Analyse & Kritik 30/2008 ( c Lucius & Lucius, Stuttgart) p.

723733
Kurt W. Rothschild
Economic Imperialism
Abstract: Economic Imperialism is the claim of some economists that the methodology
of neoclassical economics has superior scientic qualities and should be adopted by
most or all social sciences. The paper rst shows why such a dominant claim could
develop among economists but in no other science and then goes on to point out the
shortcomings of this claim of methodological superiority. These critical remarks are
also relevant for methodological controversies within economics between a mainstream
and heterodox economists.
There are only very few cases where a lable attached to a certain object of
investigation can have such double meaning as the term Economic Imperia-
lism. Its rst meaning had its birth at the turn of the 19
th
century with its
use by Hobson and Lenin for describing the economic exploitation of colonies
and underdeveloped countries by the big imperial nations. The second mea-
ning, which is to be discussed in this paper, had its origin in the second half of
last century. It was triggered by the work of Gary S. Becker (Becker 1976)
1
and
refers to the claims and attempts to show that the methods used in neoclassical
economic theory should be exemplary for all social sciences (Lazear 2000; Rad-
nitzky/Bernholz 1987). This proposition is based on the assumption that the
methods of neoclassical economics are more scientic (or the only scientic
ones) than those used by other social sciences irrespective of the subject mat-
ter. In many cases it is suggested that they should always obtain priority. This
demand is also directed towards so-called heterodox economists who use other
methodologies (Keynesians, institutionalists, evolutionary economists, Marxists
etc.). There is no question of a give and take, a mutual learning process from
dierent methodological needs and approaches.
In fact one can distinguish between two forms of Economic Imperialism (EI).
A milder form, sometimes called the economic approach, which recommends
the application of the neoclassical method in all social sciences but admits that
other basic methods may be useful too, and economic imperialism in a narrower
sense which looks at the economic method as the only or at least most dominant
path for scientic discovery. In the latter case neoclassical economic methods are
1
The 17 volume International Encyclopaedia of the Social Sciences published in the year
1967 does not contain a reference to this type of economic imperialism.
724 Kurt W. Rothschild
regarded as the standard to be adopted by all.
2
But what the milder economic
approach and the more aggressive economic imperialism have usually in com-
mon is their belief in the superiority of the economic method. It is regarded as
the only method applicable to all social sciencespsychology, sociology, political
science etc.
The exact characteristics of this method which are or should be applied are
not always dened in the same way, but they all rest on the traditional metho-
dological elements of neoclassical economics. Thus for Radnitzky and Bernholz
the decisive fact is that human action is directed towards maintaining life in its
widest sense, and this demands choice and action in a world of scarce resources.
Acting rationally, using appropriate means to achieve ones ends, in-
volves essentially taking into account costs and benets, whereby
costs and benets are taken in its widest sense. Since resources
are always scarcelast not least our life-timerational conduct is
governed by principles of economics. Hence, it is plausible that new
and important knowledge can be gained by applying the perspective
of economics, the methods and tools of economicssome of them
suitable generalizedto elds of enquiry that have traditionally be-
en thought to lie outside the competence of economics. (Radnitz-
ky/Bernholz 1987, VIII)
For Lazear EI is justied because only economics has a rigorous methodological
framework which is centered around three basic themes: (1) rational maximi-
zing individual behaviour, (2) equilibrium, and (3) eciency (Lazear 2000, 100).
Reder denes EI rather widely as the general adoption of a set of analytical pro-
cedures connected with the neoclassical RAP (Resource Allocation Paradigm)
and its associated econometric techniques (Reder 1999, 344). It is stressed that
the rigourness of the axioms excludes the use of ad hoc argumentation.
The endeavour to preserve the dominance of the neoclassical methodology ge-
nerally and particularly in economics itself is illustrated very clearly in a recent
review of a book on Behavioural Economics (Diamond/Vartiainen 2007) in the
Economic Journal of June 2008 (118/529). In a friendly and appreciative review
of the book (Binmore 2008) Ken Binmore agrees that behavioural economics
oers interesting and important contributions, which are of course oftenlike
contributions from other heterodox economic theoriesdiverging from and con-
tradictory to neoclassical approaches. But then he goes on to warn that the
solution is not to throw away the successes of neoclassical economics and revert
to the nave [sic!] psychology of the economists of the Victorian era. We need to
hold on to neoclassical theory when it can be shown to work in practice and to
rene it where it does not (251; my italics). Nobody will object to holding on
to neoclassical theory when it can be shown to work in practice, but why should
that not also hold for behavioural economics and its renement?
