when founders are ready to
accept new equity from dynamic and watchful investors the potential has often been
exhausted, and traditional VCs are no longer interested.
3 From Sequential to Simultaneous Entrepreneurship
In this paper we extend the actions of the entrepreneur from Baumol’s sequential
rotation of eﬀorts to a simultaneous one. A producer can do his best to generate
output (through acts of productive entrepreneurship), but can also seek rents and
privileges at the unjustiﬁed expense of other people’s welfare (through destructive
entrepreneurship); or strive to protect his wealth from aggression (through defensive
entrepreneurship). All at the same time.⁷ Put diﬀerently, simultaneous entrepreneurship
means that founders often ﬁnd it rewarding to engage in unproductive ac⁵
A non traditional VC would be one who engages in the political market. That is unlikely to happen,
though: Rent-seeking is frequently labour intensive, it often requires a long-term vision (developing
personal contacts with bureaucrats and policy-makers), is not transparent. That explains why VC funders
are interested neither in direct rent-seeking (if anything, their job would be funding rent-seekers,
not carrying out rent-seeking themselves), nor in supporting rent-seekers (too many intangibles, exceedingly
high monitoring costs, lack of transparency).
⁶ See Berger and Udell (2002) on small ﬁrms and ‘relationship lending’.
⁷ Following the deﬁnition proposed in section 1, the notion of productive entrepreneurship also includes
the activities of those who try to destroy rent-seeking situations, such as tariﬀ barriers or
normative constraints to the freedom of contract. On the other hand, lobbying to introduce tariﬀ
barriers is an act of destructive entrepreneurship, whereas lobbying to avoid legislation to that eﬀect
is here considered an act of defensive entrepreneurship.
6 New Perspectives on Political Economy
tivities as long as they succeed in acquiring or maintaining some privileges (e.g. as a
consequence of special personal relations with the authorities) so as to monitor and
possibly aﬀect the design of future regulation, reduce the bureaucratic entry costs
they are facing in some industries or geographical areas, increase barriers to entry for
potential competitors.
By means of these two notions – destructive and defensive entrepreneurship –
we also believe we clarify an intrinsic ambiguity that characterizes Baumol’s original
deﬁnition, in that his term ‘unproductive’ is limited to what we here call ‘destructive’
and omits to consider defensive activities, which are in fact far from negligible
(Tullock 1993). In particular, defensive activities cannot be captured by Baumol’s productive/unproductive
distinction for they absorb resources which could otherwise be
devoted to productive purposes (hence they are unproductive), but at the same time
happen to avoid greater destructions of wealth (hence they are also productive).
The reward structure of the economy – institutions – determines how proﬁtseeking
eﬀorts are allocated and which company features are conducive to better results.
For instance, one may expect that productive entrepreneurship be intense when
institutions encourage and protect economic freedom – private property and freedom
of contract.⁸ Instead, entrepreneurship will reveal signiﬁcant wealth-destructive features
if rent-seeking opportunities are promising. And will bring to light defensive elements
whenever economic freedom is jeopardized, either by other individuals or by
state organizations. Uncertainty and fears about policy-makers’ discretionary power
usually discourage (productive) entrepreneurs from taking action, but may create a
favorable environment for those who believe they can successfully inﬂuence political
mechanisms, or stand better chances to protect their property from outside interference.
Under these circumstances one observes a mix characterized by acts of
destructive and acts of defensive entrepreneurship.
In a simple extension of this principle entrepreneurs may also conceive strategies
that recognize the role of unproductive activities in order to enhance the proﬁtability
of their productive eﬀorts. As an example, this explains why small-size software
developers are inclined to sell their business to larger companies, which are indeed
⁸ See for instance La Porta, Lopez-de-Silanes, Shleifer and Vishny (1997), Laﬀont (2000), Glaeser and
Shleifer (2003), Weede (2006).
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 7
able to reach a wider market (economies of scale in distribution); but are also more
eﬀective in making sure that the government does not force them to reveal the codes
of their newest products, thereby transforming a private proﬁt into a free ride.
As hinted above, company size is also likely to be aﬀected. For instance, in countries
where labour-market regulations increase in severity with the size of the ﬁrm,
defensive entrepreneurs would restrict the size of their company well below what
would allow full exploitation of the economies of scale. On the other hand, in countries
where rent-seeking is more pervasive, large ﬁrms are going to be more eﬀective
in lobbying national governments, and size might increase beyond what would be
optimal from a strictly manufacturing standpoint.
3.1 Simultaneous Entrepreneurship and Transaction Costs
Simultaneous entrepreneurship also helps to shed new light on the role of transaction
costs, an issue of particular importance in the economics of venture capitalism.
When analyzing eﬃciency, economists usually focus on how to reduce transformation
and (standard) transaction costs (Williamson 1981). The former are just a matter
of good engineering, as taught in introductory economics. Standard transaction costs
are the expenses that producers incur in order to reach their business counterparts,
specify the nature of the contract, deliver the goods that form the object of the transaction.
Contract enforcement also belongs to the category of standard transaction
costs. As a result, according to the traditional approach producers are eﬃcient when
they equalize relative factor prices to the marginal rate of technical substitution (optimality
condition); and minimize standard transaction costs.
Although this paper does not deny the importance of standard transaction costs,
the introduction of simultaneous entrepreneurship lays considerable emphasis on
another category of costs, which we call ‘total institutional transaction costs’. In the
present context they describe the diﬀerence between the costs incurred by a proﬁtmaximizing
ﬁrm operating in the real world, where the rules of the game generate incentives
to engage in unproductive and defensive activities; and the costs incurred by
8 New Perspectives on Political Economy
an eﬃcient company in an ideal Lockean institutional context where private property,
freedom of contract and freedom from coercion are guaranteed.⁹ For instance, other
things being equal, companies that make use of labour-intensive techniques usually
enjoy more bargaining power when engaging in rent-seeking vis-à-vis politicians, who
tend to be particularly sensitive to the dynamics of employment, even on a local scale;
as a result, they might be encouraged to use labour-intensive techniques even when
ineﬃcient from a purely engineering/transformation standpoint. In fact they do so,
because such ineﬃcient techniques turn out to be more proﬁtable, once the rentseeking
opportunities are taken into account. Similarly, small-company size might
be an advantage when it comes to evading taxation or dodging regulation, even when
larger size would allow lower average costs. In order to exploit at least some of the
economies of scale, this may thus lead to the existence of networks of more or less
cooperating small companies, or even sets of small companies that are de facto run
by the same management – among other things, with problematic consequences on
the statistics.¹⁰
As will be explained shortly (see section 4), total institutional transaction costs are
likely to play a crucial role in the VC context, for they explain both the features of the
ﬁrm and its relationship with outside actors, among whom funders. In the presence
of signiﬁcant interdependence between institutional transaction costs and the other
costs of production – transformation and standard transaction – the producer might
appear to be ineﬃcient according to the traditional approach, but not necessarily under
the simultaneous view proposed here. The consequence on the funders can be
mixed. Although these actors might understand the importance of optimizing total
institutional transaction costs, they may be unwilling to accept their implications: re⁹
The terms ‘total’ is explained by the fact that these expenses do not refer to the mere cost of dealing
with the institutions (e.g. lobbying). Rather, they concern the cost suﬀered as a result of the decision
to take advantage of the institutional context in order to reap a rent or to avoid a loss. Thus, they
often imply less than ideal engineering; and might also aﬀect the choice of the standard transaction
costs, say when it comes to selecting the appropriate contract.
Surely, the mainstream literature is well aware of the presence of institutional costs. Nevertheless,
it generally considers them as a set of variables that aﬀect localization strategies, i.e. the decision
about where to start a new business and which the relevant time horizon should be. Contrary to
our approach, the traditional view on institutional costs overlooks the analysis of their implications
regarding the operational features and requirements of a ﬁrm, and largely ignores the consequences
for the funding of newly-born companies.
¹⁰ Both examples are typical of the Italian economic context, which is certainly not unique in this respect.
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 9
liance on informal rules, personal and often implicit contracts, exorbitant monitoring
expenses.
Finally, total institutional transaction costs can also play a role in a dynamic perspective.
Contrary to the original view held by Baumol, where the switch from productive
to unproductive activities would take place only in the very long run (historical
periods), we posit that today an entrepreneur must be ready to reshape the mix
of his eﬀorts relatively quickly. Then, total institutional transaction costs are the appropriate
concept to use to account for those entrepreneurial decisions that must be
taken in order to adapt the company characteristics to the new normative conditions.
Once again, a funder who is not involved in daily management and is unwilling to depart
from the business strategies set up at the beginning of the investment period –
which is not unusual in the VC world – might not be quick and ﬂexible enough to
go along with what the institutional dynamics requires. Or might simply be afraid to
blindly trust the founder-manager and take the responsibility for giving up on strict
monitoring.
To sum up, we maintain that the ability of entrepreneurs to combine their productive
and unproductive eﬀorts following the reward structure has a bearing on the
relations between them and the venture capitalists (funders). In the remaining part of
this paper we make use of simultaneous productive/unproductive entrepreneurship
in order to explore the interaction between such two partners in this new light.
4 The diﬀerent Roles of a Venture Capitalist
Venture capitalists are a special category of entrepreneurs, loosely deﬁned as “limited
partnerships in which the managing partners invest on behalf of the limited partners”
(Denis 2004, p. 304). VCs typically acquire a substantial share of equity in relatively
small companies with diﬃcult-to-assess prospects, most of the time because of the
presence of intangibles and strong information asymmetries, thereby requiring close
monitoring.¹¹ In particular, VCs pick young companies with a potential to expand
¹¹ VCs may be minority shareholders, but by introducing preferred equity, convertible securities and
suitable contract clauses they usually make sure they retain control on strategic decisions, create
10 New Perspectives on Political Economy
and improve,¹² possibly in growing industries; and where they sense that they could
make the diﬀerence by contributing as active funders. They share ownership with
the incumbent stockholders (the founders), provide expertise in various domains,
often appoint new managers and revise human-resource policies in the early stages.
However, although their eﬀorts are crucial to obtain success, VCs do not usually engage
in daily management. Furthermore, since they raise their capital from investors
that do not abstain from taking risks and expect high returns within a relatively short
time period, VCs operate on a relatively short-term basis. The life of a venture-capital
fund averages 10 years, while the funding of a projects usually extends over a 3-7 year
period (Tykvová 2007, p.80). That contributes to explaining why their investments
tend to concentrate in high-tech sectors.¹³ And also why VCs tend to disregard large
ﬁrms, where structural change occurs more slowly and/or with greater diﬃculties,
where assessing the personal qualities of the entrepreneurs and the potential of their
ideas in terms of proﬁts is more complex; and where monitoring could be considerably
more expensive.¹⁴ Not to mention that if VCs concentrated on large ﬁrms and
wished to acquire control, each VC would be obliged to concentrate resources on a
very small number of enterprises, thereby reducing the beneﬁts of diversiﬁcation.
Surely, large ﬁrms might oﬀer the opportunity to exploit some economies of scale in
contracting and monitoring. But these are more likely to be reaped by other actors
in the ﬁnancial markets, such as investment funds and merchant banks, which could
do the same without suﬀering neither from the lack of diversiﬁcation, nor from the
time constraints imposed by those who ﬁnance VCs.
incentives to cooperate with the founders and do not reduce the founders’ motivation. See also Gompers
and Lerner (2001), who provide information about the operational features and history of venture
capitalists in the US; as well as Carpenter and Petersen (2002), who document the role of new equity
for small, high-tech companies.
¹² Not necessarily brand-new ﬁrms, though. As reported in Denis (2004, pp. 307-310), start-ups are
ﬁnanced by “angel investors”, while VCs are more likely to intervene at a later stage. Each venture
capitalist follows some eight, nine companies at the same time, which allows him to attend periodical,
non-technical meetings with the management, study the reports and occasionally sit on the board of
directors.
¹³ This also matches the founders’ needs of course. For risky, high tech projects presented by young
ﬁrms with little tangible collateral are unlikely to be ﬁnanced by debt (see for instance Carpenter and
Petersen 2002).
¹⁴ The cost and the eﬀectiveness of monitoring are indeed crucial (Gompers 1995): VCs know that most
companies they are involved in will not produce the expected results. In fact, the diﬀerence between
a good and a bad VC ensues both from his ability to select potential winners and from his talent to
drop losers before too much money has been disbursed.
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 11
On the other hand, founders appreciate the contribution of venture capitalists,
since they provide capital and often contribute also by reducing the cost of risk. Unlike
banks, they do not require collateral, and unlike public providers of equity funds
they do not force the founders to disclose their plans to a wide public (including
potential competitors). In addition, VCs usually oﬀer an extensive network of contacts,
which in many cases represents a welcome contribution to the marketing opportunities
of the founders. Perhaps even more important, venture capitalists fulﬁll
an important signaling function. For instance, most recently-born companies, especially
if started with limited ﬁnancial resources, ﬁnd it diﬃcult to persuade potential
customers about the quality of their products and suppliers about their creditworthiness.¹⁵
Similarly, VC support makes it easier to approach additional funders (e.g.
banks) or qualiﬁed workforce, for under such circumstances the presence of VCs certiﬁes
the producer.
Of course, from the VC standpoint the ideal partnership with a founder is a situation
where competitive pressures reduce the cost of monitoring – the size of the
proﬁts (or of the losses) will signal whether the company is badly managed – and
the personal role of the founders is limited, so as to make sure that the value can be
transferred with the company, rather than with the founders. For although VCs risk
their resources along with the founders, founders and funders do not always share
the same preferences and behavioral patterns: the potential for conﬂict on strategic
decisions looms high (Kaplan and Schoar 2005).
In fact, the founders-funders relationship presents a number of problems, most
of them typical of the principal-agent literature. The agent/founder may fail to disclose
relevant information about the company and the industry (or the market niche)
where it operates; by managing daily business he has opportunities to siphon oﬀ company
proﬁts into his own personal accounts; by having a smaller and time-constrained
stake in the company the founder might reduce his entrepreneurial eﬀorts and possibly
deploy them somewhere else. On the other hand, the founder might be persuaded
that the short-run strategies enacted by VCs ultimately damage the company,
and that his increased accountability reduces his degrees of freedom, his willingness
¹⁵ See Gompers (1995, p.1403), Cable and Shane (1997, p.171), Hellmann and Puri (2000, p. 960).
12 New Perspectives on Political Economy
to experiment, his ability to create a suitable environment within the ﬁrm. Not to
mention the risk of being replaced by outside CEOs (Hellmann and Puri 2002).
According to this (widely accepted) version of the principal-agent problem, there
are three ways to mitigate the conﬂict. First, the investor (principal) can collect information
about the entrepreneur before funding the project with the intention to
screen out bad undertakings. Second, the investor can write an elaborate contract
with the entrepreneur. This contract will allocate the beneﬁts of the real investment
– cash ﬂow, control and termination rights – between the two parties so as to provide
incentives to the entrepreneur, while defending the interests of the VC. Third,
during the contract the VC can monitor the entrepreneur in order to ascertain that
the project is run properly. All three, of course, require additional costs, which can
further increase if VCs choose to intervene under syndicated forms.
In general, however, tensions between the two groups remain frequent. Sophisticated
contracts and eﬀective enforcement might reduce the cost of opportunism
when entrepreneurship is entirely productive. Still, if the incentives to engage in defensive
entrepreneurship are signiﬁcant, the role of personal relations and informal
routines can become very important, especially when the original founders and managers
are few, often times close relatives to each other or at least long-time friends
(Faccio and Lang 2002). Under such circumstances company loyalty and personal
loyalty are one. Transparency and accountability need to be replaced by trust, formal
procedures by informal dealings. As a whole, entrepreneurial speciﬁcity increases, in
that the success of a ﬁrm engaged in defensive activities depends more and more on
the characteristics and personal contacts of the founding entrepreneurs, on whom
most informal contracts are built. Obvious examples in this direction are tax evasion
and tax avoidance (both require several people turning a blind eye, or both),
underreported work performances (they involve side payments to the worker), dodging
regulation, coordinated access to government procurement (whereby diﬀerent
companies take turns and decide ex ante each time who the winner is going to be).¹⁶
Within this (defensive) framework venture capitalists are still an attractive possi¹⁶
Entering a cartel that distributes rents (government procurement) can be necessary – and thus qualify
as a defensive move – when staying out and trying to outcompete the cartel may lead to violent
retaliation by cartel members. Of course, in this case the border line between defense and destruction
can become very thin, for a defensive decision might easily degenerate into a destructive activity.
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 13
bility for the founders (and vice versa) if they are friends or relatives. But they tend to
become a questionable choice otherwise. In the end, outsiders feel uneasy, as they remain
unwilling to invest in enterprises that might be proﬁtable, but require substantial
investments in the enforcement of informal contacts.¹⁷ Eventually either funding
partners become founders themselves, or VCs prefer to move out and founders
increase their dependence on personal or family funds, bank loans guaranteed by
personal asset.
5 Venture Capitalisms and Institutions
The previous sections have outlined the elements of the relation between funders
and founders. This section develops those insights by focusing on two institutional
categories: property rights and regulation. For each category diﬀerent situations will
then be evaluated in order to assess how the role of VCs is aﬀected, to what extent
the presence of VCs is necessary, how companies are likely to evolve.
5.1 Property rights: Weak Enforcement and Discretionary Assignment
Today much of the academic debate about property rights regards the principles underlying
their assignment and the incentive structure that those principles imply.
However, it is often overlooked that the main problem with private property rights is
not their theoretical regime as deﬁned – say – in the constitution. Stability and credibility
are far more important. Property rights may be clearly deﬁned, but they are of
little relevance if they can frequently be altered by ordinary politics, or ‘interpreted’,
bypassed, watered down by the judiciary.¹⁸ That explains why, when enforcement is
¹⁷ See Hart (2001), Cable and Shane (1997) and Kaplan and Strömberg (2004) on the design of optimal
contracts that would reduce conﬂicts of interests between VCs and entrepreneurs. Contrary to this
(prevailing) literature, we claim that VC contracts do not break down when badly speciﬁed. Instead,
they break down because they are no longer suitable when informal agreements become more important.
Informal agreements are reliable when they are the result of repeated interaction over a
long period of time among individuals that share the same time horizons and the same structure of
accountability. This is not what happens in the VC framework.
¹⁸ For instance, Opper (2008) notes that the success of transition in the former Soviet-bloc countries
does not depend on the quality of institutions (assignment), but rather on the degree to which they
can be put into eﬀect (enforcement).
14 New Perspectives on Political Economy
weak or inconsistent, eﬀorts are generally devoted to protecting existing assets from
outside aggression, or to joining the aggressors. As a result, entrepreneurship is likely
to present a high degree of defensive or even destructive elements. Instead, productive
activities are limited to short-term initiatives and VC funders are likely to shy
away: proﬁts tend to be hidden, personal relations and possibly some forms of corruption
become critical, accountability to outsiders turns out to be a heavy liability.
In a word, tackling institutional transaction costs raises the cost of monitoring in the
founders-funders interactions and easily becomes intolerable for the latter.
