And what does itmean when we model actors as rational, but they in practice follow individual,diversely distributed heuristics or “ersatz-economics” as Klamer terms the variouseveryday t
Trang 1Ethical Economy Studies in Economic Ethics and Philosophy
Birger P Priddat
Communication and Economic
Theory
How to deal with rationality in a
communicational environment
Trang 3Ethical Economy Studies in Economic Ethics and Philosophy
Series Editors
Alexander Brink,University of Bayreuth
Jacob Dahl Rendtorff,Roskilde University
Editorial Board
John Boatright,Loyola University Chicago, Chicago, Illinois, USA
George Brenkert,Georgetown University, Washington D.C., USA
James M Buchanan{,George Mason University, Fairfax, Virginia, USA
Allan K.K Chan,Hong Kong Baptist University, Hong Kong
Christopher Cowton,University of Huddersfield Business School, Huddersfield,United Kingdom
Richard T DeGeorge, University of Kansas, Lawrence, Kansas, USA
Thomas Donaldson,Wharton School, University of Pennsylvania, Philadelphia,Pennsylvania, USA
Jon Elster,Columbia University, New York, New York, USA
Amitai Etzioni,George Washington University, Washington D.C., USA
Michaela Haase,Free University Berlin, Berlin, Germany
Carlos Hoevel,Catholic University of Argentina, Buenos Aires, ArgentinaIngo Pies,University of Halle-Wittenberg, Halle, Germany
Yuichi Shionoya,Hitotsubashi University, Kunitachi, Tokyo, Japan
Philippe Van Parijs, University of Louvain, Louvain-la-Neuve, Belgium
Deon Rossouw, University of Pretoria, Pretoria, South Africa
Josef Wieland,HTWG - University of Applied Sciences, Konstanz, Germany
For further volumes:
http://www.springer.com/series/2881
Trang 4and Economic Theory
How to deal with rationality in a communicational environment
Trang 5Birger P Priddat
Universita¨t Witten/Herdecke Pra¨sident
Witten
Germany
ISSN 2211-2707 ISSN 2211-2723 (electronic)
ISBN 978-3-319-06900-5 ISBN 978-3-319-06901-2 (eBook)
DOI 10.1007/978-3-319-06901-2
Springer Cham Heidelberg New York Dordrecht London
Library of Congress Control Number: 2014941627
© Springer International Publishing Switzerland 2014
This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part
of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed Exempted from this legal reservation are brief excerpts
in connection with reviews or scholarly analysis or material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work Duplication
of this publication or parts thereof is permitted only under the provisions of the Copyright Law of the Publisher’s location, in its current version, and permission for use must always be obtained from Springer Permissions for use may be obtained through RightsLink at the Copyright Clearance Center Violations are liable to prosecution under the respective Copyright Law.
The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.
While the advice and information in this book are believed to be true and accurate at the date of publication, neither the authors nor the editors nor the publisher can accept any legal responsibility for any errors or omissions that may be made The publisher makes no warranty, express or implied, with respect to the material contained herein.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)
Trang 6This book deals with different topics on communication, language, and decisionswithin the field of economics and markets The chapters give insights into newaspects of economic processes and are reflecting economics in a particularnew way.
It is argued that in markets most of the transactions are embedded in multiplecommunicational processes (Chaps 4 and 6) This is no cheap talk, but highlyrelevant for determining supply and demand (Chaps.2,6, and7) Relevance andacceptation of new products in dynamic markets, for example, get launched byadvertising and promotion, which is the very branch of an economics of persuasion.The markets are not only stringed by networks (Chap.2) and stabilized by institu-tions (formal and informal ones) (Chaps 2 and 5), but also overwhelmed by
‘semiosphere’ (see Chap.6), semantical and semiotical fields, producing differentmeanings, relevancies and intents
20 October 2013
v
Trang 7.