2
The most aggressive economic imperialists aim to explain all social behaviour by using the
tools of economics. (Lazear 2000, 103) See also Stigler (1988) where he stresses the superiority
of the economic methodology (191205).
Economic Imperialism 725
Discussions and quarrels about methodological questions relating to science
3
in general and to specic branches of science in particular have a long and li-
vely history. Philosophers, methodologists, and practical scientists have debated
and continue to debate the contents and quality of methods which should be
permitted and applied in order to regard the results of studies as scientic.
4
These debates can be very heatedclassically shown in economics in the famous
Methodenstreit between Schmoller and Mengerand though not necessarily
leading to generally accepted results help the acting scientists to be aware of the
cognitive aspects of their work. This awareness is sharpened by the interchange
and mutual stimulation of ideas. In this ongoing methodological discussion EI
occupies a special and unique position. It is the only methodological prescrip-
tion emanating from a special eld which postulates exclusively for itself to be
applicable and often preferable in all branches of social science. Psychologists,
sociologists, biologists etc. would never dream of such imperialistic claims.
How can the emergence of this special position of a certain type of economic
methodology be explained? To answer this question one has to roam about a
bit over a somewhat wider eld. The study of economic phenomena is dealing
with the contents and consequences of human actions in a restricted eld. In this
respect it is in the same boat as all social sciences which dier only with regard
to the special eld with which they are concerned (politics, society etc.). When
economic is understood in its widest sense, viz. referring to all actions connected
with the production and consumption of goods and services, economics would
cover a very wide eld indeed. For instance the actions of housewives would
be partly caring (e.g. dealing with children), partly social (e.g. charitable
activities), and partly economic (e.g. cooking). Economic motives and activities
would act side by side with other aspects and motives. There would be no reason
to regard economic action as something special to be treated as outside and/or
dominant in the study of social facts. There would be no need to have in some
universities a Department of Economics and Social Sciences. Department of
Social Sciences would suce.
But the situation became dierent with the emergence of a more or less
universal dominance of anonymous markets and market processes regulated by
competition and prices. The term economic ceased to refer to all individual pro-
ductive and consumptive activities and became restricted to the description and
analysis of market processes (in the widest sense). In the analysis of these pro-
cesses the main problem was to discover and explain how anonynomous markets
and the impersonal pricing process could bring about socially acceptable and
even ecient macro-economic results. Adam Smiths hidden handthe pricing
processopened the path for the development of a special place for economics in
the eld of social sciences. This Smithian revolution found a reformulation and
partial revision in the so-called marginal revolution of economic theory in the
seventies of the 19
th
century which provided the basis of what we call neoclas-
sical economics. It was simultaneously and independently introduced in three
variants by Menger in Austria, Jevons in England, and Walras in Lausanne. It
3
I use science in its wider sense comprising both natural and social sciences.
4
Or in a more demanding formulation as a path to discover truth.
726 Kurt W. Rothschild
is the Walrasian approach and the Walrasian methodology which has become
with many modications and renementsdominant in present-day neoclassical
economic theory and which lies behind the EI-postulate. To this background we
now turn.
In order to understand what Walras was driving at it is useful to remember
that the 19
th
century had seen an enormous expansion in the development of
the natural sciences in general and of physics and chemistry in particular. This
growth in knowledge and its successful application in practice had created great
prestige for Science in its narrower sense of natural science. This led to a desire
among many social scientists to achieve similar theoretical foundations as the
natural sciences and a comparable exactness in their conclusions. The achieve-
ments of mechanical physics acted as a particular attraction. This background
had a profound inuence on Walras (Mirowski 1989) and its stamp is left on
later developments of neoclassical economic theory.