On the other hand, when property-right enforcement is credible, but assignment
is unstable (discretionary), rent-seeking activities at a national or super-national level
are encouraged. Thus, in large countries destructive entrepreneurship pays oﬀ for
large-size companies that stand a chance of internalizing a signiﬁcant share of their
rent-seeking eﬀorts. Once again, these are not an ideal target for VCs, though. In
addition, smaller companies can be successful only by engaging in defensive activities,
which also cuts down VC presence. To conclude, VCs are unlikely to be numerous in
this institutional environment.
5.2 Regulation
Venture capitalists’ attitudes are also inﬂuenced by regulation, which generates two
sets of consequences. By interfering with individual decision making, regulation reduces
eﬃciency and productivity overall. As a result, proﬁt opportunities are also
less attractive. Furthermore, the weight of regulation is not uniform across industries,
size, production techniques and classes of actors. Hence, in regulated environments
some categories of ﬁrms can do better than others; and some defensive
strategies are more proﬁtable than others. For instance, if compared with a largecompany
manager, the owner-manager of a small company appropriates a larger portion
of the residual created by successful defensive entrepreneurship. Thus, when it
comes to defensive entrepreneurship, competitive selection is likely to reward small
companies with respect to large ﬁrms. On the other hand, large companies might
be favoured when engaging in destructive activities, since a large company enjoying
political connections and aﬀection may be able to drive the regulatory bodies to their
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 15
own advantage and be more proﬁtable than smaller companies open to competition,
but without appropriate contacts.
True enough, VCs are not necessarily deterred by regulated environments, as long
as transparency is satisfactory and the rules are enforced consistently, so that monitoring
costs do not become prohibitive. It is however important to observe that the
demand for VC funders is going to remain modest. As emphasized in section 2, within
a regulated environment part of the proﬁts are in fact either established rents guaranteed
by normative barriers to entry, or the reward to defensive entrepreneurship,
hard to monitor, heavily characterized by intangibles (trust) and by asset speciﬁcity
(high dependence on the founder). Thus, VCs’ hopes to make proﬁts are realistic only
for incumbents who might have dissipated the rent through bad management, rather
than for newcomers competing for market shares and possibly displacing obsolete or
ineﬃcient producers.
In regulated environments venture capitalists will therefore be eager to provide
ﬁnancing to companies that had been poorly-managed and where the incumbent
management can be easily replaced. Under such circumstances the VC proﬁt would
amount to the previously dissipated normative rent. Still, why should founders need
venture capitalists to get rid of the (bad) incumbent management and replace it with
new staﬀ? In fact, they often don’t, to the advantage of headhunters as well as of prestigious
consultants that take the responsibility of the changes that the owners alone
wouldn’t have been able to enforce.
6 Growth and Structural Change
Surely, growth with modest venture capitalism is possible, as the European experience
has shown. Still, the absence of VCs is important in two respects. VCs are a
tell-tale sign about the prevailing institutional conditions (low institutional transaction
costs) and the incentives to concentrate entrepreneurial talents on productive
ventures. Hence, they act as some kind of ‘entrepreneurial multiplier’ by motivating
other categories of funders with an interest in productive projects. In addition,
growth without VC often implies the presence of undesirable path-dependence processes.
As noted previously, when companies grow by carrying out substantial de-
16 New Perspectives on Political Economy
structive and defensive activities, ﬁnancial sources tend to be provided by banks or
personal funds. In particular, banks are relied upon when founders exhibit long-time
horizons,¹⁹ or wish to supplement or replace personal funds (personal collateral is
however required to cover risk). Put diﬀerently, the banking sector plays an important
role either in order to manage the rent-seeking economy, thereby signalling some
forms of crony-capitalism; or to provide support to small, ﬂexible companies, possibly
in the presence of substantial personal assets (family capitalism). Such companies
solve the private property right problems – poor enforcement and/or instability – by
shortening their time horizon, whereas ﬂexibility allows them to reduce the cost of
regulation.
As a result, the absence of VCs reveals property-right and/or regulatory features
that drive the economy towards a structure characterized by a few large companies
engaging in destructive entrepreneurship and a substantial number of small ﬁrms engaging
in defensive entrepreneurship. The former are unlikely to expand and display
a propensity to collude with the banking sector. Lack of transparency scares stockexchange
investors to a larger extent than (domestic) bankers familiar with relationship
lending.²⁰ The latter are family funded. They are seldom based on high-tech
breakthroughs, for this area is too risky and usually requires considerable investment
in R&D. In addition, they are unable to increase in size. They would be beyond reach
for family ﬁnancing and too vulnerable to regulatory authorities to be proﬁtable.
In the end, one might speculate that an economy that does without venture capitalism
is characterized by relatively high barriers to entry (otherwise rent-seeking
would be eroded or deeply resented) and a somewhat inadequate framework for impersonal
trade to take place satisfactorily. Product innovation does remain attractive,
especially if conceived by small-size companies that subsequently sell their rights to
larger ﬁrms interested in further development for commercial purposes. Nonetheless,
in a regulated economy where personal networks play a signiﬁcant role a large
¹⁹ Family loans are usually short term. Otherwise they tend to become semi-gifts: they are paid back if
and when possible, but if the debtor is in trouble it is unlikely that the credit is terminated, or that
the creditor is taken to court.
²⁰ This may be typical of areas where capital market globalization is not welcome. If so, by restricting
competition established rents are protected and producer’s proﬁts are high enough not to require
external ﬁnancing. Following this insight, Johnson et al. (2002) have observed that local companies in
the former communist bloc were not suﬀering from signiﬁcant ﬁnancial constraints during transition.
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 17
part of technological growth is likely to be imported from outside. New technologies
are bought, imitated and/or adapted, for the domestic rewards to the search for
potential technological breakthroughs would be rather modest, as VCs know all too
well. Surely, despite disregard for signiﬁcant and systematic product or process innovation
(supplanted by imitation), productive entrepreneurship is not necessarily
absent. For instance, one can devise new or more eﬀective organizational structures,
discern new sources of latent demand or conceive new ways to reach that demand.
Still, once the potential for catching-up has been exploited, growth prospects are going
to be constrained by the unwillingness to take risks on a large scale or – more
precisely – to share risk and thus allow productive entrepreneurs to pursue their intuitions.
Sometimes the constraint is severe, e.g. in Italy and France; sometimes its
bite is minimal, as in China or India.
7 Summary and Conclusions
The previous paragraphs have suggested that venture capitalism is just one among
the various possibilities of transforming founders into successful producers of goods
and services. The appeal of this form of ﬁnancing depends on many variables that can
be summarized by referring to the institutional features within which entrepreneurs
operate. In particular, venture capital tends to be a poor choice when informal rules
are at odds with the formal institutional framework; or when productive entrepreneurship
is stiﬂed by violations in property rights, regulation, privileges. In the end,
either the time horizon becomes very short – intangibles and asset speciﬁcity are
dominant – and the outcome depends on the founders’ action; or uncertainty prevails
and decision-making becomes some sort of systematic groping. Both scenarios imply
high institutional transaction costs and thus high monitoring costs. Often times too
high for VCs’ tastes.
Indeed, the need to allocate eﬀorts across productive and unproductive activities
creates an extra dimension of asymmetry between the two parties. Because of the
asymmetric information between founders and funders, the founders ﬁd it cheaper
to raise capital from family and friends, whereby the cost of trusting is lower than the
cost of monitoring.
18 New Perspectives on Political Economy
Bottazzi and Da Rin (2004) rightly note that European venture capitalists fund
less than one third of the projects ﬁnanced by their American counterparts. Contrary
to common beliefs, however, this does not imply that Europe is lacking entrepreneurship
or that free enterprise in Europe is about to die. For diﬀerent capitalist practices
are just one of the consequences of the deep institutional disparities between − say −
Continental Western Europe and the US. Indeed, diﬀerences in ﬁnancing, formal and
informal company structures and industrial specialization are the logical responses to
diverse incentive structures. By developing Baumol’s original insight, this paper has
claimed that such responses can be framed and understood according to the features
of entrepreneurial eﬀorts. One way or another, individuals that engage in productive
eﬀorts frequently – perhaps always – engage in some kind of destructive and/or defensive
activity as well. And the presence of venture capitalism depends on the mix
of the varieties. Put diﬀerently, institutional features aﬀect agents’ behaviour, which
in turn determines to what extent cooperation with VCs is mutually proﬁtable. As a
result, institutional incentives and ﬁnancing options lead to alternative structures of
development and diﬀerent potential feed-back eﬀects upon the institutional context.
Further insights on the nature of the interaction between venture capitalism and
entrepreneurship can be produced by testing the implications of the arguments outlined
in the previous sections and developing the work pioneered by Jeng and Wells
(2000). For instance, the presence of entrepreneurship could be further explored by
comparing the dynamics and features of newly-formed companies with the prevailing
rules of the game (institutions). The role and industrial concentration of VCs across
countries could be explained by referring to the formal and informal institutional
features of each area, as hinted in section 5. In this light the nature of regulation,
the role and eﬀectiveness of the judiciary system, informal contracting, the extent of
corruption are all obvious explanatory variables. The indirect causal links between
VC intervention and family business (size, funding, informal networks) is an additional
area where the theoretical instruments proposed in the previous pages could
be applied.
More challenging from a speculative viewpoint is however the analysis of the eventual
feedback mechanisms, which also lies beyond the scope of this work. Still, one
normative clue of some consequence seems apparent at this stage already. In partial
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 19
contrast with the recent literature – see for instance Antonelli and Teubal (2007) –
venture capitalism per se is not the solution to stagnant growth, in Continental Europe
or elsewhere. But it does signal the presence of a healthy business environment
where entrepreneurial energies are more likely to be channeled towards productive
purposes.
Bibliography
[1] Antonelli, Cristiano and Morris Teubal. 2007. “Venture Capitalism as a Mechanism
for Knowledge: the Emergence of the Markets for Knowledge Intensive
Property Rights”. ICER Working Paper Series.
[2] Baumol, William J. 1990. “Entrepreneurship: Productive, Unproductive and Destructive”.
Journal of Political Economy, 98, pp. 893-921.
[3] Berger, Allen and Gregory Udell. 2002. “Small Business Credit Availability and
Relationship Lending: the Importance of Bank Organisational Structure”. Economic
Journal, 112 (477), February, pp. F32-F53.
[4] Boettke, Peter and Christopher J. Coyne. 2003. “Entrepreneurship and development:
cause or consequence?” In R. Koppl (ed.), Austrian Economics and Entrepreneurial
Studies, Amsterdam: JAI Press, pp. 67-88.
[5] Bottazzi, Laura and Marco Da Rin. 2004. “Financing Entrepreneurial Firms in
Europe: facts, issues and research Agenda”. In C. Keuschnigg and V. Kanniainen
(eds.), Venture Capital, Entrepreneurship and Public Policy, Cambridge (Ma):
MIT Press.
[6] Cable, Daniel and Scott Shane. 1997. “A prisoner’s dilemma approach to
entrepreneur-venture capitalist relationships”. Academy of Management Review,
22 (1), January, pp. 142-176.
[7] Carpenter, Robert and Bruce Petersen. 2002. “Capital Market Imperfections,
High-tech Investment, and New Equity Financing”. Economic Journal, 112 (477),
February, pp. F54-F72.
20 New Perspectives on Political Economy
[8] Denis, David J. 2004. “Entrepreneurial ﬁnance: an overview of the issues and
evidence”. Journal of Corporate Finance, 10 (2), March, pp. 301-326.
[9] Faccio Mara L. and H.P. Lang. 2002. “The Ultimate Ownership of Western European
Corporations”. Journal of Financial Economics, 75, pp. 375-395.
[10] Glaeser, Edward and Andrei Shleifer. 2003. “The Rise of the Regulatory State”.
Journal of Economic Literature, 31, pp.401-425.
[11] Gompers, Paul. 1995. “Optimal Investment, Monitoring, and the Staging of Venture
Capital”. Journal of Finance, 50 (5), December, pp.1461-1489.
[12] Gompers, Paul and Josh Lerner. 2001. “The Venture Capital Revolution”. Journal
of Economic Perspectives, 15 (2), Spring, pp. 145-168.
[13] Hart, Oliver. 2001. “Financial Contracting”. Journal of Economic Literature, 34 (4),
December, pp. 1079-1100.
[14] Hellmann, Thomas and Manju Puri. 2000. “The interaction between product
market and ﬁnancing strategy: the role of venture capital”. Review of Financial
Studies, 13 (4), Winter, pp. 959-984.
[15] Hellmann, Thomas and Manju Puri. 2002. “Venture capital and the professionalization
of start-up ﬁrms: empirical evidence”. Journal of Finance, 57 (1), February,
pp. 169-197.
[16] Jeng, Leslie and Philippe Wells. 2000. “The determinants of Venture Capital
Funding: Evidence across Countries”. Journal of Corporate Finance, 6, September,
pp. 241-289.
[17] Johnson, Simon, John McMillan, Christopher Woodruﬀ. 2002. “Property Rights
and Finance”. American Economic Review, 92 (5), December, pp. 1335-1356.
[18] Kaplan, Steven N. and Per Strömberg. 2004. “Characteristics, Contracts and Actions.
Evidence from Venture Capitalist Analysis”. Journal of Finance, 59, pp.
2177-2210.
[19] Kaplan, Steven N. and Antoinette Schoar. 2005. “Private Equity Performance:
Returns, Persistence and Capital Flow”. Journal of Finance, 60, pp. 1791-1823.
Colombatto and Melnik: Productive and Non-Productive Entrepreneurship … 21
[20] Laﬀont, Jean-Jacques. 2000. Incentives and Political Economy, Oxford: Oxford
University Press.
[21] La Porta Rafael, Florencio Lopez-de-Silanes, Andrei Shleifer and Robert Vishny.
1997. “Legal Determinants of External Finance”. Journal of Finance, 52, pp. 113-
1150.
[22] Opper, Sonja. 2008. “New institutional economics and its application on transition
and developing economies”. In E. Brousseau and J.-M. Glachant (eds.), New
Institutional Economics, a Guidebook, Cambridge: Cambridge University Press,
pp. 389-406.
[23] Tullock, Gordon. 1993. Rent Seeking, Cheltenham: Edward Elgar.
[24] Tykvová, Tereza. 2007. “What do economists tell us about venture capital contracts?”
Journal of Economic Surveys, 21 (1), pp. 65-89.
[25] Weede, Erich. 2006. “Economic freedom and development: new calculations
and interpretations”. Cato Journal, 26 (3), Fall, pp. 511-524.
[26] Williamson, Oliver E. 1981. “The Modern Corporation: Origins, Evolution and
Attributes”. Journal of Economic Literature, 19, pp. 1537-1568.
ISSN 1801-0938
New Perspectives on Political Economy
Volume 4, Number 1, 2008, pp. 23 – 40
Collective Consumption Externalities and Charitable
Giving
Scott A. Beaulier and Joshua C. Hall
JEL Classiﬁcation: D60, H41
Abstract: Charitable giving suﬀers from market failure, which means government
redistribution and other forms of social insurance can be justiﬁed. According to market
failure theorists, charitable giving for poverty alleviation suﬀers from a free-rider
problem. In a free market, giving will, therefore, be underprovided. Using insights
from public choice theory and Austrian economics, our paper ﬁnds market failure
arguments to be an insuﬃcient justiﬁcation for government intervention. Private
charity and redistribution respects autonomy and is more robust in creating incentives
for people to help the poor.
An earlier version of the paper titled “Help Me Help Myself: The Economics of Private Charity and
Redistribution” was presented at the fourth annual Mises Seminar in Sestri Levante. The authors
would like to thank Pierre Garello and other seminar participants for their helpful comments and
suggestions.
Scott A. Beaulier – Department Chair and Assistant Professor of Economics, Stetson School of Business
and Economics, Mercer University, Macon, GA 31207, beaulier_sa@mercer.edu.
Joshua C. Hall – Department of Economics and Management, Beloit College, Beloit, WI 53511,
halljc@beloit.edu.
24 New Perspectives on Political Economy
1 Introduction
In America today, classical liberals and conservatives are alarmed by the runaway level
of government involvement in our lives. Interventions aﬀect and complicate nearly
every facet of our lives and impose tremendous explicit and implicit costs. The sheer
number and complexity of the rules governing our behavior seems to be at an alltime
high, and the rapid increase of government interventions in our everyday lives
in order to “make them better oﬀ” deserves explanation.
Alternative opinions about the overall scale and scope of government intervention
depend, of course, on one’s more general theoretical understanding of the welfare
state. Analysis of the current state of poverty alleviation programs in America cannot
be separated from one’s general model of the welfare state. We take as our point of
departure Thomas Sowell’s discussion of the welfare state (2003, B7). According to
Sowell, “The welfare state is not really about the welfare of the masses. It is about the
egos of the elites.” Like Sowell, we think government intervention hampers people’s
lives and increases in intervention beget more intervention. Our main purpose is
to critically examine one argument for government intervention into the market for
charitable giving to the poor.
Historically, the economic justiﬁcation for government involvement in programs
for the poor was based on public goods theory. In his classic book, Capitalism and
Freedom, Milton Friedman (1962, 191) discussed the public goods justiﬁcation for
poverty alleviation when he wrote, “I am distressed by the sight of poverty; I am beneﬁted
by its alleviation; but I am beneﬁted equally whether I or someone else pays
for its exclusion.”
Friedman’s claim was picked up by others.¹ Shortly after Friedman, Hochman
and Rodgers (1969) claimed that private charity and transfers were underprovided in
a free market because of people free-riding on the donations of others. At the core
of their argument was the notion that people have interdependent utility functions,
i.e., the utility of person A is dependent in part on the utility of person B.
¹ Lee and McKenzie (1990) credit Friedman for being the ﬁrst to make this connection.
Beaulier and Hall: Collective Consumption Externalities and Charitable Giving 25
In this paper, we question whether charity will be underprovided in a free market
because of free-riding on others donations. We ﬁrst look at the overall role interdependent
utility functions play in a person’s decision to give. Most charitable giving
seems to occur because of the direct beneﬁts received by the donor. To the extent
people do have interdependent utility functions, interdependence does not seem to
matter on the margin since most attempts to solicit donations do not appeal to the
welfare of those helped by the additional donation. Private motivations for giving
signiﬁcantly reduce the argument for government intervention because private charity
will reduce or eliminate the gap between the “market failure” level of charity and
redistribution and the socially optimal level of charity and redistribution.
Second, even if interdependent utility functions exist and free-riding occurs, the
resulting outcome is not ineﬃcient because no resources are misallocated. Interdependent
utility functions generate pecuniary externalities, which do not misallocate
resources. The “ideal” condition, where no free-riding occurs, is analogous to a market
environment with no competition. Just as an absence of competition would not
be good for competitive markets, an absence of pecuniary externalities in charitable
consumption would not be good because much of the innovation in charitable
fund-raising ﬂows from the pecuniary externality associated with charitable giving.
An economic analysis of charitable giving yields important insights into the
proper role of government towards charity and redistribution from the rich to the
poor. The desire to help the worst-oﬀ in society is strong and can engender considerable
support for the welfare state in its various forms, from transfer programs to a
paternalistic regulatory state. We maintain that the best approach to helping people
help themselves is through decentralized markets.