Trang 81 The New Population of Economics: Multiple, Fair, Ignorant and
Emotional Actors How Are the Markets Ordered in Accordance
with Diversified Knowledge Bases? 1
1.1 Many “I”’s 3
1.2 The Modern “I”: In a Language Game 9
1.3 Embedded Individuals: Networks 12
1.4 New Dynamics 14
References 17
2 Mutual and Self-Enforcing Agreements Contracts as the Basic Institution of Economics: Network Knowledge Instead ofRational Choice 21
2.1 Mutuality of Advantage 23
2.2 Contractual Processes 29
2.3 From Choice to Contract (I) from Contract to Cooperation (II) 31
2.4 Summa and Amplifications: Contracts and Network Environments 33
References 36
3 Morals: Restrictions, Metapreferences: Adjusting an Economics of Morality 39
3.1 Morals as Restriction: G.S Becker’s ‘Economic Approach’ 39
3.2 Morality as Metapreference: A Sen et al 45
3.2.1 ‘Belief Systems’, Communication and Changing Preferences 49
3.2.2 Preference and Semantic 51
3.3 Morality as Resource 55
References 61
vii
Trang 94 Communication of the Constraints on Action K J Arrow
on Communication 65
4.1 Introduction 65
4.2 ‘Judgement’ Versus ‘Rational Choice’ 65
4.3 ‘Probabilities’, ‘Common Experience’ and ‘Convergence of Beliefs’ 67
4.4 Communication as Learning Process 71
4.5 Communication as Mechanism and Interpretation 73
4.6 Arrow’s Scepticism 75
4.7 Communication as Equilibrium of the Second Order 77
4.8 Context Neutrality of Communication 79
References 82
5 Communication as Interpretation of Economic Contexts: The Example of Culture and Economy: D.C North 85
5.1 Culture as Institutions: D.C North 87
5.2 Deficit of Conception: The Unexplained Semantic Dimension 91
5.3 Stability and Contingence: Consequences for the Construction of Economic Theory 95
References 98
6 Ludonarrative Dissonance: Economy as a Diversified Language Game Landscape 101
6.1 For Example: “Stocks” in E and D 108
6.2 The Difference Between E and D 111
6.3 Everyday Theories: TH (D) 114
6.4 Consequences 117
References 123
7 Rationality, Hermeneutics, and Communicational Processes On L Lachmann’s Approach of Hermeneutical Economics 127
7.1 The Hermeneutical Approach of Ludwig M Lachmann 127
7.2 Further Evidences on Hermeneutical Economics 131
7.3 Some Epistemological Problems 134
7.4 Some Critics of Hermeneutical Economics 138
7.5 From ‘Intersubjectivity’ to ‘Communication’ 139
7.6 Possible Consequences 142
References 146
Trang 10The New Population of Economics: Multiple, Fair, Ignorant and Emotional Actors How
Are the Markets Ordered in Accordance
with Diversified Knowledge Bases?
We associate economics with individuals acting rationally in competitive markets,who, so to speak, provide its Basic Law The rationality attributed to the actorsconsists less, however, in their individual competence than in the disposition toachieve the best results for themselves in their exchange transactions.-and this isbased on mutual agreement between the parties Transactions rest upon contractsand contracts, in turn, rest upon congruence and consensus
The fact that we establish this consensus in a routine, automatic, almost scious fashion in every purchase or sales agreement does not dispense us from theneed to take its structure as a form of mutual agreement into account What is atissue is not the best an individual can wish for, but the best that can be achievedunder the competitive conditions of a transaction In this sense it is the tendency tomaximize personal benefit, limited by what the market yields – and this meansultimately that another actor must agree to the transaction Rationality as individualrationality is not in itself an economic concept; only when a bilateral transaction hasbeen performed can we speak of an economic event If we wish to give rationality
uncon-an economic meuncon-aning we must speak of the successful coherence of two individualrationalities; it is exclusively a matter of systemic interference and not of exclu-sively individual actions and decisions In this sense individual rationality is alwaysalready limited by a transaction procedure in which two actors must be successful atthe same time This is the fundamental basis of the idea of balance The generalbalance consists in the sum of all successful transactions
But in the last 20 years economic behavioural research has demonstrated theexistence of so many anomalies, deviations and irrationalities that these can nolonger be seamlessly covered by the assumption of individuals taking rationaldecisions It is in the meantime regarded as an established fact that individuals donot, for example, approve of unfair behaviour If they feel disadvantaged theyprefer to forgo the optimization of their own benefit in order to punish unfair actors.Other forms of consideration are reciprocity, altruism etc (e.g Fehr andFischbacher2004; Fruehwald2009); additional emotional forms come from studies
in neuroeconomics (Frith and Singer 2008; Kabalak and Priddat 2010) ingly, individual benefit maximization is accompanied by other behavioural forms,
Accord-B.P Priddat, Communication and Economic Theory, Ethical Economy 47,
DOI 10.1007/978-3-319-06901-2_1, © Springer International Publishing Switzerland 2014 1
Trang 11which we characterize as norms (social norms (Binmore2011; Gintis2009)), rules
or “informal institutions” (Douglas C North2010) Armin Falk believes that thehomo oeconomicus has already been replaced by the homo reciprocans (Falk2003;conceptually more mature Gintis 2009) Above all, economic behaviour is nolonger free and independent, but context-related The behavioural spectra haveexpanded to include, among others, the analysis of heuristics (Gigerenzer andGaissmaier2011), i.e of simpler substitutes forrational choice (see also behaviour