We now turn to those aspects of the structure of Walrasian neoclassical theory
which are relevant in connection with the EI-problem. Looking at the economic
process as a whole (and trying to explain it) one can divide it into two interde-
pendent parts. On the one side we have the individual actors with their decisions
regarding consumption and production within the market and on the other si-
de we have a complex interrelationship of all prices (including wages, interest)
which results in a certain equilibrium in the competitive markets. The indi-
vidual decisions inuence the production and pricing processes and are in turn
inuenced by them.
To analyse this complex relationship one has to deal with decision processes of
the individuals on the one hand and the working of the interrelated price eects
of the market mechanisms on the other. As far as the rst aspect is concerned
economics is confrontedlike all social (or human) scienceswith the actions,
the motives, the idiosyncrasies of human beings. The situation is dierent when
one deals with the interrelationships of prices and quantities on the markets.
These are dependent on objective (physical) interdependencies shaped by pro-
duction processes in a world of scarce resources (raw materials, human labour,
time). In a fully employed economy every economic act of obtaining additional
goods or services for consumption or investments (benets) entails necessarily
(because of the scarcity problem) a renunciation of some means of production
for other elements of consumption or investment (costs). To obtain more motor
cars (or military equipment etc.) one must give up some part of the production
of butter, or schools, or . . . There is no free lunch. Thus as far as market pro-
cesses are concerned there exists a complex network of interrelations between
alternative benets and costs based on objective interdependencies which are
made comparable by the possibility of measuring them under a single common
aspect: the prices established in competitive markets.
It is this latter part with its mechanical aspects of market interrelationships
which is a specic characteristic of economics which enabled Walras and his
tradition to develop an axiomatic theoretical structure with an anity to the
exactness of physical science. The creation of such a rigorous formal model for
the study of a complex network of interrelated markets was and is without doubt
Economic Imperialism 727
an ingenious achievement which has its uses in many elds of economic analysis.
But it is notas some claima sucient or suitable methodological basis for
economics as a whole, let alone for social sciences in general.
Two aspects have to be noted in this connection. One is the fact that this
mechanical picture is only one part of the total analysis of the economy, and
the second is that this part is a specic element of economics. As far as the rst
item is concerned we are faced with methodological problems which are connec-
ted with the diculty of combining the market mechanism with the individual
actions which inuence the market events and are inuenced by them. When
this important question has to be incorporated the human element with all its
diversities and diculties enters and threatens to destroy the exactness and de-
terminism of the market model. All the diculties and uncertainties with which
psychology, sociology etc. have to ght in view of the many-sidedness of human
nature would enter and disturb the formal rigour of the neoclassical economic
model and its results. The way out of this diculty taken by the neoclassical
approach was the adoption of an extremely abstract and simplied picture of
human motives and decision-making when acting in the market sphere. This
picture is embedded in the representative homo economicus, the rational and
informed individual who aims at and usually achieves an optimal result from his
market-based actions (maximization of utility and maximization of prots).
The fact that human beings dier with regard to their wishes and preferences is
set aside by taking these dierences as given facts which are not and/or cannot
be ascertained. All that is necessary for the completion of the neoclassical mo-
del is that the rational actors base their actions in the market sphere (i.e. their
purchases and sales in the widest sense) on the common aim of optimising their
given individual preferences. In doing this they are constrained by the budgets at
their disposal. Any expenditure to obtain preference-satiscing goods or services
involves a reduction of some purchases of other goods or services, i.e. the bene-
ts obtained by the purchase involve costs (opportunity costs) of renouncing
some other possibilities. The assumption of optimising behaviour involves for
all market actions the weighing of such benets in relation to costs
5
and this
calculus is dependent on the ruling market prices of all goods and services. The
logic of the model is now closed. Human actions and the mechanical productive
processes are interlinked through a price system which is determinedin the
environment of competitive marketsby a balancing of costs and benets in a
world of material and budgetary constraints. An equilibrium is achieved when
the price structure (or alternatively: demand and supply) corresponds to the
constrained aims of the individuals and the interrelations of scarce goods and
materials.
6
Before going on to a critical look at the priority claims of neoclassical me-
thodology I want to make a short remark on the concept homo economicus in
order to prevent possible misunderstandings. Some people resist the use of this
5
Jevons, a contemporary of Walras in laying the foundations of neoclassical economics,
talks of a calculus of pleasure and pain.
6
The picture of dierent forces inuencing each other and leading to a clearly dened
equilibrium illustrates the anity of the neoclassical model to the laws of mechanical physics.