The remainder of our paper proceeds as follows. Section 2 discusses market failure
theory in the context of helping others in society. Section 3 critically examines the
notion that charitable giving suﬀers from a market failure, while Section 4 concludes
with some policy implications of our analysis and some insights from private charity
and paternalism.
26 New Perspectives on Political Economy
2 Helping Others and Market Failure Theory
The paternalism of contemporary America comes from an obvious and pernicious
source: politicians. Politicians – both liberal and conservative – have been promising
voters a world of increased health, safety, and wealth, but they never succeed at
fulﬁlling their promises because of the elastic and ﬂeeting nature of the goods they
are asked to provide. But, in their attempt to satisfy our every desire, politicians have
made good on delivering one thing to voters: a massive increase in the scale and scope
of government. In a fundamental sense, Americans are afraid to be free. Politicians
are often the “messengers” delivering policies that exploit our fears.²
On the opposite side of the political process, voters themselves often contribute
to a more expansive state.³ When it comes to economic policy, the general public
exhibits a tremendous amount of ignorance that borders on irrationality (Caplan
2007). Numerous contradictions can be found in the general public’s social welfare
function. The general public wants higher levels of education and defense spending,
but lower taxes. They want higher levels of economic prosperity, but believe in a more
progressive system of taxation. Thus, while politicians are often acting as “political
entrepreneurs” (Wohlgemuth 2002) interested in discovering new political opportunities,
they are at times simply responding and giving voters the policies they want
and deserve.
The promises made by paternalistic politicians have led to a large increase in government.
In addition to growing in overall size, the government has become increasingly
concentrated at the Federal level because more local sources of authority cannot
make sense out of all of the contradictory regulations being created by politicians.
For those trained in Austrian economics, power becoming increasingly concentrated
comes as no surprise. As Mises (1983 [1944]), 3) put it when describing the bureaucrat,
He has arrogated a good deal of legislative power. Government commissions
and bureaus issue decrees and regulations undertaking the management and direction
of every aspect of the citizens’ lives. Not only do they regulate matters
² Buchanan (2005) argues that big government will remain an enduring problem because of man’s
yearning to escape, evade, and even deny the responsibilities that come with being free.
³ We thank an anonymous referee for pointing out and encouraging us to discuss the “double-sided
dynamic” of the political process.
Beaulier and Hall: Collective Consumption Externalities and Charitable Giving 27
which have hitherto been left to the discretion of the individual; they do not
shrink from decreeing what is virtually a repeal of duly enacted laws… Every day
the bureaucrat assumes more power… There cannot be any doubt that this bureaucratic
system is essentially anti-liberal, undemocratic, and un-American …
Over time, piecemeal interventions beget more interventions, and the interventions
continue until we have arrived at a totalitarian state.
While man’s desire for more regulation has played an important role in the rise
of the regulatory state, the market failure literature in economics has also helped to
justify and make popular many government proposals. According to market failure
theory, people do not care about the spillover beneﬁts/costs of their behavior. Therefore,
government must enter markets with externalities to correct for the alleged failures.
The remedies to market failure can take many speciﬁc forms, but, in general,
disincentives are necessary to curb negative externalities and positive incentives are
needed to subsidize markets where positive externalities are present.
For advocates of market failure theory, correcting for market failures was a rather
straightforward task (Samuelson 1947; 1961). More sophisticated critics of market failure
theory recognize the epistemological challenge involved in sorting out the failing
and well-functioning markets, estimating the size of the failure, and prescribing corrective
remedies (Wagner 1989). Public choice economists also criticized market failure
theorists for their naiveté: even if bureaucrats and politicians had the necessary
information about external costs, they lack the proper incentives to be correcting for
market failures (Shleifer and Vishny 1998).⁴
Over time, policymakers in Washington have ignored the criticisms of market
failure theory and latched on to the academic writings of market failure theorists.
The new literature on market failure theory helped politicians justify interventions,
and it undoubtedly aﬀected the general public’s attitude towards government. As
Lord Keynes (1936, 383) put it:
[T]he ideas of economists and political philosophers, both when they are
right and when they are wrong, are more powerful than is commonly understood.
⁴ Levy (2002) and Boettke and Leeson (2004) argue for a “robust” approach to political economy that
makes “worst case” assumptions where politicians and bureaucrats are said to be both non-benevolent
and non-omniscient.
28 New Perspectives on Political Economy
Indeed the world is ruled by little else. Practical men, who believe themselves to
be quite exempt from any intellectual inﬂuences, are usually the slaves of some
defunct economist.
Hayek (1960 [1949], 371) concurred with Keynes, when he wrote, “… [intellectuals]
have probably never exercised so great an inﬂuence as they do today … by shaping
public opinion.”
For both Keynes and Hayek, the views of the intellectual come to shape the political
landscape in an extremely watered down fashion. When policymakers latch onto
an academic idea, errors often occur. Many of the ideas selected by policymakers
are not the “best” ideas or the “right” ones, but, rather the ideas progressive intellectuals
and policymakers stand behind. Market failure theory is one such idea. It is
hard to imagine there ever being a time where a majority of economists supported
widespread interventions to correct for market failure. Yet, market failure arguments
gained traction because they provided intellectual support for a variety of government
interventions.
The implicit assumption of the market failure literature is that with enough regulation,
all problems can be solved. By constantly passing new laws and massaging our
fear to be free, politicians supply the false sense that government is “solving” market
failures. Like Demsetz (1969), we think market failure theory suﬀers from the nirvana
fallacy. In its simplest form, the nirvana fallacy is a comparison of real markets with
perfect government. When it comes to charity and redistribution, the public good
characteristics of charity makes it prone to free-riding behavior. The mere possibility
of public good characteristics in charity is then taken as suﬃcient proof of the need
for government intervention.
But, government interventions are determined by fallible men and women with
their own biases and agendas. A proper stocktaking of the desirability of intervention
would: (1) evaluate a market failure argument on its own terms (i.e., is it really
a market in the sense of misallocating resources), and (2) compare the robustness of
markets and government in addressing the failure in the short and long run. Comparing
the robustness of markets over the short and long run is important because
government could be more eﬃcient in mitigating the short-run eﬀects of a market
Beaulier and Hall: Collective Consumption Externalities and Charitable Giving 29
failure but crowd out the evolution of decentralized institutional mechanisms, which
would be superior in the long run. With an eye towards both market and government
failure, we critically evaluate one of the most prominent arguments for government
involvement in charity and redistribution
3 Collective Consumption Externalities and Helping the Poor
One of the primary economic arguments for the public provision of charity is that it
the supply of charitable activities will be underprovided in a free market because charitable
giving has the properties of a collective consumption good. Technically speaking,
collective consumption goods are those for which a many people can simultaneously
consume the good without reducing the amount available for consumption.⁵
National defense is the classical example of a collective consumption good. An increase
in a country’s population increases the number of people consuming national
defense without reducing any other citizen’s consumption. The incentive to free ride
increases with collective consumption goods, such as national defense, because people
can enjoy the beneﬁts of national defense without contributing to it. If we make a
few simplifying assumptions about a person’s willingness to pay for collective goods,
government interventions, such as compulsory taxation, can improve upon the voluntary
outcome.
People donate time and money to charitable organizations for many reasons.
Some donate for personal ediﬁcation while others donate because they care about
the increased utility of those helped through an organization’s programs. When a
person’s utility depends on the utility of others, they are said to have interdependent
utility functions. Charitable donations increase the utility of people with interdependent
utility functions because when we know the recipients of charity are better
oﬀ; we are made better oﬀ. If donors have interdependent utility functions, charitable
giving is a collective consumption good where all potential donors beneﬁt when
⁵ In his seminal paper that essentially created public goods theory, Samuelson (1954, 387) describes
collective consumption goods as those “which all enjoy in common in the sense that each individual’s
consumption of such a good leads to no subtractions from any other individual’s consumption of that
good....”
30 New Perspectives on Political Economy
someone gives to charity. If I am a potential donor whose utility depends on the utility
of the African poor, my utility increases when The Bill and Melinda Gates Foundation
donates $47 million to help treat tropical diseases plaguing Africa’s poor (Dugger
2006).
Just as in the case of national defense, people can enjoy the “warm glow” from
knowing the poor are being made better oﬀ without having to contribute. According
to Hochman and Rodgers (1969), some people will free ride on the charitable donations
of others, which means the supply of charitable giving in a free market will
be underprovided. Maximizing utility, they argue, requires government intervention
to ensure the “optimal” level of transfers is made to the poor (Holcombe and Sobel
2000). The possibility of an insuﬃcient supply of giving has led to government programs
that aim to increase giving, such as the charitable deduction (Hochman and
Rodgers 1977).⁶
There are many problems with this argument. At the most basic level, what if people
do not have interdependent utility functions? Or to put it another way, why don’t
all donors free ride? Many people obtain personal pleasure in giving even though
they cannot observer the eﬀect of their actions on the welfare of the recipients (Arrow
1972). While Arrow’s formulation of donor motivation is surely not true for all
donors (Rose-Ackerman 1982), a considerable amount of donor motivation for giving
does not appear to be about the welfare of the recipient. Donors appear to care about
the “warm glow” associated with charitable giving, and beneﬁt from the wealth signal
they send when they give (Glazer and Konrad 1996).
When we look at the way in which charities solicit donations, they are clearly not
appealing to donors’ altruistic motives. Development oﬃcials do not rely primarily
on stories regarding the charity’s work to raise money. While the charitable works
are always part of the marketing campaign, the actions of development oﬃcials point
towards donors being motivated by the approbation ﬂowing from giving or the signaling
value of a donation. One manifestation of the signaling motivation is in “naming
rights,” where a new facility or program is named after a donor. Or charities provide
diﬀerent “tiers” of giving, with the choicest of status gifts being given to the high⁶
Gergen (1988) provides an excellent discussion of arguments for and against the charitable deduction,
including that of Hochman and Rodgers (1977).
Beaulier and Hall: Collective Consumption Externalities and Charitable Giving 31
est donors (e.g., “gold circle members receive regular updates from the president and
monthly donor receptions”).
The warm glow is also a powerful motivator. Donors not only want to see the poor
better oﬀ, but they want to feel good about having been part of the eﬀort. Andreoni
(1990) quotes the Red Cross using the slogan “Feel good about yourself – Give blood!”
Donors appear to be motivated by many reminders of their charitable deeds, either
for internal reasons or because of external approbation of their eﬀorts. Spreading
charitable giving around instead of focusing on the one cause a person feels “best”
about is more evidence against the altruistic giving explanation.⁷ McGranahan (2000)
examines seventeenth-century English wills and ﬁnds more bequests left to the poor
when the deceased had fewer immediate family and friends. Even in death, people
appear to be more concerned with how they are perceived by others.
In economic models where preferences are modeled as being purely altruistic,
public giving reduces private giving dollar-for-dollar: if all you care about is the welfare
of the recipient, government ﬁnanced transfers will cause you to reduce your
voluntary donation by the amount of the forced transfer. Yet ﬁeld tests fail to ﬁnd
considerable “crowding out” of private charitable donations (Clotfelter 1985; Kingma
1989; Ribar and Wilhelm 2002).⁸ The fact that there is not complete crowding out
suggests that donations to private charities are often motivated by factors other than
the welfare of the recipient.⁹ If for example, you also care about the warm glow you
receive from donating to help the poor, you are not going to stop doing so because
the government is now taxing you to provide charity.¹⁰
We are not making a normative claim about people’s motivations for giving.
⁷ This argument is made by Landsburg (1997), who argues that if our motivations were purely altruistic,
we would give all charitable donations to the charity we think does the greatest good. Giving to
ten diﬀerent charities is evidence, Landsburg argues, that people care about their own sense of self-
satisfaction.
⁸ At best, the evidence on crowding out is mixed. Some recent papers by Andreoni and Payne (2003)
and Gruber and Hungerman (2005), for example, show strong evidence of some crowding-out.
⁹ As pointed out by an anonymous referee, the sphere of charitable giving is wide and motivations for
giving vary considerably across individuals and organizations. Crowding out could be high in certain
contexts such as religious giving where fundraising eﬀorts are based on observable need. This, for
example, could explain the high-levels of crowding out found by Gruber and Hungerman (2005) in
Depression-era churches in response to the New Deal.
¹⁰ It should be clariﬁed, however, that the low level of crowding out does not mean that government
charitable activity is not harmless. There is the not so insigniﬁcant matter of the administrative and
32 New Perspectives on Political Economy
Rather, we are arguing against the idea that people are primarily motivated by altruistic
concerns. We do not believe giving for purely altruistic reasons has any moral
superiority over other kinds of giving. Adam Smith (1759 [1790], section III.I.6) said
it best in The Theory of Moral Sentiments when he wrote,
To be amiable and to be meritorious; that is, to deserve love and to deserve
reward, are the great characters of virtue; and to be odious and punishable, of
vice. But all these characters have an immediate reference to the sentiments
of others. Virtue is not said to be amiable, or to be meritorious, because it is
the object of its own love, or of its own gratitude; but because it excites those
sentiments in other men.
Giving in order to be perceived in a favorable light by others is not lacking in virtue,
and in fact is far more virtuous than any other alternative.
Compared to private charity and redistribution, transfers to the poor done
through democratic means are almost certainly less virtuous. The replacing of voluntary
transactions with a compulsory democratic one transforms helping the poor from
a virtuous activity into an obligatory one. The self-reinforcing cycle where people give
because they seek the approbation of their fellow citizens is instead replaced by representative
democracy, where resources are often transferred to the middle class in
the name of helping the poor. While giving for purely altruistic reasons might be
nice to contemplate, we do not live in such a world and charities must appeal to selfinterest
as well as altruism in order to get donations. If it were possible to compare
the transfers to the poor under a completely private system of charity versus a more
public system, a completely private system would be superior because of closer link
between donation and outcome. Even if many United Way fundraisers are organized
by real estate agents and owners of car dealerships because it is good for business
(Hartford 2006), a less than ideal arrangement is superior to a world where people
have no incentive to become involved in private charity at all.¹¹
deadweight losses associated with raising tax revenue for alleviating poverty. In addition, government
solutions typically prevent the institutional experimentation that might lead to institutional
innovations (such as mutual aid societies) that found voluntary solutions to the problems associated
with poverty alleviation.
¹¹ Lee and McKenzie (1990) show that once private beneﬁts from charitable giving are taken into account,
the public good argument for redistribution is hard to justify.
Beaulier and Hall: Collective Consumption Externalities and Charitable Giving 33
Even if people gave out of entirely altruistic reasons and had interdependent utility
functions, however, government intervention is not warranted because there is no
misallocation of resources.¹² The collective consumption externality pointed out by
Hochman and Rodgers (1969) is a pecuniary externality, which does not lead to market
failure (Holcombe and Sobel 2000). In fact, pecuniary externalities are necessary
for competitive markets. To understand why pecuniary externalities do not lead to
economic ineﬃciency, it is important to better understand the distinction between
pecuniary and technological externalities.
Nearly every activity in markets generates eﬀects on third parties. There are two
types of third-party eﬀects (also called externalities): technological externalities and
pecuniary externalities. Technological externalities are externalities that directly affect
the production of the individual or ﬁrm. For example, suppose a factory moves
next door to a dry cleaning establishment and begins emitting pollution. The emissions
make it impossible for the dry cleaner to launder clothing while the factory is
operating. As a result of the externality, the proﬁts of the dry cleaner owner fall.
Pecuniary externalities, on the other hand, are third-party eﬀects transmitted
through prices. Instead of a factory moving next door to a town’s only dry cleaner, let
us suppose that another dry cleaning establishment quickly engaged the incumbent
dry cleaner in a price war. As competition occurs, the proﬁts of the incumbent dry
cleaner fall by the same amount as in the factory example. Clearly both pecuniary
externalities and technological externalities lead to welfare losses for the dry cleaner.
The policy implications for diﬀerent types of externalities vary. For example, technological
externalities can lead to market failure because resources are misallocated.
When factory owners do not take the eﬀect of their pollution into account, there may
be too much pollution. In the pecuniary externality case where a new dry cleaner
comes in next door, however, no resources were being misallocated and no ineﬃciency
resulted. In fact, the truth is exactly the opposite. Pecuniary externalities are
necessary for the eﬃcient operation of markets as they are the byproduct of compe-
tition.
¹² Reece (1979) ﬁnds, using the Bureau of Labor Statistics Consumer Expenditure survey, that people
do not have interdependent utility functions. His evidence is that there is no relationship between
charitable giving and the consumption level of potential recipients of said charity.
34 New Perspectives on Political Economy
Holcombe and Sobel (2000) take the distinction between technological and pecuniary
externalities from the production literature and extend it to the consumption
literature. In their analysis, consumption externalities, such as the case of interdependent
utility functions identiﬁed by Hochman and Rodgers (1969), are pecuniary
externalities because they do not directly aﬀect household production. Individuals
maximize their utility by allocating household resources to various activities based
on the shadow prices of those activities (Becker 1981). Given a ﬁxed set of inputs,
free-riding does not inﬂuence a household’s level of charitable giving. Instead, utility
falls because the actions of others lower the return associated with a particular set of
inputs. Just as intervention is not required because competition among dry cleaners
reduces economic proﬁts, neither is intervention required when the actions of others
reduce the utility a person receives from charity. Contrary to Hochman and Rodgers,
Holcombe and Sobel (2000) conclude that the Pareto-optimal public policy is to have
no government intervention.
To further illustrate this concept, imagine a world with two income groups – rich
and poor – both groups having interdependent utility functions a la Hochman and
Rodgers (1969). In this world, every rich person free-rides oﬀ the charitable giving of
other rich people and the level of charitable giving is zero. In such a world, the utility
of both the rich and the poor would be reduced because of the free riding. From the
perspective of Hochman and Rogers, government transfers from the rich to the poor
can make both parties better oﬀ.
This perspective, however, overlooks the distinction between technological and
pecuniary externalities. Placing it in the Holcombe and Sobel (2000) framework, the
larger charitable donation has merely changed the “price” of charitable giving, not the
amount of charitable activities that a household can undertake given the resources at
their disposal.¹³ A household charitable donation of $1000 produces $1000 of charity
regardless of whether other individuals decide to donate or not.¹⁴ The actions of
¹³ On the recipient side, recipients of charity have lower utility because they have less money (because
of reduced transfers), but their household production function is not aﬀected at all by the interdependent
utility function. For any given set of household inputs they can produce the same level of
outputs.
¹⁴ To make a clearer analogy to household production, a household purchasing $1000 of food receives
$1000 worth of groceries. The utility derived from that bundle of goods depends on the relative prices
of diﬀerent food stuﬀs in the household’s consumption bundle. An increase in the relative price of
Beaulier and Hall: Collective Consumption Externalities and Charitable Giving 35
others might aﬀect the utility received from giving, but from the standpoint of public
policy that change in utility is the result of a change in the relative price of giving and
thus is not Pareto-relevant.
This should not be taken to imply that pecuniary consumption externalities do
not have real eﬀects; they certainly do. Just as pecuniary externalities can reduce
a ﬁrm’s proﬁts they can also reduce a household’s utility. From the standpoint of
public policy, however, these harms are not policy relevant. In fact, they are necessary
for the eﬃcient operation of markets because it is pecuniary externalities that cause
ﬁrms and, we would argue, households to engage in entrepreneurship that that is the
wellspring of progress.
Just as the dynamic nature of the market process causes ﬁrms to come up with
new products and new ways of doing things, pecuniary externalities cause individuals
concerned for the poor to be “social entrepreneurs.” If pecuniary consumption
externalities reduce charitable donations, then those wanting to help the poor have
to take advantage of other facets of individuals’ utility functions, such as the warm
glow received from giving. Making high levels of donations status goods, for example,
links donors’ interests in being seen as someone who can aﬀord to give a lot with
the interests of the poor. The bundling of desired goods (such as public recognition
or exclusive access to social networks) with donations is another such innovation.
In ﬁnding new ways to align the self-interest of potential donors with the interests
of the poor, development oﬃcials help to expand the level of charitable giving in an
economy through experimentation. While government oﬃcials can raise taxes or reallocate
funds to the poor, they do not have the same ability to innovate and ﬁnd
solutions to the free-rider problem. Moreover, taking and redistributing does not
create a “culture of giving” where others give out of a desire for approbation (and
might, in fact, crowd out future giving).
groceries, to the extent it lowers household utility is not Pareto relevant because no resource misallocation
has occurred.
36 New Perspectives on Political Economy
4 Conclusion
This paper has examined a particular economic justiﬁcation often given for government
involvement in helping the poor. The lack of charitable giving may not be a
market failure at all. In addition, it is unclear that interdependent utility functions
are as rampant as some theorists suggest. Moreover, even if the collective consumption
characteristics of charity did lead to a reduction in charitable giving because of
interdependent utility functions, the undersupply of charitable giving would not be
economically ineﬃcient in the sense of misallocating resources. More importantly,
private charity and redistribution are more “robust” in the sense of ﬁnding ways to
align the incentives of potential donors and the poor. In doing so, free markets help
to foster sympathy for the less well-oﬀ and a culture of approbation for those who
help them. Government intervention, on the other hand, transforms a private, voluntary
transaction into a public obligation. This places a strong presumption in favor
of laissez-faire.
Ultimately, the diﬀerence between those who want government involved in charity
and redistribution and those opposed comes down to a belief in the individual’s
ability to help himself and his fellow man. Just because some people are diﬀerent
than others is not, ipso facto, an argument for paternalism, but, rather, an argument
for leaving people alone. Eﬀorts to force people to contribute to anti-poverty programs
takes away from the richness and general culture of society and inhibits trial,
error, and learning.
Those in favor of government involvement in poverty alleviation are guilty of making
the perfect the enemy of the good. This attitude regards almost all government
intervention as desirable and sees many areas where more government involvement
could improve market outcomes. But, it suﬀers from the same problem that all arguments
for big government encounter: it assumes other people – the experts – are
outside and above the economic order. The experts are said to be better at deciding
for everyone what’s best. By trying to make their values (or, to be more precise, the
values of the “rational actor”) the values everyone should live by, they are failing to
recognize the beauty of millions of people pursuing their own interests according to
their own values.
Beaulier and Hall: Collective Consumption Externalities and Charitable Giving 37
Government as a corrective to market failures is an inferior solution to alleviating
poverty problems when compared to the private sector. The crowding out of a
whole range of private charitable organizations – most of which were quite eﬀective
at dealing with “free riders” and deviant behavior – is one of the greatest, and largely
unaddressed, costs of big government. In the absence of government, we do not and
cannot know what speciﬁc mechanisms might arise to deal with the “free rider,” but
we do have some historical experiences where mutual aid arrangements worked quite
well (Beito 2000; Chalupníček and Dvořák 2007; Friedman 1962, 190-91).
Bibliography
[1] Andreoni, James. 1990. “Impure Altruism and Donations to Public Goods: A Theory
of Warm-Glow Giving.” Economic Journal vol. 100, no. 401 (June): 464-477.
[2] Andreoni, James, and Abigail Payne. 2003. “Do Government Grants to Private
Charities Crowd out Giving or Fund-Raising?” American Economic Review vol.
93, no. 3 (June): 792-812.
[3] Arrow, Kenneth J. 1972. “Gifts and Exchanges.” Philosophy and Public Aﬀairs vol.
1, no. 4 (Summer): 343-362.
[4] Beito, David. 2000. From Mutual Aid to the Welfare State. Chapel Hill, NC: University
of North Carolina Press.
[5] Becker, Gary. 1981. A Treatise on the Family. Cambridge, MA: Harvard University
Press.
[6] Boettke, Peter, and Peter Leeson. 2004. “Liberalism, Socialism, and Robust Political
Economy.” Journal of Markets and Morality 7 (1): 99-112.
[7] Buchanan, James. 2005. “Afraid to Be Free: Dependency as Desideratum.” Public
Choice 124: 19-31.
[8] Caplan, Bryan. 2007. The Myth of the Rational Voter. Princeton, NJ: Princeton
University Press.
38 New Perspectives on Political Economy
[9] Chalupníček, Pavel, and Lukáš Dvořák. 2007. “Giving as Investment: Towards
a Theory of State Interventions in Non-Proﬁt Sectors.” Paper presented at 2007
Mises Seminar in Sestri Levante, Italy.
[10] Clotfelter, Charles T. 1985. Federal Tax Policy and Charitable Giving. Chicago:
University of Chicago Press.
[11] Demsetz, Harold. “Information and Eﬃciency: Another Viewpoint.” Journal of
Law and Economics 10: 1-12.
[12] Dugger, Cecilia. 2006. “Gateses Give $47 Million to Bolster Coordinated Assaults
on Diseases.” New York Times, December 20.
[13] Friedman, Milton. 1962. Capitalism and Freedom. Chicago: University of Chicago
Press.
[14] Gergen, Mark P. 1988. “The Case for a Charitable Contributions Deduction.” Virginia
Law Review 74, no. 8 (November): 1393-14.
[15] Glazer, Amihai, and Kai A. Konrad. 1996. “A Signaling Explanation for Charity.”
American Economic Review 86, no. 4 (September): 1019-1028.
[16] Gruber, Jonathan, and Daniel M. Hungerman. 2005. “Faith-Based Charity and
Crowd Out During the Great Depression.” National Bureau of Economic Research
Working Paper 11332.
[17] Hartford, Tim. 2006. “Charity is Selﬁsh: The Economic Case Against Philanthropy.”
Slate, October 14. Online at: http://www.slate.com/id/2151244.
[18] Hayek, F.A. 1949. “The Intellectuals and Socialism.” In George de Huszar’s (Ed.)
The Intellectuals: A Controversial Portrait. Glencoe, IL: The Free Press, 371-84.
[19] Hochman, Harold M. and James. D. Rogers. 1969. “Pareto Optimal Redistribution.”
American Economic Review vol. 59, no. 4 (September): 542-557.
[20] Hochman, Harold M., and James D. Rodgers. 1977. “The Optimal Tax Treatment
of Charitable Contributions.” National Tax Journal vol. 30, no. 1 (March): 1-18.
Beaulier and Hall: Collective Consumption Externalities and Charitable Giving 39
[21] Keynes, John Maynard. 1936. The General Theory of Employment, Interest, and
Money. London: MacMillan.
[22] Kingma, Bruce R. 1989. “An Accurate Measure of the Crowd-out Eﬀect, Income
Eﬀect, and Price Eﬀect for Charitable Contributions.” Journal of Political Economy
vol. 97, no. 5 (October): 1197-1207.
[23] Landsburg, Steven. 1997. “Giving Your All: The Math on the Back of the Envelope.”
Slate, January 11. Online at: http://www.slate.com/id/2034/.
[24] Lee, Dwight R. and Richard B. McKenzie. 1990. “Second Thoughts on the PublicGood
Justiﬁcation for Government Poverty Programs.” Journal of Legal Studies,
vol. 19, no. 1. (January): 189-202.
[25] McGranahan, Leslie M. 2000. “Charity and the Bequest Motive: Evidence from
Seventeenth-Century Wills.” Journal of Political Economy vol. 108, no. 6 (December):
1270-1291.
[26] Mises, Ludwig von. 1983 [1944]. Bureaucracy. New York: Center for Futures Ed-
ucation.
[27] Reece, William S. 1979. “Charitable Contributions: New Evidence on Household
Behavior.” American Economic Review vol. 69, no. 1 (March): 142-151.
[28] Ribar, David C., and Mark O. Wilhelm. 2002. “Altruistic and Joy-of-Giving Motivations
in Charitable Behavior.” Journal of Political Economy vol. 110, no. 2 (April):
425-457.
[29] Rose-Ackerman, Susan. 1982. “Charitable Giving and ‘Excessive’ Fundraising.”
Quarterly Journal of Economics vol. 97, no. 2 (May): 193-212.
[30] Samuelson, Paul. 1947. Foundations of Economic Analysis. Cambridge, MA: Harvard
University Press.
[31] Samuelson, Paul. 1954. “The Pure Theory of Public Expenditure.” Review of Economics
and Statistics 36 (November): 387-389.
[32] Samuelson, Paul. 1961. Economics. New York: McGraw-Hill.
40 New Perspectives on Political Economy
[33] Smith, Adam. 1759 [1790]. The Theory of Moral Sentiments. London: A. Millar.
Sixth Edition.
[34] Sowell, Thomas. 2003. “The Vision of the Anointed.” Chattanooga Times FreePress,
April 30, page B7.
[35] Wohlgemuth, Michael. 2002. “Evolutionary Approaches to Politics.” Kyklos vol.
55, no. 2: 223-246.
ISSN 1801-0938
New Perspectives on Political Economy
Volume 4, Number 1, 2008, pp. 41 – 52
Rothbard’s Welfare Theory: A Critique
Michal Kvasnička
JEL Classiﬁcation: B49, B53, D60
Abstract: This paper analyzes Rothbard’s welfare theory. It diﬀers from other critiques
of the theory in that it does not analyze the context of the theory but only
its coherency and consistency with the rest of Rothbard’s analysis. It shows that the
theory is invalid and should be rejected because it is incoherent (its two welfare theorems
cannot be defended at the same time), and inconsistent with Rothbard’s own
claims made elsewhere.
Michal Kvasnička – Masaryk University, Faculty of Economics and Administration,
qasar@econ.muni.cz.
42 New Perspectives on Political Economy
1 Introduction
There were many attempts to create an economic welfare theory to allow an economist
to decide what action is beneﬁcial from the point of view of the society as a whole
with no reliance on exogenous ethical norms. Such attempts were in vain because of
the fundamental limits of the utility theory on which they were based – it is possible
neither to compare the utility interpersonally, nor aggregate it. The only rule to
distinguish “good” from “wrong” is then Pareto’s Unanimity Rule: increasing ‘social
utility’ (to use Rothbard’s term) is such a change that makes at least one person better
oﬀ, and no one worse oﬀ judged by their own preferences; decreasing social utility is
a change that makes at least one person worse oﬀ, and no one better oﬀ judged by
their own preferences; in the remaining cases we can say nothing about the change
of social utility. The problem with this analysis is that almost all changes that may
happen fall into the indeterminate category.
In his Reconstruction of Utility and Welfare Economics Rothbard (1956) ﬁrst criticized
the previous attempts to create the economic welfare theory, and then presented
his own version. His critique is sound and decisive, but his attempt must be
rejected, however strongly we are sympathetic to the philosophy he wanted to defend.
There are four reasons for this: 1) his approach is inconsistent with Pareto’s
rule (while claiming the opposite), 2) it is inconsistent with common sense, 3) it is
incoherent (its two welfare theorems cannot be defended at the same time), and 4) it
is inconsistent with Rothbard’s own claims made elsewhere. Thus Rothbard’s welfare
theory is invalid, and should be rejected. Otherwise it may happen that sound ideas
would be ridiculed because they are not defended on the grounds of a sound analysis
but on the grounds of the invalid welfare theory, or be even rejected by those who
believe in this invalid dogma.¹
¹ Based on Rothbard (1979) some people believe that Rothbard withdrew his welfare theory since he
said here that “we cannot decide on public policy, tort law, rights, or liabilities on the basis of eﬃciencies
or minimizing of costs. But if not costs or eﬃciency, then what? The answer is that only
ethical principles can serve as criteria for our decision”, i.e. that no purely economic norm (the welfare
theory) can distinguish “right: from “wrong”. However, Rothbard did not withdraw his theory
here. He was merely aware of many reasons why an “increase of social utility” is not suﬃcient criteria
for an ethical judgment. He said it plainly even in his older work (1956, pp. 40–41). Only three years
before Rothbard (1976) published a paper where he used both arguments together. He claimed that
praxeology cannot settle ethical judgments (p. 78), and then used his welfare theory (pp. 87–89).
Kvasnička: Rothbard’s Welfare Theory: A Critique 43
Our critique diﬀers signiﬁcantly from other critiques of Rothbard’s welfare theory
(Caplan, Cordato, Gunning and others). They analyze the broad context of the theory,
or even question Austrian economics as such (like Caplan 1999, does). Our analysis is
based on a detailed study of Rothbard’s own arguments only. We have only checked
the coherency of his arguments.
The structure of the paper is: section 2 deals with Rothbard’s methodology, especially
with the limits and implications of his concept of demonstrated preference. Section
3 shows that Rothbard’s welfare theory is incoherent because each of its two theorems
is derived under a diﬀerent and mutually exclusive set of assumptions, i.e. they
cannot hold true at the same time. Section 4 analyzes one of potentially many inconsistencies
between Rothbard’s welfare theory and the rest of his own analysis. Our
arguments are summarized in section 5.
2 Rothbard’s Methodology: Demonstrated Preference
Rothbard (1956) said explicitly that every welfare theory must be based on Pareto’s
Unanimity Rule; he then criticized older welfare theories for not satisfying this condition
(pp. 23n). However, Rothbard was also able to see that Pareto’s rule is very
strict, and that almost no change is either Pareto-improving, or Pareto-worsening.
He gave the example of envy: let us assume that two men make a mutually beneﬁcial
exchange that enriches both of them while it makes no one else worse oﬀ in terms of
goods and services he can consume now or later. Such a trade could be called increasing
social utility. However, if there is one envious person that feels worse oﬀ because
of the success of his neighbors, Pareto’s rule rejects the case as “indeterminate” since
one (envious) person is worse oﬀ (pp. 28–29).
Rothbard believed he could get a determinate solution in more cases with the tool
he called demonstrated preference.² The idea is simple: human action is purposeful,
i.e. a man acts to reach his own goals. He chooses an action because he believes that
² Gunning (2004) shows that Rothbard is wrong when he attributes this concept to Mises. Both Rothbard’s
method and its outcome (the welfare theory) diﬀer signiﬁcantly from Mises’ own ones. The
diﬀerence is clear also from Rothbards’s own critique of Mises in Rothbard (1976, pp. 90–98).
44 New Perspectives on Political Economy
the action will improve his well-being in comparison to hypothetical situations in
which he would have chosen other actions (including no action). Thus we may infer
that the action demonstrates an increase of the agent’s utility. It demonstrates he is
better oﬀ in terms of his own preference in comparison to outcomes of other actions
he might have chosen but did not. The concept of demonstrated preference is then,
according to Rothbard, “simply this: that actual choice reveals, or demonstrates, a
man’s preferences; that is, that his preferences are deducible from what he has chosen
in action.” (p. 2)
The notion that the action demonstrates a utility improvement is not new (Rothbard
quotes Fisher, p. 3). What is new is that Rothbard proposes to ignore all changes
of a person’s utility except those the person demonstrates through his actions. Rothbard
rejects from analysis everything which is not demonstrated in an actual action,
i.e. what goes beyond the scope of the demonstrated preference, as a vain psychology
(pp. 13–14).³ We can read this in two ways: 1) we can know nothing that was not
demonstrated in an action, or 2) there is nothing more than what was demonstrated
in an action. While Rothbard might have the ﬁrst in mind, he spoke as if he believed
the second. This can be seen from his assertion that a man cannot be indiﬀerent.
This assertion is a corollary of the concept of demonstrated preference. Rothbard
argues thus: since each action is unique, a person can choose only one choice,
i.e. there is no way to demonstrate indiﬀerence; hence there can be no indiﬀerence at
all (p. 15). There must always be some preference even if it is established by chance
(p. 16). Rothbard then rejects the indiﬀerence curves approach on these grounds.
However, Rothbard is clearly wrong. The inability to demonstrate indiﬀerence is no
proof there is no indiﬀerence but only that an outside observer cannot observe it,
which is quite a diﬀerent thing. So there is not a fundamental diﬀerence between the
indiﬀerence curves approach and Rothbard’s own utility theory as long as the former
one yields a unique solution. The indiﬀerence curves just describe the agent’s inner
world, in which Rothbard takes no interest, or rather denies its existence altogether.
³ It is really strange that Rothbard limits knowledge an economist can have to what was demonstrated
in an action, but does not limit other social sciences in this way. Rothbard (1976, pp. 88–89) says
that “we may know as historians from interpretative understanding of the hearts and minds” more
than we can know as “scientiﬁc economists”. “Scientiﬁc economists” should “conﬁne the concept to
its strict scientiﬁc compass in demonstrated preferences”. However, he does not say why a “scientiﬁc
historian” should know more than a “scientiﬁc economist”.
Kvasnička: Rothbard’s Welfare Theory: A Critique 45
However indiﬀerent a man may or may not be, Rothbard’s critique of the indiﬀerence
analysis shows how he means to use the concept of demonstrated preference.
In the same way in which Rothbard denies any relevance (or perhaps existence)
of the inner states of man, he denies the possibility of a comparison of the two situations.
It is obvious from his defense of the corollary of impossibility of indiﬀerence.
He argues thus: one might object that the indiﬀerence may be demonstrated by a
repeated choice. If a person in the same circumstances chooses in half the cases one
action, while in the other half another action, we may assume he is truly indiﬀerent,
and chooses one of two actions by chance. However, Rothbard argues, such a comparison
is not possible because over time the agent’s preferences might have changed
(p. 16). Therefore, we cannot (according to Rothbard) compare a real situation with
another hypothetically the same, but we have to stick to the demonstrated preference
which shows the action the agent has really chosen.
To summarize, Rothbard’s methodology consists of three elements: Pareto’s rule,
the demonstrated preference (and its corollary of the impossibility of indiﬀerence),
and of the impossibility of the comparison of situations at diﬀerent times (which may
truly be seen as another corollary of the concept of demonstrated preference).