728 Kurt W. Rothschild
concept on moral and psychological grounds because they see in its assumption
of optimising (maximizing utility) behaviour with regard to ones preferences
the characteristics of a perfect egotist. This is a misunderstanding. Optimising
the satisfaction of ones preferences does not say anything about the contents
of such preferences. These can be altruistic as well as egotistic so that altrui-
stic actions will be preferred to some egotistic advantages. Thus the objections
to the Homo Economicus on moral grounds are not justied. The problems lie
behind the assumption of optimising as a unique and clear-cut strategy which
rests on the assumptions of clear and stable preferences, rationality, knowledge
of (all?) alternatives, and of the benets and costs connected with them. It is
this demanding collection of requirements which alone would make optimising
strategies a doubtful fact for describing and understanding human action, even
if restricted to market environments. Added to this is the growing evidence (1)
that preferences are not stable and do not only inuence events but are also
inuenced by events, and (2) that motives are not rational in the narrow sense
of xed consequences based on given preferences, but are inuenced by the past,
by surrounding circumstances, habits, traditions, institutions etc.
All this means that the gap between the theoretical homo economicus and
the economic individual in the real world is considerable. That a gap exists is of
course no reason for rejecting a theory. Theories rest on abstractions from a com-
plicated reality. But these should be carefully chosen and two conditions should
be fullled: One should be fully aware of the abstractions and connected with
this one should be careful when one applies the theory to real problems. This
creates already a problem for the question of the status of the neoclassical model
within the eld of economic science andas we shall see laterthese diculties
are intensied when we turn to social sciences in general. As far as economics
is concerned the neoclassical approach with all its limitations has certainly a
role to play. To deal with problems which aect in particular the working of the
complex market mechanisms it is helpful to neglect the diversity of individual
reactions in order to make the analysis more manageable and eective. Neoclas-
sical economics has therefore a considerable eld for application. But there are
also many economic problems where the reality of irrationality of consumers
and producers plays an important role, where we have shifting and conicting
preferences, limits in information, uncertainty about the future, imitative be-
haviour etc. which all can be particularly prominent under certain economic
conditions and relevant for problem-oriented analysis. This fact is mirrored in
the existence of several alternative economic models and theories (e.g. Keynesi-
an, Marxist, institutionalist, behavioural, evolutionary, feminist economics etc.),
each with its own methodological apparatus and with its own limitations and
elds of applicability. This multiparadigmatic situation can be regarded as re-
grettable, but it is unavoidable when one has to deal with an extremely complex
and dynamic economic reality.
7
7
In a paper presented in Japan in 1931 Schumpeter said: No method is good for all pro-
blems, but everyone has a place and it is for this reason that I think so little of methodological
controversies. (Swedberg 1991, 285)
Economic Imperialism 729
The trouble with neoclassical theory is therefore not that it has weak points
and is not generally applicable. These characteristics it shares with all existing
theories. The trouble is that some neoclassical economists claim a superiority
and scientic exclusiveness of their approach vis--vis all the other approaches
which is based on their capacity to presentthanks to their simplications and
abstractionsa coherent picture of competitive markets and the price system
which resembles the achievements of the exact natural sciences. Other economic
theories which point out and introduce other relevant factors do not and do
not want to accept the strict axioms used by the narrow market- and price-
oriented neoclassical theory. To capture the inuence of human diversity and
other non-economic factors they necessarily use other methods which cannot
be forced easily into the deterministic neoclassical framework. But they have
to be considered if one wants to get nearer to the problems of a dynamic and
fuzzy reality. Instead of acknowledging the need for the coexistence of these
dierent theories side by side with the neoclassical tradition and its methodology
there is a tendency to draw a sharp line between the neoclassical school and its
ramications on the one hand and all the other basic theories on the other. There
is no other social science which has such a sharp division between a mainstream
which demands and obtains a leading position in research and teaching, and
heterodox theories covering all the other (non-neoclassical) theories.