The demonstrated preference concept looks simple but it poses severe limits that
may be easily overlooked. An agent can demonstrate only those changes of his utility
that are caused by his own actions, i.e. when he is active. There is no way a passive
agent may demonstrate a change of his utility caused by an external force he passively
suﬀers. Moreover, an action only demonstrates that the agent is better oﬀ choosing
the action in comparison to choosing another possible action in his situation, not
that he actually is better oﬀ.⁴
Let us illustrate the point. For instance, I cannot demonstrate that I am better or
worse oﬀ when I am given a gift. It is a situation that happens to me – I am passive in
it. The gift may considerably change my utility; however, there is no action that could
demonstrate this. It may seem that my acceptance of the gift is proof that the gift has
increased my utility (otherwise I would have rejected it), but it is not so. Rejecting a
⁴ Herbener (1997) in his defense of Rothbard’s welfare theory can clearly see these points, and he regards
it a positive virtue of the theory. However, he completely fails to notice that this destroys the coherency
of the theory. See below.
46 New Perspectives on Political Economy
gift is something quite diﬀerent from not being given it, as everyone knows who was
given an ugly present by someone whose feelings he does not want (or dare) to hurt.
In the same way, my non-resistance to a robber does not prove I enjoy being robbed – I
simply may not be bold enough to resist, i.e. I prefer no action to resistance. Nor does
resistance to the robber prove I do not like being robbed – I may like ﬁghting, and
could have come to a dangerous place to challenge it. In other words, every subject
chooses the most preferred action in any given situation; but there is no way he could
demonstrate how much he prefers the situation that happened as such.
This means that the concept of demonstrated preference is not compatible with
the usual meaning of Pareto’s rule, because it allows us to look at only those utility
changes that can be demonstrated with an action, while Pareto’s rule looks at all of
them. The usual meaning of Pareto’s rule is God-like – the analyst pretends to know
everything that is happening and how it aﬀects all agents. Obviously, such an approach
is scientiﬁcally valid (even though it could easily be abused by someone who
pretends he really is omniscient, as governments sometimes do). Rothbard goes to
the opposite extreme. He says we can know (even in principle) nothing that cannot be
demonstrated by an action. This is extreme behaviorism, as Caplan (1999, pp. 825–6)
rightly says.
To summarize, while Rothbard said that any economic welfare theory must satisfy
Pareto’s rule, he did not actually follow his own recommendation. The concept of
demonstrated preference is a subtle, yet crucial, change of the rules of the game.
This change is both conscious and purposeful.
3 Rothbard’s Welfare Theory: Analysis of Two Theorems
Rothbard’s welfare theory consists of two major theorems. The ﬁrst theorem says
that “the free market always increases social utility” (p. 31). This theorem is derived
(pp. 28–30) as a generalization of the old doctrine that both traders beneﬁt from a
voluntary exchange, which is in turn a special case of the doctrine which inspired the
concept of demonstrated preference that a man acts to beneﬁt. The traders exchange
to beneﬁt, i.e. the utility of each of them is increased by the exchange, at least ex-ante.
Kvasnička: Rothbard’s Welfare Theory: A Critique 47
To claim that the exchange increases social utility according to Pareto’s rule, one
has to prove that there is no losing subject. But this does not hold true in general. A
voluntary exchange between two or more partners (as any unilateral or multilateral
action in general) may have various impacts on a non-involved third party. It may
aﬀect it as 1) common physical externalities, 2) a redistribution of wealth or income
either through a non-voluntary transfer (say a robbery), or change in prices, 3) envy,
etc. If any non-involved third party loses because of an action from which someone
else beneﬁts, the action (or the process that generates it) cannot according to Pareto’s
rule increase social utility.
Rothbard made a couple of subtle claims that (as he believed) allowed him to ignore
these eﬀects. First, he said the negative externalities can be ignored, because
they cannot occur in the free market because “[t]hese ‘problems’ are due to insuﬃcient
defense of private property against invasion. Rather than a defect of the free
market, therefore, they are results of invasion, of property, invasions which are ruled
out of the free market by deﬁnition” (p. 38, footnote 74). The externalities are indeed
caused by ill-speciﬁed property rights, but this does not allow us to ignore them.
Since the deﬁnition and enforcement of property rights is costly, they cannot be complete
in the real world because complete property rights do not pay oﬀ and are often
not technically viable at the given state of technology. To say that the externalities are
“ruled out of the free market by deﬁnition” means either that the free market cannot
exist, or that Rothbard’s theory is of no relevance to the real world.
Rothbard also has an (invalid) answer to the problem of redistribution of wealth
and income (pp. 29–30). He rightly says that distribution is not independent from
production and exchange in a free market economy, and that a change in the distribution
is caused by voluntarily-made exchanges. However, this is not a suﬃcient
reason to ignore the impact of a change of the income and wealth distribution on
individual’s utilities under Pareto’s rule. Pareto’s rule looks for all subjects’ utility
increases or decreases in comparison to the status quo, despite their cause.
These lines of defense of the theorem are both invalid, and redundant; moreover,
they do not apply to the case of envy quoted above. There is a very simple reason why
Rothbard can ignore all these people’s losses on the basis of his concept of demonstrated
preference: the trading partners demonstrate by their actions that they beneﬁt
48 New Perspectives on Political Economy
by the trade (they must beneﬁt because possible indiﬀerence was ruled out). On the
other hand, the possibly losing non-involved third party cannot demonstrate it loses
because it loses passively – there is no action that can prove it (see section 2). Thus
under the concept of demonstrated preference the market exchange, and hence the
market process itself, increases social utility because at least one party beneﬁts, and
no one can prove he loses.⁵
Such a conclusion is possible, but it corresponds neither to the usual meaning of
Pareto’s rule, nor to common sense. Moreover, it proves too much. Under such logic,
any action, even a murder, increases social utility since the active agent proves by his
action he beneﬁts while the potential suﬀering party cannot prove it loses. Rothbard
himself did not notice this ﬂaw in his theory – perhaps because he did not use the
concept to prove obvious nonsenses. Some of his followers were more consistent
and went further. They even found “mistakes” in Rothbard’s own analysis (see next
section).
The second theorem of Rothbard’s welfare theory says that “no act of government
can ever increase social utility” (p. 31). Rothbard argues (pp. 30–32) that any governmental
action makes someone better oﬀ – at least the government beneﬁts since otherwise
it would not do it (the action demonstrates the preference). On the other
hand, the government’s action coerces someone else to do what he does not want
to do, or to abstain from an action he would like to do. This person loses. Thus
such an action does not increase social utility according to Pareto’s rule. This holds
true for every government action since every action, however beneﬁcial for anyone,
is ﬁnanced by taxes collected on an involuntary basis.
Rothbard’s argument seems to be correct, but it is inconsistent with his own
methodology because he uses Pareto’s rule in its usual meaning now, not the concept
of demonstrated preference. This is so because under demonstrated preference
the losing party cannot prove it loses because it suﬀers the loss passively (see above).
⁵ This is the purpose for which the concept of demonstrated preference was created. Rothbard (1976)
says: “Therefore, if we employ the Paretian deﬁnition of ‘social utility’ in the usual ‘psychologizing’
meaning, we can say nothing about social utility one way or the other. But if we conﬁne the concept to
its strict scientiﬁc compass in demonstrated preference, then we can state that social utility increases
from the exchange.”
Kvasnička: Rothbard’s Welfare Theory: A Critique 49
Herbener (1997, pp. 103–104) states it clearly: “It is by this inference [i.e. deduction]
and not his action under duress … that the Pareto-Inferior nature of involuntary interaction
is seen.” In other words, Rothbard abandoned the concept of demonstrated
preference here, and he used Pareto’s rule in its usual “psychologizing” meaning instead.
This is the incoherence of the theory.⁶
The result is that the ﬁrst and second theorem of the welfare theory cannot hold
at the same time. Either we use Pareto’s rule in its usual meaning, which yields the
second theorem but rules out the ﬁrst one, or we use the concept of demonstrated
preference, and then we can retain the ﬁrst theorem, but not the second one. Moreover,
as we have seen above, this later approach is inconsistent with both Pareto’s rule
and common sense. We will see in the next section it is inconsistent with Rothbard’s
own claims made elsewhere too.
The incoherency of Rothbard’s welfare theory can be illustrated with the case of a
cartel which Rothbard himself used (p. 34). He ﬁrst argued that a cartel which came
to existence on a voluntary basis increases social utility (the ﬁrst welfare theorem).
This is so because the members of the cartel beneﬁt (they can charge a higher price),
and they demonstrate it by an action – the formation of the cartel. The consumers
have to pay the higher price, but they cannot demonstrate they lose by any action,
and hence Rothbard ignores them, and calls the change Pareto-improving. Then he
argued that government prohibition of the cartel does not increase social utility (the
second welfare theorem). This is so because someone beneﬁts from the destruction of
the cartel. Surprisingly those who beneﬁt are not the customers because they cannot
demonstrate they beneﬁt with any action since they are passive, but the government
which acts, and its action demonstrates it beneﬁts. But in the same time the producers
lose. They can charge only competitive prices again. Rothbard says the prohibition
of the cartel “demonstrably injures them”. But this is not so. There is no action by
which the producers could demonstrate they lose – in precisely the same way as the
consumers could not demonstrate they lose from the cartel formation. Thus Roth⁶
Herbener is wrong when he calls the involuntary interaction ‘Pareto-Inferior’ since it is only indeterminate.
Rothbard (1976, p. 89) says it correctly. This also means that even if Rothbard’s welfare theory
was correct (which it is not), it would be a very weak basis for a critique of governmental meddling
with the economy, as Caplan (1999) shows.
50 New Perspectives on Political Economy
bard used the concept of demonstrated preference to prove the ﬁrst statement, and
Pareto’s rule to prove the later one – but these two concepts are incompatible and
inconsistent, see section 2.
4 Contradiction Between Rothbard’s Welfare and Economic Theories
Rothbard’s Welfare theory is not only incoherent, but it is clearly inconsistent with
Rothbard’s own economic theory. One example has been found by Barnett and
Block (2004): Rothbard in his works held the position that “[a]n increase in the supply
of money confers no social beneﬁt whatever” (quoted in Barnett and Block 2004,
p. 42). The same position has been defended by many economists from Hume’s days
on, including Mises and other Austrian economists.
Barnett and Block (2004) try to criticize this theory. Their analysis is more comprehensive
(and in my opinion entirely wrong). What matters to us is the fact they
use Rothbard’s welfare theory to uproot his monetary theory (pp. 44–46). They distinguish
between the ﬁat money produced by the government on a coercive basis,
and the gold standard money produced by free entrepreneurs on a voluntary basis.
In Rothbard’s monetary economics, all types of money (in this context) are treated in
the same way: the increase of the stock of money of whatever type does not increase
social utility. But according to Rothbard’s Welfare Theory, Barnett and Block quite
rightly argue, the impact of the two types of money must be absolutely diﬀerent. The
extra ﬁat money is produced on a non-voluntary basis, i.e. the second welfare theorem
applies, and social utility is not increased. The extra gold money is produced on a
voluntary basis, i.e. the ﬁrst welfare theorem applies, and social utility must increase.
We have here three problems: ﬁrst, there is inconsistency in Rothbard’s own
claims, i.e. inconsistency between his monetary and his welfare theory. Secondly,
the impact of a money injection into the economy depends neither on the quantity
of money injected, nor on the way it is injected into the economy, but on its producer.
Thirdly, the inconsistency analyzed in the previous section appears again: either we
stick to Pareto’s rule, and then no injection of money increases social utility since
some people lose (they lose a part of their purchasing power, as Rothbard knew very
well), or we stick to the concept of demonstrated preference, and then any injection
Kvasnička: Rothbard’s Welfare Theory: A Critique 51
of money increases social utility, since no one can prove by any action he loses. We
can expect that more similar inconsistencies exist.
5 Conclusion
While Rothbard’s critique of older welfare theories is sound and persuasive, his own
welfare theory failed. It is incoherent because its ﬁrst and the second welfare theorems
cannot hold true at the same time, inconsistent with Pareto’s unanimity rule
while it claims it is consistent with it, and inconsistent with common sense. Moreover,
there is at least one inconsistency between Rothbard’s welfare theory and sound
economic doctrine – the doctrine he himself held. This means that the maintenance
of Rothbard’s welfare theory may endanger sound propositions of common economic
theory, as the case quoted in section 4 shows. For these reasons we propose to reject
it.
Bibliography
[1] Barnett II, W., and W. Block. 2004. “On the Optimum Quantity of Money” in
The Quarterly Journal of Austrian Economics, vol. 7, no. 1, Spring, pp. 39–52.
[2] Caplan, B. 1999. “The Austrian Search for Realistic Foundation” in Southern
Economic Journal, 65, 4, April, pp. 823–828.
[3] Gordon, D. 1993. “Toward a Deconstruction of Utility and Welfare Economics”,
The Review of Austrian Economics, 6, 2, pp. 99–112.
[4] Gunning, J .P. 2004. “Austrian Welfare Economics?: A Misesian Response”,
available at http://knight.fcu.edu.tw/~gunning/.
[5] Herbener, J. M. 1997. “The Pareto Rule and Welfare Economics”, Review of Austrian
Economics, 10, 1, pp. 79–106.
[6] Rothbard, M. N. 2002 [1956]. “Toward a Reconstruction of Utility and Welfare
Economics”, originally published in On Freedom and Free Enterprise: The Eco-
52 New Perspectives on Political Economy
nomics of Free Enterprise, May Sennholz, ed., Princeton; quoted from the electronic
version available at The Mises Institute.
[7] Rothbard, M. N. 1973. “Value Implications of Economic Theory”, The American
Economist, 17, Spring, pp. 35–39.
[8] Rothbard, M. N. 1997 [1976]. “Praxeology, Value Judgement, and Public Policy”,
originally published in The Foundations of Modern Austrian Economics, Edwin
Dolan, ed., Sheed and War, pp. 89–111; quoted from the electronic version available
at The Mises Institute.
[9] Rothbard, M. N. 1997 [1979]. “The Myth of Eﬃciency”, originally published in
Time, Uncertainty, and Disequilibrium, Mario Rizzo, ed., DC Heath, pp. 90–95;
quoted from the electronic version available at The Mises Institute.
ISSN 1801-0938
New Perspectives on Political Economy
Volume 4, Number 1, 2008, pp. 53 – 78
In Defense of Rothbardian Welfare Economics
Jeﬀrey M. Herbener
JEL Classiﬁcation: D60, B25
Abstract: Murray Rothbard oﬀered a twofold defense of the unhampered market
economy. He argued that it upholds natural rights and promotes the life and ﬂourishing
of persons more fully than any other social system. His welfare economics
was not a defense of the unhampered market economy, but a critique of mainstream
economists’ use of welfare economics to promote state intervention. He argued that
mainstream economists accept the subjectivity of value by adhering to the Pareto rule
in welfare economics. In doing so, however, they must embrace all the implications
of the subjectivity of value and not just the impossibility of interpersonal utility comparisons.
They must accept also the principle of demonstrated preference. Recasting
welfare economics along the lines of demonstrated preference reorients it away from
artiﬁcial models of the economy based on ﬁctitious economic agents toward the real
economy of actual human persons. Rothbard’s demonstration that laissez-faire renders
the greatest social welfare in the real economy is shown to be valid.
Jeﬀrey M. Herbener – Professor of Economics, Grove City College, jmherbener@gcc.edu.
54 New Perspectives on Political Economy
I begin by observing that you cannot ﬁnd out what a man means by simply studying
his spoken or written statements … In order to ﬁnd out his meaning you must
also know what the question was … to which the thing he has said or written was
meant as an answer.¹
1 Rothbard’s Defense of the Unhampered Market
Murray Rothbard’s defense of the unhampered market economy is based on the philosophy
of natural law and the ethics of natural rights. Natural law holds that reason
can discover the characteristics, or natures, of the diﬀerent entities in the world and
how these entities behave and interrelate, i.e., natural laws. The natural law ethic is
that the good for each entity is what fulﬁlls its nature. Man, too, has a nature and
thus, reason can determine what is good for man. “The natural law,” Rothbard wrote,
“elucidates what is best for man – what ends man should pursue that are most harmonious
with, and best tend to fulﬁll, his nature.”²
Salient among the characteristics that constitute man’s nature is reason itself. Human
action is guided by reason. By it, man learns which ends satisfy him more fully
and by it, he identiﬁes the causal connections between things in the world and his
ends. In other words, reason discovers means. To act, man must do more than merely
recognize the existence of means; he must control the use of, in other words, own,
means he deems helpful in attaining his ends. He naturally owns his mind and body,
but man cannot fulﬁll his nature by attempting to act with labor alone. He must
produce to sustain himself through consumption. Production requires that he appropriate
natural resources and use them with his labor to produce goods. To fulﬁll
his nature, man must have such natural or personal ownership of property. He must
own his labor, the natural resources he homesteads, and the goods he produces.
Man exists as individual human persons each possessing the same nature. If each
person lived in isolation, then each one would act with his own personal property
and there would be no question of one person interfering in the natural development
of another. When persons interact, however, one may aggress against the per¹
Collingwood (1939, p. 31); quoted in Gordon (1993, pp. 104-105).
² Rothbard (1982, p. 10).
Herbener: In Defense of Rothbardian Welfare Economics 55
sonal property of another and thereby, disrupt another person’s natural process of
life. Rights each person has to his personal property protect each person in the fulﬁllment
of his nature. Natural rights declare it immoral for other persons to aggress
against a person by using or threatening to use violence to interfere with his acquisition,
use, or disposal of his naturally-owned property. The rights of private property
include unilateral and bilateral transfer of ownership between willing persons. Bilateral
transfers or voluntary exchanges constitute the unhampered market economy.
The hampered market economy grants legal privileges of some persons to aggress
against the private property of other persons and the command economy denies the
natural rights of persons altogether. Rothbard’s main argument in favor of the unhampered
market economy is that it upholds each person’s natural rights of private
property and thereby, achieves the ethical use of interpersonal violence more fully
than another other social arrangement.
The role economics plays in Rothbard’s defense of the unhampered market is subsidiary
to that of ethics. Rothbard wrote, “[Superior productivity is a] … result of the
free market, but it is not, to the libertarian, the prime reason for his support of the
system That prime reason is moral and is rooted in the natural-rights defense of private
property we have developed above.”³ While natural law philosophy delineates
the natural rights by which man upholds the ethical use of interpersonal violence in
society, economics delineates the natural laws by which man produces, exchanges,
and consumes in society.
Economics demonstrates that the unhampered market promotes human life and
ﬂourishing to a greater extent than any other social arrangement. The diversity of
labor skills among persons and the diversity of productivity of natural resources in
the world imply the greater productivity of the division of labor. The division of labor
furthers the life and ﬂourishing of man by sustaining a vastly larger number of
persons at far greater levels of material wellbeing than other social arrangements and
by making a place in production suited to each person’s talents and personality. For
the division of labor to advance beyond a primitive level and continue to progress
there must be a market economy. The monetary prices that emerge from the voluntary
exchange of private property can be used by entrepreneurs to make economic
³ Rothbard (1978, p. 40). Italics original.
56 New Perspectives on Political Economy
calculations of the net income of producing goods and the net worth of acquiring assets.
Within the framework of net income, entrepreneurs can determine which goods
people in society value more highly and which combinations of factors of production
used to produce the higher-valued goods they value less highly. By comparing the
anticipated revenues from selling goods to the anticipated costs of hiring the factors
of production, entrepreneurs can economize production in the division of labor.