While the claim for dominance of a given methodology is not justied even
within a certain discipline, it can at least be excused as an attempt to make it
more widely accepted. But the demand becomes absolutely unacceptable when
it is extended to all social sciences, i.e. when it becomes EI. This claim is based,
as already mentioned, on the use of a deterministic and axiomatic methodology
which has an anity to the methods used in parts of physical science and is
therefore regarded as superior. The power of economics lies in its rigour, writes
the well-known US economist Edward Lazear. Economics is scientic; [it] suc-
ceeds where other social sciences fail because economists are willing to abstract.
(Lazear 2000, 102)
8
There are several aspects which undermine this argument. They will be here
indicated without going into all the details connected with them. First of all it
must be stressed that the claim that the rigidity of the neoclassical model is
the optimal guaranty for its scientic character and for reaching reliable re-
sults cannot be maintained. Rigidity of theoretical foundations is certainly a
desirable quality but it cannot be reasonably maintained when the complexity
of the subject matter does not permit such a restriction and demands openness
for methodological diversity. This is not only true for the social sciences but for
some branches of the natural sciences as well whereas in the social sciences
complexity and dynamics are considerable and experiments impossible or dif-
cult (e.g. meteorology, climatology). In fact it can be shown that progress in
the elds of natural science has often beneted from methodological diversity,
from ad hoc additions, and from plausibility considerations (Feyerabend 1975)
which are all abhorred by EI advocates. Their ideal is inuenced by the special
8
It must be kept in mind that in Lazears paper economics is equivalent to mainstream
(neoclassical) economics.
730 Kurt W. Rothschild
aesthetic and functional qualities of theoretical physics where a high degree of
rigidity is obtainable.
9
The fact that the special methodological qualities of theoretical physics are
not (exclusively) capable or ecient to explain subjects in all natural sciences
is of course even more true when we turn to the social sciences. It has already
been pointed out that one problem these sciences have in common is the human
element with its wide diversity. And it was shown that in the study of prices and
actions in competitive markets this problem is settled in neoclassical economics
by reducing the human element to a rational maximization process under given
preferences and budget constraints. This together with the matrix of the market
network and competition provides the basis for a general cost-benet driven
equilibrium result. Irrespective of the question how useful or restricted this
approach is when dealing with economic problems it cannot be applied easily
one-to-one to other social sciences which are dealing with other aspects of social
reality and the related human actions.
The main dierence is perhaps that the picture of the rational maximizing
homo economicus cannot be (easily) tted into social sciences dealing with non-
economic spheres (e.g. psychology and sociology). One aspect springing imme-
diately to mind is that a principal task of psychology is to explain dierences
between people, their motives, and their actions. Useful as the assumption of the
standardized homo economicus may be in some economic studies, it cannot be
a sucient basis for the study of human diversity. This consideration is rejected
by EI-ideologists with the comment that what human action has in common in
all spheres is the problem of choosing between alternatives in a world of scarce
means (in the widest sense: money, time, reputation etc.). The unifying advan-
tage of EI is therefore that with its assumption of rational and informed choice
behaviour it provides a universal basis for all social sciences. But this claim
cannot be maintained for several reasons. First of all even if one accepts the
assumption that general rational maximizing behaviour in a world of scarcities
is a central aspect for all social sciences, there remains the question why dierent
people act in dierent ways and why they change their actions. In other words
what psychology and other social sciences need to explain is not (or not only)
the choice process as such, but the preferences and the change of preferences
which explain individual behaviour. The methods for attacking these questions
can be far more important than the EI approach (which rests on the assumption
of stable given preferences
10
).
But more important is that the choice as such becomes increasingly ques-
tionable once one leaves the idealized world of neoclassical market theory. The
careful weighing of alternatives for choosing an optimal result may be to so-
me extent plausible and acceptable when we deal with choice decisions in cases
where budgets provide a tight and measurable strait-jacket and where most al-
9
Economists insist on a physical-science-style equilibrium as part of the analysis. (Lazear
2000, 101) This is most clearly mirrored in explaining equilibrium as the exact result of a
balance of forces (demand and supply, costs and benets).
10
De Gustibus Non Est Disputandum is the telling title of a paper by Becker/Stigler
(1977).
Economic Imperialism 731
ternatives have market prices which provide an obvious common measure for
making comparisons between alternatives. The weighing of costs and benets
for each alternative becomes possible and an optimal choice can be achieved.