Within the framework of net worth, entrepreneurs can determine which lines of capital
investment people in society value more highly and the liabilities they associate
with each line. By comparing the anticipated asset value of the investment with its
anticipated liability value, entrepreneurs can economize the construction of the capital
structure. The command economy eliminates economizing the division of labor
for people in society by abolishing economic calculation and the hampered market
economy impairs economizing by distorting economic calculation.⁴
Rothbard’s achievement was to show that these two arguments, the ethical and
economic, are grounded on the same praxeological foundation. Both are constructed
from the nature of man and human action. This approach has been called a “science
of liberty” because both its parts are objectively established. It was Rothbard’s
acceptance of an objective ethical system that was the basis of his critique of Ludwig
von Mises’s defense of the unhampered market economy. He embraced Mises’s
economic argument, but chided him for his inadequate ethical claims for advocating
laissez-faire.⁵ As Rothbard wrote:
Thus, while praxeological economic theory is extremely useful for providing
data and knowledge for framing economic policy, it cannot be suﬃcient by itself
to enable the economist to make any value pronouncements or to advocate any
public policy whatsoever. More speciﬁcally, Ludwig von Mises to the contrary
notwithstanding, neither praxeological economics nor Mises’ utilitarian liberalism
is suﬃcient to make the case for laissez-faire and the free market economy.
To make such a case, one must go beyond economics and utilitarianism to establish
an objective ethics which aﬃrms the overriding value of liberty, and morally
condemns all forms of statism…⁶
⁴ Mises (1998, pp. 211-232, 258-311, 694-711, and 736-773).
⁵ Rothbard (1982, pp. 205-212).
⁶ Rothbard (1982, p. 212).
Herbener: In Defense of Rothbardian Welfare Economics 57
Kvasnička incorrectly juxtaposes Rothbard’s welfare economics against Mises’s approach.
He writes, “Both Rothbard’s method and its outcome (the welfare theory) differ
signiﬁcantly from Mises’ own ones.”⁷ But Rothbard accepted and defended Mises’s
praxeological method and the economic theory Mises built by it as vigorously as anyone.
He saw his own role as adding to Mises’s theoretical ediﬁce. In the preface to his
own economic treatise, Rothard wrote:
Professor Paul Samuelson has written rhapsodically of the joy of being under
thirty at the time of publication of Keynes’ General Theory. I can say the same
for the publication of Ludwig von Mises’ Human Action in 1949. For here at last
was economics whole once more, once again an ediﬁce … Suﬃce it to say that
from now on, little constructive work can be done in economics unless it starts
from Human Action … In one sense, the present work attempts to isolate the
economic, ﬁll in the interstices, and spell out the detailed implications, as I see
them, of the Misesian structure … it is my hope that this work may succeed in
adding a few bricks to the noble structure of economic science that has reached
its most modern and developed form in the pages of Human Action.⁸
2 Welfare Economics and the Defense of the Market
Rothbard’s welfare economics, then, was not an alternative to Mises’s economic defense
of the unhampered market, which Rothbard accepted. It was, instead, a critique
of mainstream welfare economics. It was directed at mainstream economists, a very
limited audience that held a narrow set of presuppositions and thought about social
welfare within a restrictive framework. Rothbard’s welfare economics was not an argument
aimed at the general public, which would fail to grasp it. For that audience,
he wrote, For a New Liberty. Rothbard’s welfare economics was not even aimed at academicians
in general, who would see its shortcomings all too well. For that audience,
he wrote, The Ethics of Liberty. Given the limited task Rothbard was trying to accomplish
and the narrow audience he was trying to reach with his welfare economics, the
import of Kvasnička’s criticism of him loses its force. He writes,
⁷ Kvasnička (2008, p. 43, fn. 2).
⁸ Rothbard (1970, p. xi). Italics original.
58 New Perspectives on Political Economy
Thus Rothbard’s welfare theory is invalid, and should be rejected. Otherwise
it may happen that sound ideas would be ridiculed because they are not defended
on the grounds of a sound analysis but on the grounds of the invalid welfare
theory, or be even rejected by those who believe in this invalid dogma.⁹
Presumably by “sound ideas” Kvasnička means laissez-faire or the unhampered market
economy. But, as shown above, Rothbard’s arguments for the unhampered market
economy were that it upholds natural rights and fosters the life and ﬂourishing of
man. Even if his welfare theory were invalid, then, it would matter little to Rothbard’s
defense of “sound ideas.”¹⁰
Rothbard’s reconstruction of welfare economics attempted to show mainstream
economists that the only sound basis for both utility and welfare economics is the
fundamental principles of human action. To the extent that mainstream economists
were dissatisﬁed with the disarray in which both ﬁelds found themselves in 1956,
Rothbard was oﬀering a corrective. As he wrote:
Both theories have lately been ﬂoundering in stormy seas. Utility theory is
galloping oﬀ in many diﬀerent directions at once; welfare theory, after reaching
the heights of popularity among economic theorists, threatens to sink, sterile
and abandoned, into oblivion.
The thesis of this paper is that both related branches of economic theory can
be salvaged and reconstructed, using as a guiding principle of both ﬁelds the
concept of “demonstrated preference.”¹¹
⁹ Kvasnička (2008, p. 42).
¹⁰ Rothbard himself lamented just such confusion about the role of welfare economics in his defense of
the unhampered market economy. Edward Stringham cites an audio tape of a speech Rothbard made
in a lecture series “A Short Course on Free Market Economics,” on which Rothbard states:
Unfortunately … it has been maintained that my whole basis for laissez-faire rests
on this whole social utility nonsense. Of course it really doesn’t. It’s all really a gimmick
to show that if you really go along with this whole Pareto-optimality-social-utility then
you have to conﬁne yourself to laissez-faire. It’s not my major argument for laissezfaire.
At any rate, the trouble with those people who think it’s my major argument are
so inamorate [sic] that that’s all they can focus on. (Rothbard, Tape 6, “Cost of the Firm”
Side B, 35:57 to 37:44 in Stringham 2007, p. 27)
¹¹ Rothbard (1997a, p. 211).
Herbener: In Defense of Rothbardian Welfare Economics 59
Rothbard began by attempting to establish the proper meaning of the concept of
preference.¹² He argued that the meaning of preference can be found in its role in
human action. Action is purposeful behavior. Given the circumstances that make
up the situation in which a person ﬁnds himself, his action aims at attaining an end.
The situation of acting contains both general conditions, elements of the situation a
person does not control in his action, and means, elements of the situation he does
control. Action is using means in an attempt to attain an end. For something to be
a means, the person must own it, i.e., control the use of the thing in action. Because
of the scarcity and diversity of means, action requires choice. A person must choose
among competing ends to pursue with each set of means and among competing sets
of means capable of attaining each end. Preference is the judgment a person makes
in ranking the value of alternatives in action.
There are several distinguishing characteristics of preference. Preference is the
logical requisite of choice, which is necessary for action. If a person could not establish
a preference in his mind, then he could not choose and therefore, could not act.
Choosing only requires establishing a rank order of alternatives; one option is better
another not as good. Preference does not measure value. It grades it. Preference
is established within the circumstances of action, both the psychological conditions
internal to the mind and the means and general conditions external to the mind. Preference
does not refer to a hypothetical state of aﬀairs in which the actor achieves bliss
or happiness or even a state of aﬀairs that he likes. It is not wishful thinking about
what one would do if conditions were more favorable. Preference is not a stipulation
made by the economist about a person’s state of mind. Preference refers to the rank
order of more preferred and less preferred alternatives that a person makes in his
mind as the basis for choosing in action. It incorporates, therefore, all possible cases
of psychological disposition. Preference, being a state of mind, is known experientially
only by the person acting and no one else. Only he can know the level of utility
he enjoys from attaining his higher-valued alternative and how much more value he
gets from it than from his lower-valued alternative. Another person’s knowledge of
the preferences of a person acting is limited to the evidence empirically available from
¹² On these basic principles, see Rothbard (1970, pp. 1-66).
60 New Perspectives on Political Economy
his action and what can be logically inferred from this evidence and the conceptual
meaning of action. That objective knowledge can be acquired from a person’s action
about his preferences is the basis of the concept of demonstrated preference.
Demonstrated preference, then, is not a special concept of welfare economics, but
the basis of all action. Rothbard wrote, “The concept of demonstrated preference is
simply this: that actual choice reveals, or demonstrates, a man’s preference; that is,
that his preferences are deducible from what he has chosen in action.”¹³
Because a person’s preference can be deduced from his action, someone observing
his action can have objective knowledge about it. The deduction comes from combining
the conceptual meaning of action and the empirical evidence of the action. For
example, a farmer plants a corn crop. An economist can deduce from this action that
the farmer preferred planting a corn crop to the next best alternative that he considered.
This conclusion is not evident from empirical evidence alone, but implied from
conceptual understanding of action, speciﬁcally the understanding that any action
aims at a preferable state of aﬀairs. That to be scientiﬁc a treatment of a subject must
be objective, i.e., based on facts and reason, is the source of Rothbard’s claim that a
scientiﬁc treatment of utility and welfare economics can include only demonstrated
preference.¹⁴
Although there is no welfare economics of the person acting in isolation, Rothbard
claimed that the economic theory of the person acting in isolation is the foundation
for welfare economics since it underlies any and all action, including social
interaction. Thus, voluntary exchange is mutually beneﬁcial because in choosing to
exchange ownership of goods both traders aim at a preferable state of aﬀairs. Kvasnička
himself, along with most other economists, accepts these basic principles.¹⁵
Rothbard’s welfare economics was an appeal to mainstream economists to make these
fundamental principles of economics that they accept the foundation of welfare economics.
In advancing this line of argument, Rothbard was hoping to bring mainstream
economists around to accepting the laissez-faire conclusions that stem from
¹³ Rothbard (1997a, p. 212). Italics original.
¹⁴ On his view of science, see Rothbard (1997b, pp. 32-34).
¹⁵ Kvasnička (2008, p. 46) writes, “[Rothbard’s ﬁrst welfare theorem] is derived as a generalization of
the old doctrine that both traders beneﬁt from a voluntary exchange, which is in turn a special case
of the doctrine which inspired the concept of demonstrated preference that a man acts to beneﬁt.”
Herbener: In Defense of Rothbardian Welfare Economics 61
basic economic theory. In retrospect, Rothbard’s quest may seem quixotic, but not, I
think, incoherent as Kvasnička claims.¹⁶
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.
These conclusions must be deduced from the conceptual meaning of action and
the empirical evidence of the action. They are not evident from the empirical evidence
of the action alone. The mutual beneﬁt of voluntary exchange is deduced from
the understanding the economist has that each person in any action he takes, given
its circumstances, chooses what he prefers. That deduction is necessary to arrive at
social welfare conclusions has particular importance in the case of involuntary exchange.
What one observes about involuntary exchange appears on the surface to
be no diﬀerent than what one observes about a voluntary exchange. In both cases, a
person chooses to transfer the ownership of something to another person. That he
is not demonstrating a preference for the transfer in the case of aggressive violence
used against him is inferred from his initial property ownership and that aggressive
violence is an implicit, non-consensual claim on his property. Without such conceptual
understanding, the economist could not deduce the harm to the victim from
aggression. It cannot be seen from the empirical evidence of involuntary exchange
alone.
¹⁶ Kvasnička (2008, p. 43).
62 New Perspectives on Political Economy
Determining the social welfare consequences of voluntary exchange and involuntary
exchange relies upon another basic principle about preference that economists
generally assent to, viz., its subjectivity. As discussed above, value is a state of mind.
As such, it has no extensive property that can be measured. For this reason, there
can be no interpersonal comparisons of utility. To conclude that social welfare has
improved from a social interaction, therefore, it must be shown that the interaction
makes at least one person better oﬀ and no one worse oﬀ. Such interactions can be
called Pareto superior. If a social interaction makes some persons better oﬀ and others
worse oﬀ, then it cannot be concluded that social welfare has improved. Such
interactions can be called Pareto inferior.¹⁷
Even though it can be deduced that both traders beneﬁt from a voluntary exchange,
the consequences for social welfare are indeterminate until third-party effects
are considered. It is possible that a third person who is not a party to the exchange
made by two other persons has an adverse psychological response to the exchange.
He is envious, let us say. The economic-theoretical meaning of this case is
that the conditions of action external to his mind, and which he does not control,
have changed for the envious person. Other persons have used their personal property
in a way that he does not like. The relevant question for determining the eﬀect
on social welfare is: Does this psychological change aﬀect his preferences? If it does
not, then his action does not change. He demonstrates that he is no worse oﬀ in preference
than before and, to reiterate, preference is the objective standard for social
welfare considerations. If the feeling of envy is strong enough to change the person’s
preferences, then he changes his action toward a more preferred alternative, say complaining
to his friend. In this case, too, his action, given the new circumstances, is
more-preferred, not less. It can be conceded, though never objectively established,
that his level of utility is higher before the voluntary trade of others than after it,
but that does not imply that the voluntary trade of others made the envious person
worse oﬀ in preference. Preference refers to the judgment a person makes in his mind
comparing the value of alternatives in action given the circumstances of the action.
¹⁷ Kvasnička seems to think that in using the phrase Pareto inferior to describe cases of state intervention,
I meant that a reduction of social welfare had occurred. Pareto inferior, however, refers to cases
in which it cannot be concluded that social welfare has improved. No cases could exist for the deﬁnition
Kvasnička implies for the phrase Pareto inferior, because the only way to have an unambiguous
deterioration in social welfare is if both parties to a social interaction are made worse oﬀ. Obviously,
neither party would act to bring about such a result. See Kvasnička (2008, p. 49, fn. 6).
Herbener: In Defense of Rothbardian Welfare Economics 63
Deducing the eﬀects on social utility from voluntary and involuntary exchanges
requires considering each action in turn given the conditions as they are at that point.
Nothing can be deduced about the level of utility a person has at the beginning of a
series of actions compared to the level of utility he has at the end of the series of
actions. For example, a person having dinner with his friends orders steak from the
menu. The economist observing him can objectively conclude that, given his options,
he selected what he preferred. He is enjoying the conversation when it turns to a
subject he dislikes, but he stays and endures it. The economist observing him, lacking
access to what he is experiencing in his mind, can objectively conclude that he prefers
to continue dining with his friends. At some point, one of his companions makes a
remark so objectionable to him that he says, “Anymore such talk and I shall leave.”
The economist observing him can objectively conclude that he preferred to make this
remark. The economist cannot objectively conclude that this line of conversation has
lowered the level of his utility. To conclude that would require the economist to make
a judgment about his utility. The economist would have to interpret the meaning of
the remark as it relates to his utility. The economist would have to decide whether it
was a serious remark or a joke and if it was serious did making the remark push his
utility up or down. Bullies, after all, like to intimidate others with such remarks. No
such judgments are necessary for the economist to conclude that he preferred making
the remark. It follows from the objective evidence of his action and the conceptual
meaning of action. And so it goes for the rest of the evening with the level of his utility
sometimes rising and sometimes falling, but he continues dining with his friends and
leaves only after the party breaks up. Is he enjoying a higher level of utility after the
evening is over compared to before it began? Who can objectively say but the person
himself? He is the only person with experiential knowledge of his own utility. What
another person can objectively deduce is that he preferred what he did each step of
the way.
Moreover, recognizing that the given conditions of a social interaction are necessary
to determining its social welfare consequences leads to a regression theorem of
social interactions. The action taken by the envious person occurred in the conditions
that existed because of the prior voluntary exchange of other persons. But that voluntary
exchange occurred in the conditions that existed because of the prior actions
of other persons and so on. Tracing these interactions backwards step-by-step leads
64 New Perspectives on Political Economy
to the starting point of each person’s ownership of his mind and body. Each person’s
ownership of his labor is the initial endowment of property in which he begins the
process of social interaction. Each step of acquisition of property in its natural state
by someone demonstrates that he gains in preference and no one is demonstrably
worse oﬀ in preference. With their naturally-acquired property, persons exchange
voluntarily and thereby, demonstrate that they prefer their social interaction while
no one is demonstrably worse oﬀ in preference.¹⁸
The determination of social welfare is not an end-state of optimality to be
achieved. Instead it is a step-by-step comparison of whether or not any particular
social interaction improves social welfare. To be consistent with fundamental economic
principles, the analysis can use the concepts of Pareto superior and Pareto inferior,
but not Pareto optimal. In each voluntary exchange on the unhampered market
traders attain what they prefer more without third-parties suﬀering what they prefer
less. Each step taken in the market economy improves social welfare and hence,
is Pareto superior. In each involuntary exchange aggressors attain what they prefer
more and victims suﬀer what they prefer less. Each step taken by the state fails to
improve social welfare and hence, is Pareto inferior.¹⁹
3 Kvasnička’s Critique
Kvasnička claims that his critique diﬀers from the criticism of Rothbard oﬀered by
others in that “they analyze a broad context of the theory, or even question Austrian
¹⁸ See, Herbener (1997, pp. 98-101). That naturally-owned property is the initial endowment from which
the process of social interaction begins is the Rothbardian rebuttal to justiﬁcations for state redistribution
of income that mainstream economists make in their welfare economics to achieve equity. See
note 43 below.
¹⁹ In writing about welfare economics, a leading historian of economic thought recognizes this distinction
between the step-by-step process analysis of Austrian economics and the end-state equilibrium
analysis of mainstream economics. He wrote:
[Adam Smith] held what I have elsewhere called “a process conception of competition,”
nowadays associated with Austrian economics, in contrast to the orthodox “endstate
conception of competition,” in which all emphasis is directed to the nature of the
ﬁnal equilibrium, regardless of how that equilibrium is attained. (Blaug 2007, p. 189).
Herbener: In Defense of Rothbardian Welfare Economics 65
economics as such… Our analysis is based on a detailed study of Rothbard’s own arguments
only. We have only checked the coherency of his arguments.”²⁰
The main thrust of Kvasnička’s critique, however, is not to point out internal inconsistencies
in Rothbard’s argument, but to criticize him for holding a distinctly
Austrian instead of mainstream view of utility. In the mainstream view, economic
theory consists of models constructed by the economist to explain market phenomena.
The model contains “agents” whose “actions” are determined by stipulating
mathematically tractable utility functions for them. “So there is not a fundamental
diﬀerence between the indiﬀerence curves approach and Rothbard’s own utility
theory as long as the former one yields a unique solution,” Kvasnička writes, “The
indiﬀerence curves just describe the agent’s inner world, in which Rothbard takes no
interest, or rather denies its existence altogether.”²¹ Concerning welfare economics,
Kvasnička writes, “The usual meaning of Pareto’s rule is God-like – the analyst pretends
to know everything that is happening and how it aﬀects all agents. Obviously,
such an approach is scientiﬁcally valid…”²² Setting aside the question of the scientiﬁc
validity of this approach, Rothbard rejects it in favor of praxeology, which builds economic
theory as a body of logical implications from the nature of human action as set
out above. As a praxeologist, Rothbard is interested in the inner world of real human
beings in so far as they establish preferences. He has no interest in the inner world of
economic agents, which is nothing more than mathematically tractable utility functions
invented by mainstream economists. It is not clear that by being interested in
the inner world of agents of their own invention that mainstream economists themselves
have any interest in the inner world of real human beings. In any case, with this
line of criticism, Kvasnička can, at most, merely show that Rothbard’s conclusions are
inconsistent with mainstream analysis, but not that they are internally inconsistent.
Because he sees utility as a mathematical function stipulated by the economist, it
appears to Kvasnička that Rothbard is merely asserting the concept of demonstrated
preference as a constraint on the utility function necessary to permit him to make
²⁰ Kvasnička (2008, p. 43).
²¹ Kvasnička (2008, p. 44).
²² Kvasnička (2008, p. 46).
66 New Perspectives on Political Economy
social welfare claims in the face of the diﬃculties of interpersonal utility relationships
such as third-party envy. He writes,
Rothbard was also able to see that Pareto’s rule is very strict, and that almost
no change is either Pareto-improving, or Pareto-worsening. He gave the example
of envy … Rothbard believed he could get a determinate solution in more cases
with the tool he called demonstrated preference.²³
But, Rothbard did not invent demonstrated preference as a means of creating a class
of events that satisfy the Pareto criterion as Kvasnička implies. Demonstrated preference
was, for Rothbard, a fundamental concept of human action. And it is precisely
Rothbard’s acceptance of demonstrated preference as a concept logically established
before the chain of logic reaches the question of interpersonal utility relations that
leads him to deny that envy of a third party constitutes an exception to the Pareto
superior status of a voluntary exchange.²⁴
Kvasnička then poses two interpretations of what Rothbard meant by the concept
of demonstrated preference. The ﬁrst is “we can know nothing that was not
demonstrated in an action” and the second is “there is nothing more than what was
demonstrated in an action.” Kvasnička claims that Rothbard held the second view.²⁵
But, he clearly held the ﬁrst view. Rothbard wrote:
Now since praxeology shows, by the concept of demonstrated preference,
that both the newsdealer and I gain in utility from the exchange, and nothing has
demonstrably happened to anyone else, we can conclude scientiﬁcally, as praxeological
economists, that social utility has increased from the sale and purchase
of the newspaper – since we have deﬁned social utility in the Paretian manner.
It is true, of course, that third parties may well be grinding their teeth in hatred
²³ Kvasnička (2008, p. 43). Italics original.
²⁴ Even the typical mainstream treatment of utility in welfare economics does not include third-party
envy in utility functions. As Blaug (2007, p. 185) writes, “The ﬁrst fundamental theorem states that,
subject to certain exceptions – such as externalities, public goods, economies of scale, and imperfect
competition – every competitive equilibrium is Pareto-optimal.” Third party envy is not on the
list of exceptions and, save for such exceptions, every competitive equilibrium is Pareto optimal. If
mainstream economists would have taken Kvasnička’s view of the pervasiveness of third-party envy,
they would never have developed welfare economics at all since under this view the social welfare
consequences of every social interaction are indeterminate.
²⁵ Kvasnička (2008, p. 44).
Herbener: In Defense of Rothbardian Welfare Economics 67
at the exchange. There may be people, for example who through envy suﬀer
psychic loss because the newspaper dealer and/or I have gained. Therefore, if
we employ the Paretian deﬁnition of “social utility” in the usual psychologizing
sense, we can say nothing about social utility one way or the other. But if we
conﬁne the concept to its strict scientiﬁc compass in demonstrated preference,
then we can state that social utility increases from the exchange. Still further,
we may know as historians, from interpretive understanding of the hearts and
minds of envious neighbors, that they do lose in utility. But we are trying to
determine in this paper precisely what scientiﬁc economists can say about social
utility or can advocate for public policy, and since they must conﬁne themselves
to demonstrated preference, they must aﬃrm that social utility has increased.²⁶
Rothbard’s point in insisting on the use of demonstrated preference is precisely that it
is consistent with the empirical evidence from an action and the conceptual meaning
of action and, therefore, it can be part of a scientiﬁc, i.e., praxeological, treatment of
human action. Rothbard does not “ignore all changes of a person’s utility except those
the person demonstrates through his actions,” as Kvasnička claims; he just recognizes
that they play no role in the logical structure of action and hence, are not part of economic
theory.²⁷ This distinction between the abstract and concrete aspects of action
is the key to solving Kvasnička’s puzzlement that Rothbard can hold at the same time
that aspects of action beyond demonstrated preference can be known or understood
by the historian while only demonstrated preference can be accepted as objective
evidence by the economist.²⁸ The diﬀerence is like a scientist who is conducting a
controlled experiment to determine the physical eﬀects of smoking. He may know
a lot about the lifestyle of his grandfather, who has smoked for forty years with no
obvious physical repercussions, but he cannot include this knowledge in his scientiﬁc
study. As a scientist, he can only include data that has been scientiﬁcally generated in
his experiment. Another analogy is a mathematician who deduces the Pythagorean
Theorem from prior postulates and theorems of geometry. Whatever can be know
about actual right triangles drawn on paper or embedded in architecture is not relevant
for his proof. He is not ignoring, and does not need, other knowledge about
right triangles that he may have acquired from drawings and architecture in making
²⁶ Rothbard (1997c, pp. 88-89).
²⁷ Kvasnička (2008, p. 44).
²⁸ Kvasnička (2008, p. 44, fn. 3).
68 New Perspectives on Political Economy
his proof. In like manner, the economist is not concerned with the particular, concrete
aspects of action as the historian is, but with the universal, conceptual structure
of action.²⁹
Kvasnička’s god-like approach does not avoid this distinction between the conceptual
and concrete and its consequent diﬃculties. This approach does not say that the
economist should conduct utility and welfare economics using every aspect of what
he can actually come to know about real human persons. Instead, it counsels him
to construct mathematically tractable utility functions for mythical economic agents
and deduce, by the use of mathematics, utility and welfare theorems from them. Such
constructs are held up as a model by which the economist is to understand the action
of real human persons. He is still left with the question of how these models relate
to the human action of real persons.³⁰
Kvasnička backs up his view that Rothbard held the position that “there is nothing
more than what was demonstrated in action,” by claiming that Rothbard argued that
“there is no way to demonstrate indiﬀerence, hence there can be no indiﬀerence at
all.”³¹ But, Rothbard did not deny that psychological states of indiﬀerence can exist.
He wrote:
The concept of “indiﬀerence” may be important for psychology, but not for
economics. In psychology, we are interested in ﬁnding out intensities of value,
²⁹ As with economic theory, Rothbard accepted Mises’s views of the relationship between theory and
history. See his preface to Mises (1985).
³⁰ Blaug claims that mainstream economists have used the diﬃculty of reconciling the abstract and the
concrete as an opportunity to push their own ideologies. He writes,
This sort of intellectual schizophrenia [of mainstream economists over the Coase
Theorem] would be excusable provided it were well understood that the irrelevance of
the initial distribution of property rights to a ﬁnal allocation of resources is, like Paretooptimal
properties of competitive equilibrium, a truth about economic models and not
a truth about the real world. But I doubt that is suﬃciently underlined, because such
underlining would destroy the ceremonial value of formal theorems in a subject like
economics. Many economists know in their heart that the market is eﬃcient, and know
it as an unexamined ideological belief but do not want to admit to themselves that it
is ideological. So they revel in mathematical theorems that satisfy their self-respect
… [They] accept the judgment that the theorems are close to reality, while those that
distrust markets do not. (Blaug 2007, p. 201)
³¹ Kvasnička (2008, p. 44). Italics original.
Herbener: In Defense of Rothbardian Welfare Economics 69
possible indiﬀerence, and so on. In economics, however, we are only interested
in values revealed through choices. It is immaterial to economics whether a man
chooses alternative A to B because he strongly prefers A or because he tossed a
coin. The fact of ranking is what matters for economics, not the reasons for the
individual’s arriving at that rank.³²
What Rothbard claimed was that indiﬀerence has no relevance for the conceptual
structure of action. If such states of indiﬀerence mean that no preference exists then
no action can come from them since preference is the requisite for choice. And if in
such states of indiﬀerence a person can still establish a preference, then action can
take place. In the ﬁrst case, indiﬀerence is not part of the conceptual structure of action
and in the second case it is just one of the many psychological, physiological, and
other conditions inﬂuencing the establishment of a preference. As such, no reason
exists to single indiﬀerence out of the multiplicity of other such conditions for special
treatment. Even in the indiﬀerence curve approach of mainstream economists, a
person could not choose, and therefore could not act, on the basis of an indiﬀerence
curve alone. He chooses from among the combinations of goods that give him the
same utility, the one that he can obtain for the lowest budgetary expenditure. Or,
what amounts to the same thing, for a given budgetary expenditure he selects the
combination of goods that places him on the indiﬀerence curve which gives him the
greatest utility. Either way, he chooses what he prefers, either the same utility with a
smaller monetary expenditure or greater utility for the same monetary expenditure.
Indiﬀerence, as such, has nothing to do with choosing between alternatives even in
the mainstream approach. The assumption of indiﬀerence is made, instead, to justify
the use of a utility function. Indiﬀerence in the mainstream approach means, not
merely that a person cannot choose between alternatives, but that the utility of the
alternatives is equal. If the utility of alternatives is equal, then utility can be presumed
to be cardinal. Cardinality is required to establish utility functions and work through
the mathematical logic of the utility-function approach.
There is one more point of indictment against Rothbard that Kvasnička infers
from his discussion about indiﬀerence. Against the claim that indiﬀerence might be
demonstrated by repeated trials of action in which a person chooses each alternative
³² Rothbard (1997a, p. 226). Italics original.
70 New Perspectives on Political Economy
half the time, Kvasnička says that “Rothbard argues such a comparison is not possible
because over time the agent’s preferences might have changed.”³³ Once again, the
problem with Kvasnička’s claim is that Rothbard’s praxeological approach develops
the conceptual structure of the action of real human beings. It is not a model of economic
agents. Of course, it is possible to invent agents with utility functions that give
statistical results of whatever type the economist wants. But with a real human person,
the economist cannot presume that in repeated trials (even those in which every
external circumstance is the same) his preference with respect to the circumstances
of action remain the same. On this point, Rothbard is again following Mises who argued
that functional analysis of human action is ruled out by the lack of constants.
Functional equations have variables and constants, but in human action nothing exists
but variables. No constant quantitative relationships exist between a dependent
variable (e.g., the quantity of a good a person purchases) and independent variables
(e.g., the price of the good) because the quantitative eﬀect on the dependent variable
of a change in an independent variable depends on the judgment made in the mind
of the person acting which can change over time.³⁴
Although Rothbard claims that the economist cannot compare the results of real
actions at diﬀerent times as a means of determining a person preferences or indifferences,
he does not hold that “we cannot compare a real situation with another
hypothetically the same,” as Kvasnička claims.³⁵ Precisely such comparisons are the
basis for many economic laws, e.g., the law of demand. The law is logically deduced
from the concept of diminishing marginal utility by posing a hypothetical situation
diﬀerent from the actual buying situation in only one way, viz., that the price is conjectured
to be lower. A person would have bought at least as much of the good in
the hypothetical case compared to the amount he actually bought. Rothbard accepts
such ceteris paribus comparisons in cases for which it is acceptable to say that a person’s
preferences cannot change, e.g., at the moment a person buys a good.³⁶ He does
not, however, accept the presumption that a real human person’s preferences cannot
change over time in a series of buying opportunities.
³³ Kvasnička (2008, p. 45).
³⁴ Mises (1998, pp. 347-354).
³⁵ Kvasnička (2008, p. 45).
³⁶ Rothbard (1970, pp. 91-108).
Herbener: In Defense of Rothbardian Welfare Economics 71
Kvasnička then turns to what he considers the limitations of the concept of
demonstrated preference. First, he claims that “an agent can demonstrate only those
changes of his utility … when he is active. There is no way a passive agent may demonstrate
a change of his utility caused by an external force he passively suﬀers.”³⁷ Rothbard,
he asserts, uses demonstrated preference to ignore the utility losses of persons
who do not actively demonstrate their losses but passively accept them. He writes,
“the possibly losing non-involved third party cannot demonstrate it loses because it
loses passively – there is no action that can prove it …”³⁸ Action, however, can be either
“active” or “passive” and hence, can demonstrate a preference in either case. As
discussed above, preference is inferred from what we know about action conceptually
and the empirical evidence from an action. It cannot be known from empirical observation
alone, e.g., that a person is passive. Second, Kvasnička writes, “Moreover, an
action only demonstrates that the agent is better oﬀ choosing the action in comparison
to choosing another possible action in his situation, not that he actually is better
oﬀ.”³⁹ His examples of being oﬀered a gift and not resisting aggression do not show
that demonstrated preference is of no use in social welfare analysis, but that one must
take account of the circumstances of action in addition to demonstrated preference
in such an analysis.⁴⁰ As discussed above, there must be circumstances of each action
and some of them are outside the mind and control of the person. Welfare economics
takes these circumstances as given and deduces a person’s preferences from what we
know conceptually about action and what the person chooses to do in the given circumstances.
This point also applies to Kvasnička’s claim that Rothbard “ignore(s) the
impact of a change of the income and wealth distribution on individual’s [sic] utilities
under Pareto’s rule” because “distribution is not independent from production
and exchange … and that a change in the distribution is caused by voluntary-made
exchanges.”⁴¹ Rothbard’s argument, however, is that changes in income and wealth
distribution are conditions of action in which a person chooses what he prefers. Each
social interaction is a separate event with its own conditions determined by prior actions.
When an interaction takes place it changes the conditions for future actions.
³⁷ Kvasnička (2008, p. 45).
³⁸ Kvasnička (2008, p. 48). Italics original.
³⁹ Kvasnička (2008, p. 45). Italics original.
⁴⁰ Kvasnička (2008, pp. 45-46).
⁴¹ Kvasnička (2008, p. 47).
72 New Perspectives on Political Economy
Social welfare consequences can be deduced for each action, one at a time, as persons
face changing circumstances. Kvasnička’s claim that “Pareto’s rule looks for all
subjects’ utility increases or decreases in comparison to the status quo, despite their
cause,” may be true in the mainstream approach with the caveat that what counts as a
factor changing utility is stipulated by the economist in the agents’ utility functions.⁴²
The typical treatment of income distribution in mainstream welfare economics, however,
is not to include it as a factor in utility functions as Kvasnička implies, but to
see it as an initial endowment that conditions the particular Pareto optimal point
that is reached by voluntary exchange. Such a treatment permits the state to select
what it considers the preferable Pareto optimal outcome and to redistribute income
to set the initial conditions so that the process of the market achieves the selected
Pareto optimal point. Even if the state cannot intervene in the unhampered market
to improve social welfare, it can redistribute income to achieve equity while letting
the market achieve eﬃciency.⁴³
Kvasnička then applies the limitations that he claims for demonstrated preference
to the two theorems of Rothbard’s welfare economics. The two welfare theorems are
“that the free market always increases social utility” and “that no act of government
can ever increase social utility.”⁴⁴ His argument about the ﬁrst welfare theorem is
that Rothbard uses demonstrated preference to ignore the utility losses suﬀered by
⁴² Kvasnička (2008, p. 47). Italics original.
⁴³ Blaug writes:
Modern welfare economics is formally summed up in two so-called fundamental
theorems. The ﬁrst fundamental theorem states that, subject to certain exceptions –
such as externalities, public goods, economies of scale, and imperfect competition – every
competitive equilibrium is Pareto optimal. The second fundamental theorem states
that every Pareto-optimal allocation of resources is an equilibrium for a perfectly competitive
economy, provided a redistribution of initial endowments and property rights
is permitted; alternatively expressed, every Pareto-optimal allocation of resources can
be realized as the outcome of competitive equilibrium after a lump-sum transfer of
claims on incomes. (Blaug 2007, p. 185)
An example of what Blaug is referring to can be found in a leading Intermediate Microeconomics
textbook. About the Pareto criterion, the authors write, “it evades most of the interesting questions
about the distribution of income. The criterion has nothing to say about a change that beneﬁts one
group of people and harms another.” They then provide an example from Hal Varian of how diﬀerent
initial endowments of income permit the satisfaction of both equity and eﬃciency goals. See,
Mansﬁeld and Yohe (2004, pp. 612-613).
⁴⁴ Rothbard (1997a, p. 243).
Herbener: In Defense of Rothbardian Welfare Economics 73
third parties when others voluntarily exchange. He writes, “Thus under the concept
of demonstrated preference the market exchange, and hence the market process itself,
increases social utility because at least one party beneﬁts, and no one can prove
he loses.”⁴⁵ To illustrate his point, Kvasnička applies his analysis to Rothbard’s welfare
economics of a voluntary cartel. Kvasnička agrees that members of a voluntary
cartel demonstrate their beneﬁt in forming the cartel when selling to consumers at
higher prices but that “the consumers have to pay the higher price, but they cannot
demonstrate they lose by any action, and hence Rothbard ignores them, and calls the
change Pareto-improving.”⁴⁶ But consumers are demonstrating their preferences. By
purchasing the good, consumers demonstrate that they prefer to buy at the current
prices. Determining this is done in exactly the same way as it would be before the
cartel formed and prices were lower. In that case, too, consumers demonstrated that
they preferred to buy at the current prices, even if they were also higher than previous
prices. Social welfare consequences can be determined only for each action, one
at a time. The conditions of a person’s action determined by other people are given
to him for each action. As discussed above, whether or not consumers have higher
utility or lower utility from the beginning of a series of actions compared to its end
cannot be objectively determined. For this reason, Rothbard does not include such
comparisons over time in welfare economics. To do so, the economist would have to
account not only for the eﬀect higher prices have on consumers’ states of mind, but
any other factor aﬀecting their utility in the interim between buying at lower and buying
at higher prices. Of course, in the abstract and ceteris paribus, consumers always
prefer lower to higher prices but welfare economics is about the consequences of real
actions in their actual circumstances and not about ceteris paribus conjectures.
Turning to the second welfare theorem, Kvasnička argues that Rothbard switches
uses of the Pareto criterion from one based on demonstrated preference, used to eliminate
third-party losses of utility from voluntary exchange, to one based on the godlike
view of utility, used to show that a person coerced by the state loses. He writes,
“[Rothbard] uses Pareto’s rule in its usual meaning now … because under demonstrated
preference the losing party cannot prove it loses because it suﬀers the loss
⁴⁵ Kvasnička (2008, p. 48). Italics original.
⁴⁶ Kvasnička (2008, p. 49). Italics original.
74 New Perspectives on Political Economy
passively …”⁴⁷ In applying this analysis to Rothbard’s case of the voluntary cartel,
Kvasnička concludes that Rothbard has not shown that a state enforced breakup of
the cartel fails to improve social welfare. Perhaps this would be correct if Rothbard
was claiming that “producers lose [because] they can charge only competitive prices
again” as Kvasnička cites as the source of loss.⁴⁸ But Rothbard argued that the loss
to members of the cartel can be inferred from the coercion the state uses to prevent
them from doing what they prefer to do in the given situation and not from comparing
the higher cartel prices before state intervention to the lower non-cartel prices
after state intervention.⁴⁹ Kvasnička himself characterizes Rothbard’s argument in
this way: “the government’s action coerces someone else to do what he does not want
to do, or abstain from an action he would like to do. This person loses.”⁵⁰ Rothbard
did not switch meanings of the Pareto rule in his second welfare theorem, but consistently
applied the concept of demonstrated preference, that a person’s preference can
be deduced from the empirical evidence of his action and the conceptual meaning of
action.