Indeed Cost-Benet-Analysis can be regarded as a shorthand description of the
economic approach. It is the approach which is regarded as applicable and
recommended for use in all social sciences.
11
An immediate and fundamental re-
servation against this recommendation can be based on the extended attacks on
the use of the homo economicus even in economic analysis on the grounds that
the assumptions regarding rationality of choice are too far away from reality.
Psychological research in particular has contributed much to point out existing
gaps (Ariely 2008). Even if such reservations could be disregarded in economics
they are decisive when one deals with non-economic sectors where questions of
motivation are dominant and/or measurement is dicult.
When the economist speaks of rational choice he has in mindquite realistic-
allythe decision to achieve a given alternative in the most ecient way, i.e. by
renouncing a necessary quantity of other alternatives which are of least import-
ance to the person concerned. How much has to be given upthe opportunity
costsdepends on the prices of all the alternatives. A good choice is an ecient
choice and the economic approach regards this as a good picture of human choi-
ce both in the positive (real) and normative sense. But in many other elds of
human action (marriage, getting children, education, health, democracy, crime
etc. which all have been subjected to the economic approach
12
) this question
of eciency is not the only or normal motivational factor. Not only the ques-
tion of what one wants is important, but also the question what one ought to
do turns up, i.e. moral and cultural factors are inuencing decisions. This does
not only point to the fact that they are to be considered as a background; they
also change the decision process. The wants and the oughts can come in-
to conict and actions involve not only an eciency calculus but also moral
doubts and decisions. To study these the mere reliance on cost-benet conside-
rations is hardly satisfactory. One can, of course, argue (and something like this
is sometimes happening) that the nal actual action is always the result of an
optimising cost-benet process in which the oberservance or breaking of moral
or reputational rules enter as benets or costs respectively. Adding them to
the material benets and costs yields the nal result. But such an argument
would not explain the choice process as the result of an optimising process, it
would make that process a predetermined assumption. In other words, in many
social elds a good choice in the normative sense is not only or even not at all
an ecient choice but is just a good choice.
Again neglecting all the diculties already mentioned and looking at non-
economic processes in which moral and other special motivating factors are not
important so that benecial and cost aspects can be compared we are still con-
11
Man is a chooser. All rational choices involve the weighing up of benets and costs [. . . ]
[t]he key concepts of Cost-Benet-Analysis can be generalized. Not only in business dealings
but also in daily life we operate within the CBA-frame, most of the time. (Radnitzky 1987,
285/86. Italics in the original)
12
For an indication of some examples see Radnitzky/Bernholz 1987 and Lazear 2000.
732 Kurt W. Rothschild
fronted with a measurement problem. Many elements and particularly those on
the cost side will be non-market phenomena like time, health, life expectan-
cy, pleasure and pain etc. which can be enumerated but have no secure basis
for an exact and/or agreed valuation which could serve as a basis for reliable
comparisons and the achievement of optimum and equilibrium results.
13
The
attempts to nd reasonable valuations for such non-economic factors are some-
times ingenious, but necessarily subjective and can dier widely in some cases
(e.g. valuations of dangers to health, reputational loss etc.). This problem ari-
ses occasionally also in economic Cost-Benet studies, but there it is usually
restricted to a few items for which no actual or shadow prices are available. On
the whole economics has the advantage of prices as an objective measuring rod
which one can accept or not but which provides a rm basis that is missing in
many elds of the other social sciences.
In view of these and some other minor diculties one can understand that
doubts regarding the EI priority claims are expressed by scientists in non-
economic social sciences and by many economists
14
as well. Even a well-known
political scientist who is in strong favour for learning from neoclassical eco-
nomic methodology reserves half of his book to point out the limitations of it
(Rhoads 1985). Yet in spite of such doubts and criticisms the EI ideology and
papers inspired by it survive. How can this be explained?