Kvasnička makes two criticisms of Rothbard’s welfare economics that do not stem
from his claim that Rothbard’s views should be assessed in terms of the mainstream
approach.
First, he claims that Rothbard ignores the eﬀect negative externalities have on
the welfare improving character of the unhampered market because they are “caused
by ill-speciﬁed property rights” and hence, “ruled out of the free market by deﬁnition.”⁵¹
Rothbard’s argument, however, is that the welfare loss of negative externalities
cannot justify government intervention to correct it because the cause of negative
externalities is not the working of the market but the failure of government to enforce
property rights or government interference in the establishment and defense of
property rights. In other words, Rothbard is saying that to correct the ineﬃciency of
negative externalities, the state must remove its own ineﬃciency-generating activity.
State intervention into the activity of the market will not correct negative externalities
⁴⁷ Kvasnička (2008, p. 48). Italics original.
⁴⁸ Kvasnička (2008, p. 49). Parentheses removed.
⁴⁹ Rothbard (1997a, pp. 242-243).
⁵⁰ Kvasnička (2008, p. 48).
⁵¹ Kvasnička (2008, p. 47).
Herbener: In Defense of Rothbardian Welfare Economics 75
but impose its own additional ineﬃciency on the market. Kvasnička’s argument that
“since the deﬁnition and enforcement of property rights is [sic] costly, they cannot be
complete in the real world,” and therefore, Rothbard’s claim that negative externalities
are ruled out of the free market by deﬁnition “means either that the free market
cannot exist, or that Rothbard’s theory is of no relevance to the real world” misses
the mark.⁵² Even if Rothbard were trying to argue for the superior welfare results of
the unhampered market versus the hampered market with respect to negative externalities,
all he would have to show is that the former results in smaller welfare losses
than the latter, not, as Kvasnička claims, that there are no such welfare losses in the
unhampered market. The unhampered market need not achieve Nirvana to be justiﬁed.⁵³
It only needs to be better than any other realizable arrangement.
Second, Kvasnička claims that Rothbard’s welfare economics contradicts his general
economics. Following Barnett and Block, he claims that Rothbard’s welfare economics
must conclude that an increase in money production by the state cannot improve
social welfare while an increase in money production by private enterprise does
improve social welfare. Rothbard, however, did not follow his own welfare economics
in this case. On the basis of his general economics, he argued instead that any amount
of money is socially optimal.⁵⁴ The apparent inconsistency can be cleared up by removing
the ambiguity in the phrase “socially optimal.” Rothbard meant by this phrase
that any amount of money can perform the entire medium of exchange function in
society.⁵⁵ The entire set of exchanges people desire to make can be made with any
amount of money. No inconsistency exists between the claim that any amount of
money is socially optimal in this sense and the existence of social welfare eﬀects from
money production described by Barnett and Block. To demonstrate that any amount
of money is socially optimal, the economist postulates a given amount of money and
underlying preferences and other conditions that give rise to particular demands for
goods and factors which result in particular prices of those goods and factors. If each
person’s money holdings were instead twice as large, with the same underlying con⁵²
Kvasnička (2008, p. 47).
⁵³ On the Nirvana Fallacy, see Demsetz (1969).
⁵⁴ Kvasnička (2008, p. 50). See Barnett and Block (2004).
⁵⁵ “Any amount” of money means any amount larger than a technologically necessary minimum if such
exists. The argument refers to the adjustment of prices to diﬀerent amounts of money, not to the
implications of a technologically minimum amount of money.
76 New Perspectives on Political Economy
ditions, then prices would be twice as high and if they were half as large, prices would
be half as high. But no matter the amount of money, people would be able to make
all the exchanges they desire to make. There are no social welfare implications of this
analysis. It is designed to compare end- states that the market arrives at from given
initial conditions that diﬀer in one respect. It shows that in contrast to consumer and
producer goods, a larger stock of money in society does not permit people to attain
to a greater extent the end to which they put money. To demonstrate the social welfare
eﬀects of money production, the economist postulates given market conditions
and then that more money is produced by money producers and spent by them on
some goods and factors, which in turn is spent by the producers of those goods and
factors on other goods and factors and so on. In this case, there are social welfare
consequences to be analyzed. The analysis traces the step-by-step process by which
the market moves from one end-state to another.⁵⁶
4 Conclusion
Rothbard’s main argument for the unhampered market economy was that it achieves
the ethical use of interpersonal violence more fully than any other social arrangement.
Working within natural law philosophy, he showed that this ethic requires the defense
of natural rights against aggression and that the unhampered market economy
is integral to the free society in which there is no legitimate aggression against natural
rights. Complementary to his main argument concerning ethics was his economic argument
that the unhampered market promotes the life and ﬂourishing of man more
fully than any other social arrangement. Working within praxeology, he accepted
Mises’s argument that the life and ﬂourishing of man is promoted by the development
of the division of labor and that the unhampered market develops the division
of labor more fully than other social arrangements by bringing decisions concerning
it under economic calculation as completely as possible. His welfare economics was
not written as a defense of the unhampered market but as a critique of mainstream
welfare economics. By grounding welfare economics in fundamental economic principles,
he showed that every interaction on the unhampered market improves social
⁵⁶ Admittedly, Rothbard’s discussion of this matter lacks his usual clarity. See Rothbard (1970, pp. 670-
671).
Herbener: In Defense of Rothbardian Welfare Economics 77
welfare and no interaction of the state can improve social welfare. He oﬀered his argument
to convince mainstream economists to follow through the logic of what is implied
by the fundamental economic principles they hold to be true to its laissez-faire
conclusion. Mainstream economists have failed to embrace Rothbard’s welfare economics
because they reject his praxeological method in favor of a modeling method.
As a praxeological exercise, however, his welfare theorems are valid.
Bibliography
[1] Collingwood, R.G. 1939. An Autobiography. Oxford: Oxford University Press.
[2] Gordon, David. 1993. “Toward a Deconstruction of Utility and Welfare Economics.”
Review of Austrian Economics, vol. 6, no. 2.
[3] Rothbard, Murray. 1982. The Ethics of Liberty. Atlantic Highlands, N.J.: Humanities
Press.
[4] Rothbard, Murray. 1978. For a New Liberty, rev. ed. New York: Collier Books.
[5] Mises, Ludwig von. 1998 [1949]. Human Action, Scholar’s Edition. Auburn, Ala.:
Mises Institute.
[6] Kvasnička, Michal. 2008. “Rothbard’s Welfare Theory: A Critique.” New Perspectives
on Political Economy, Vol. 4, No. 1.
[7] Rothbard, Murray. 1970. Man, Economy, and State. Los Angeles: Nash Publishing.
[8] Stringham, Edward. 2007. “Economic Value and Costs are
Subjective.” Working Papers. Auburn, Ala.: Mises Institute,
http://www.mises.org/journals/scholar/stringham4.pdf.
[9] Rothbard, Murray. 1997a. “Toward a Reconstruction of Utility and Welfare Economics.”
In The Logic of Action I: Method, Money, and the Austrian School. Cheltenham,
U.K.: Edward Elgar.
[10] Rothbard, Murray. 1997b. “Praxeology as the Method of the Social Sciences.” In
The Logic of Action I: Method, Money, and the Austrian School. Cheltenham, U.K.:
Edward Elgar.
78 New Perspectives on Political Economy
[11] Rothbard, Murray. 1997c. “Praxeology Praxeology, Value Judgments, and Public
Policy.” In The Logic of Action I: Method, Money, and the Austrian School. Cheltenham,
U.K.: Edward Elgar.
[12] Herbener, Jeﬀrey M. 1997. “The Pareto Rule and Welfare Economics.” Review of
Austrian Economics, vol. 10, no. 1.
[13] Blaug, Mark. 2007. “The Fundamental Theorems of Modern Welfare Economics,
Historically Considered.” History of Political Economy, vol. 39, no. 2.
[14] Mises, Ludwig von. 1985. Theory and History. Auburn, Ala.: Mises Institute.
[15] Mansﬁeld, Edwin, and Gary Yohe. 2004. Microeconomics. 11th edition, New York:
W.W. Norton.
[16] Demsetz, Harold. 1969. “Information and Eﬃciency: Another Viewpoint.” Journal
of Law and Economics, vol. 12, no. 1, pp. 1-22.
[17] Barnett, William, and Walter Block. 2004. “On the Optimum Quantity of Money.”
Quarterly Journal of Austrian Economics, vol. 7, no. 1, pp. 39-52.
ISSN 1801-0938
New Perspectives on Political Economy
Volume 4, Number 1, 2008, pp. 79 – 85
Book Review
Christopher Coyne. After War: The Political Economy of Exporting
Democracy. Stanford Press, 2007.
80 New Perspectives on Political Economy
West Germany, Japan, Somalia, Haiti, Afghanistan and Iraq. What do these six
countries have in common? Each of these countries has been the site of a U.S. led postwar
military occupation and reconstruction. At various points in the past century, the
United States has endeavored to export liberal democracy to nations abroad. While
some of these reconstructions have succeeded – West Germany and Japan are clear
success stories – others, like Somalia, have failed miserably. The outcomes of other
endeavors, speciﬁcally Iraq and Afghanistan, are still undetermined. When looking at
the various outcomes in these countries, an important question arises: What factors
contribute to successful installation of liberal democracy; and conversely, what factors
doom reconstruction to failure?
After War: The Political Economy of Exporting Democracy, is the latest attempt
at understanding the spread of liberal democracy. Christopher Coyne, author of the
book and professor of economics at the University of West Virginia, explains the dynamics
behind post-war reconstruction using tools from economics and game theory.
After War, Coyne examines the “economics of reconstruction” (31). The book uses
game theory tools to better understand the strategic interaction between the participants
in the reconstruction process. The book looks at post-war reconstruction as a
game involving several players – citizens of the occupied country, the occupier and
its agents (military oﬃcials, politicians), and international parties like the United Nations.
Each actor has his own agenda; each player pursues a set of goals to the best
of their abilities and within the constraints that he faces. The simultaneous pursuit
of diﬀerent objectives by diﬀerent actors creates a shaky interplay that inﬂuences the
outcome of the reconstruction.
After War has a neat structure. The book comprises four parts. The introduction
deﬁnes the reconstruction and lays out the main lessons of the book. Then, in
chapters two through four, the book deﬁnes relevant tools from economics and game
theory and identiﬁes how these tools can contribute to our understanding of the reconstruction
process. Chapters ﬁve through seven engage in a comparative study of
sixty-years worth of reconstructions. In this section, the author looks at six cases, divided
into three categories – successes, failures, and undetermined – and discuss the
factors that inﬂuenced the varying outcomes using the logic he develops in previous
Book Review 81
chapters. The book concludes in chapter eight by recommending new approaches to
exporting democracy.
According to the book, citizens of an occupied country face two choices during
reconstruction: a) to cooperate, by working with occupying forces and by bargaining
with rival factions; or, (b) to defect, or to create conﬂict through obstructive behavior
such as corruption, inﬁghting, and breaking agreements. Each citizen’s choice
depends upon the potential rewards and costs, which, in turn, are inﬂuenced by the
choices made by other citizens in the country.
Furthermore, After War underscores that the goal for occupiers is to make the reconstruction
process a game of cooperation rather than defection. The author notes
that when conﬂict is the dominant choice among the citizens of the occupied country,
those who cooperate face huge costs. If people in the occupied country perceive
the occupier as an invader rather than a liberator, then individuals who cooperate
with the occupier will attract hostility from others. In addition, those who invest
in liberal democratic institutions will receive no returns in the absence of credible
adoption by a majority of the citizens of the occupied country. After War emphasizes
that cooperation is only beneﬁcial for the individual under a scenario of collective
cooperation.
On the same note, the book asserts that a major factor for successful reconstruction
is the “art of association” (51), a tendency for citizens of a country to interact and
to create meaningful social networks. The author borrows this concept from nineteenth
century political theorist Alexis de Tocqueville, who lauded Americans for
their “capacity for interaction” that created a “robust civil society” in America (51).
These “associations” exist at the midway point between the public and private sectors
and provide individuals a forum in which they can address the relevant issues
without government interference. According to the book, countries that have mastered
this art are more likely to turn reconstruction into a game of cooperation. It
is the author’s view that these associations “create a shared identity that facilitates
social interaction and allows individuals to cooperate to get things done.” (52)
The book oﬀers some relevant examples that help clarify this concept. The author
asserts that, in the cases of Japan and West Germany, strong associations among the
82 New Perspectives on Political Economy
citizens facilitated the reconstruction of democratic institutions following the Second
World War. In Japan, a strong sense of national identity existed prior to the
war, and revealed itself in the country’s economic and social institutions. Coyne offers
the example of the zaibatsu, large industrial conglomerates that dominated the
Japanese economy in the nineteenth century. These conglomerates, which began as
exclusive family operations, evolved over time into more public enterprises supported
by external managers. Over time, the liberalization of Japanese zaibatsu produced a
widespread culture of cooperation in Japanese society. According to Coyne, the development
of these associations prior to the war, contributed greatly to the achievement
of credible collective cooperation in the post-war reconstruction game.
Similarly, the book points out that West Germany beneﬁted from a preexisting
art of association that contributed to the success of reconstruction after the war. In
the case of Germany, associations had grown out of the nineteenth century liberalization
of the economies of the members of the German Confederation. Increased
trade between the German economies strengthened the connections between members
of German society and spurred the development of liberal political institutions
that would serve as the foundation for new institutions installed during reconstruc-
tion.
The book contrasts these successful cases with reconstruction failures such as Somalia.
Somalia did not beneﬁt from a culture of cooperation that pervaded civil society
in Japan and Germany. Coyne points out that because Somali society comprises
many diﬀerent clans with varying historical allegiances, the art of association is much
weaker. Prior to independence, there was little economic or political cooperation
between clans. Clans existed within their own political and economic institutional
structure. The book argues that because of the entrenched separatism in Somalian
society, democratic institutions formed after independence emphasized clan identity
rather national identity. The failure of Somalia’s reconstruction is linked to a preexisting
system of exclusion and a widespread lack of the art of association.
Additionally, After War emphasizes that expectations and credibility can play a
critical role in the reconstruction process. In the reconstruction game, the actions
taken by the various agents are dependent on the outcomes each expects. Coyne
Book Review 83
presents his central point regarding expectations as such: “if the expectations of the
citizens of the country being reconstructed are aligned...with the aims of the reconstruction,
there will tend to be a greater degree of coordination and cooperation” (73).
Along the same lines, the book argues that if expectations of the indigenous population
does not match the outcome of the reconstruction, conﬂict persists. Coyne
asserts that expectations should be low, but not so low as to dissuade credible commitment
by agents involved in the reconstruction.
The book oﬀers a good example of how low expectations can prohibit the establishment
of liberal democratic institutions. After War points to the case of Haiti
where the “repressive history of national institutions” marked by “coercion and predation”
have fundamentally altered citizen perceptions, creating widespread doubt
in government institutions (151). The mid-1990s U.S. occupation of Haiti was met by
widespread defection by indigenous actors. Coyne attributes this to pervasive skepticism
among Haiti’s lower class toward the reconstruction process. Due to historical
patterns of repression, the Haitian elite lack credibility, and were seen as selﬁsh players
in the reconstruction process. The book claims that as a result, even though the
occupation succeeded in keeping U.S.-backed leader Aristide in power, it did not succeed
in strengthening liberal democratic institutions in Haiti. Worse, Aristide continued
on to siphon international aid money and suppress the domestic population.
The spread of liberal democracy has received some attention in recent literature.
Democracy without Borders, by Mark F. Plattner, discusses the relationship between
liberalism and majority rule. The book argues that the ideals of liberal democracy
– the individualism of liberal society and the collectivism of majority rule – are not
inherently compatible, and this incoherence makes it diﬃcult to build new democracies.
The 2006 release, The Spirit of Democracy, by Larry Diamond, holds a more optimistic
view of the global desire for democracy. The book argues that liberal democracy
can spread under the right conditions, namely, in an environment that promotes
good governance and shared economic prosperity.
A major message from the book is that the dynamics of reconstructions are changing.
The successes of previous reconstruction are not good predictors for future endeavors.
Cultural and historical forces that aﬀect reconstruction have changed in the
84 New Perspectives on Political Economy
past sixty years. The world has changed since the Cold War when two world superpowers
created a geopolitical stalemate. Today, the author argues, failed and corrupt
states present the major challenge to America.
After War concludes by oﬀering two alternative means to social and political reform:
principled non-intervention and unilateral free trade. The author asserts that
these two alternatives oﬀer “liberal means to liberal ends” (173). The book emphasizes
that because free trade is voluntary, political change that stems from free trade
will lack the bitter aftertaste typical of military occupation. Free trade also provides
mutual economic and cultural gains. Countries exchange not only goods and services,
but moreover, countries exchange ideas and institutions and incorporate the
best aspects of each culture to produce greater eﬃciency in both societies. The book
notes that although non-intervention and free trade are not the most direct route to
liberal democracy, these strategies sidestep many of the geopolitical problems that
can impede reconstruction.
The book’s great weakness is that its discussion overlooks relevant actors in the
reconstruction game. After War takes a two-pronged approach which looks at agents
of the occupying country and indigenous actors in the occupied county. Unfortunately,
this approach does not suﬃciently address the role of third party actors in
the process. Third parties comprise international agents like the United Nations. International
institutions play a key role in reconstruction because they inﬂuence the
foreign relations between countries. Third parties also include regional powers, such
as Iran in the case of the Iraq reconstruction. These third parties are relevant players
in the reconstruction game. They often have a vested interest in the outcome. In
the case of a regional third party, failed reconstruction can produce dreadful spillover
eﬀects that threaten a region’s stability.
After War is a sophisticated piece of writing. Although the book packs in a lot of
theory, it is never inundating. It draws upon many schools of thought and requires the
reader to integrate various concepts in order to fully digest the book’s message. The
book beneﬁts from its evolving structure. After three chapters emphasizing theoretical
arguments, the historical narratives of the third section are a welcomed surprise.
The author does a great job of reintroducing the central themes in each chapter; and
because of this, the book’s message resonates throughout. By identifying the moti-
Book Review 85
vations behind the actions of participants in post-war reconstruction, one can design
policy that will build a cooperative environment and ensure liberal democracy will
prosper.
After War is an essential read for students of economics and public policy who
wish to gain a better understanding of the forces that aﬀect the spread of liberal
democracy. Even the casual reader will ﬁnd some thought-provoking asides that will
add to any coﬀee table discussion of America’s role in the world. The book is a refreshing
look at the spread of liberal democracy.
Brian Mokoro
Duke University