A motivational factor among neoclassical economists can of course play a
part. Who would not like to see his approach to be regarded as the most perfectly
scientic approach, his science as the queen of social sciences. But while such
motives probably exist I am far from suggesting that it refers to all or even many
economists, andat any ratethis could not by itself explain the persistence of
EI. The explanation rests probably on a fact that has already been mentioned
in connection with the discussion of the methodological divisions within the
economic community. There are problems and constellations in many spheres
of human relations which suggest similarities and analogies to situations met in
market processes and which permitwithout being forced to bring in outlandish
assumptions
15
the use of some sort of Cost-Benet-Analysis. In such cases the
application of economic methodology may not only be possible but sometimes
13
A forerunner of such attempts of Cost-Benet-Analysis was perhaps the philosopher Leib-
nitz of whom the following anecdote is told. When he considered marrying a lady for whom he
had some sympathies he sat down and wrote a list of all the advantages the marriage would
entail, and side by side he set a list with all the disadvantages involved. Seeing that the list of
disadvantages was longer than the list of advantages he decided to give up the idea of marriage.
14
As an example I want to quote a passage from the Thnen-Lecture of the Nobel laureate
economist Reinhard Selten (my translation). A theory of bounded rational behaviour cannot
be built on the assumption that decisions are based on consistent preferences and probability
judgements. Consistency cannot be expected [. . . ]. The target foundations of the traditional
micro-economic theory, utility and prots, can be called optimisation targets (italics in the
original). In practical decision situations it is not even possible to aim at optimal targets. A
theory of bounded rationality will have to describe decision processes which are not optimi-
sing, not even approximately. Nothing exists which could be optimised. Several factors are
considered, but this is not done through maximizing some function, but in a manner which is
fundamentally dierent [. . . ]. There is no function to be maximized. (Selten 2000, 152/53)
15
As a recent example of the questionable nature of costs and benets see for instance the
study of the economics of indelity by Elmslie/Tebaldi (2008).
Economic Imperialism 733
even superior. This is shown by several papers written in this spirit which have
opened new vistas on special aspects which were not taken into account by the
traditional methods used in the respective social science. But these success stories
do not support the EI ideology with its claims for universality and superiority.
They rather support the view that in the complex world in which we live we
need several theories and methods as a box of tools from which one has to
choose according to the nature of the object and the questions asked. Analogies
and heuristic elements can and do play a role in this process. But there is no
reason to assume that one method (which is particularly suited for a certain
branch!) is a superior ideal to be adopted in all cases. What is required is a give-
and-take relationship and this is also true for economics which benets from
the research results of other social sciences obtained by dierent methods (Frey
1993; Himmelstrand 1992).
Bibliography
Ariely, D. (2008), Predictably Irrational: The Hidden Forces That Shape Our Decisi-
ons, New York
Becker, G. S. (1976), The Economic Approach to Human Behaviour, Chicago
/G. Stigler (1977), De Gustibus Non Est Disputandum, in: American Economic
Review 67(2), 7690
Binmore, K. (2008), Review of Diamond/Vartiainen (2007), in: Economic Journal 118,
F242F251
Diamond, P./H. Vartiainen (2007) (eds.), Behavioural Economics and its Applications,
Princeton
Elmslie, B./E. Tebaldi (2008), So What Did You Do Last Night? The Economics of
Indelity, in: Kyklos 61(3), 391410
Feyerabend, P. (1975), Against Method: Outline of an Anarchistic Theory of Know-
ledge, London
Frey, B. (1993), From Economic Imperialism to Social Science, in: Public Choice 77(1),
95105
Himmelstrand, U. (1992), Interfaces in Economic and Social Analysis, London
Lazear, E. P. (2000), Economic Imperialism, in: Quarterly Journal of Economics 115,
99146
Mirowski, P. (1989), More Heat than Light, New York
Radnitzky, G. (1987), Cost-Benet Thinking in the Methodology of Research: The
Economic Approach Applied to Key Problems of the Philosophy of Science, in:
Radnitzky/Becker, 283-331
/P. Bernholz (1987), Economic Imperialism. The Economic Method Applied Outsi-
de the Field of Economics, New York
Reder, M. W. (1999), Economics. The Culture of a Controversial Science, Chicago
Rhoads, S. E. (1985), The Economists View of the World. Government, Markets and
Public Policy, Cambridge
Selten, R. (2000), Eingeschrnkte Rationalitt und konomische Motivation, in: Schrif-
ten des Vereins fr Socialpolitik 274, 129157
Stigler, G. (1988), Memoirs of an Unregulated Economist, New York
Swedberg, R. (1991), Joseph Schumpeter. The Economics and Sociology of Capitalism,
Princeton