in singular situations: Karpik2011; Hutter2011) Heuristics is the name given to allpossible self-defined behavioural types, which are not designed to achieve ratio-nality but rather the ability to make decisions and connections
This applies not only to cognitive processes but also to the affects of individuals.Here neuroeconomics has brought new insights, above all the fact that all cognitive(decision) processes are affectively co-determined (Frith and Singer2008; Kabalakand Priddat2010; Priddat2007; Hermann-Pillath2012), which can mean that everydecision which is rationally taken can be implemented completely differently byeach individual on the basis of his affects This finding of a particular department ofbehavioural economics has not yet reached economic theory For it means thatcognitive operations are not valid alone, but only together with the mood of theactor (and/or the situation which affects him) In research on the path “frommolecules to markets”, for example, it turns out that the testosterone level in theblood increases among stock exchange traders when they achieve above-averagegains; in turn, rising testosterone values increase the willingness to take risks Incontrast, in the phases of uncertainty that follow the concentration of stress hor-mones increases (Coates2012) “Persistently high cortisol values have the effectthat the same traders would become excessively anxious and averse to taking risksand make the crisis even worse” (Coates in: Shafy2012) This contradicts, at least
to date, the basic disposition of economics, which argues purely cognitively and isnow confronted with a science (neuroscience) which biomedicinally breaks downthe logical unambiguity (of the statistical mechanics) about which economics feltsecure We are faced with theoretical upheavals which we have hitherto notperceived (or wanted to perceive) Contexts and atmosphere become just as impor-tant as the neurophysiological states of the actors (and hence the relationship ofindividual constitution/context (see also psychoanalytical analyses of economicbehaviour: Gourge´2001))
But if the actors no longer decide rationally but in accordance with a variety ofother criteria, heuristics, etc., there can no longer be any guarantee that the marketwill come up with the best result The proof of the efficiency of actions is, however,theproprium of economics as opposed to other social sciences For this reason themany different economic theories of action in all their variants endeavour to provethat behaviour in each case is rational, albeit with limitations, i.e they argue thanunder specified conditions (constraints, restrictions) no better behaviour is possi-ble The problematic side of this line of argument is that any behaviour can betermed rational if it can be shown that under the prevailing conditions nothing betterwas possible The analytical focus shifts from the explanation of individual behav-iour to the analysis of the conditions, the situation, the constellation of the market
2 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 12and the perceptual abilities of the actors (or their convictions or attitudes:beliefs).This in turn requires a new orientation of attention to the epistemic competence ofthe actors: what do they know, what can they know and what not? It is not only aquestion – as economics prefers to see it – of measuring the amount of informationthe actors possibly do not have at their disposal but also of another competence:how far are they in a position to interpret their situation or their context adequately?
1.1 Many “I”’s
Let us pick out one aspect Markets are interwoven with institutions: legal, informalaction patterns, norms etc (North1992) But if individuals do not only act ratio-nally as individuals but at the same time follow norms and rules – let us keep for themoment just to this case – two differently motivated forms of behaviour aremanifested in a single action Following rules is a completely different kind ofbehaviour from making an individualrational choice Anyone who follows a rule
no longer decides rationally in a situation but follows the primary decision he hadtaken earlier when he agreed to accept the rule (Priddat 2004a; Kabalak 2009:Chap 2; Herrmann-Pillath2011a) He frees himself from the burden ofchoice, asthe rule tells him how he should behave In contrastrational choice is, in a strictsense, anti-traditionalist: it rests upon the maxim that in each situation an individualmust weigh up anew what the best alternative is Every newly takenrational choicewould hence also be anexit from the nexus of rules Because the rule develops a
“balance of expectations” (shared beliefs) for all participants (Aoki 2001) everynewly taken individualrational choice is a breach of the rules, which disturbs theachievement of the institution: thecoherence of beliefs Vice versa, every adher-ence to the rules involves, strictly speaking, anexit from rational choice We mustemphasize this point here, becausebehavioural economics proves that more norms/rules are at play than economics normally concedes
Every individual is, therefore, “two individuals” simultaneously: he is rationallygifted and rule orientated at one and the same time But that is not all Individualshave several forms of behaviour at their disposal which they apply according to thesituation or the context John Elster has termed this themultiple-self – phenomenon(Elster1987) One can speak here of individuals as agents who dispose of severalactors or actor types and use them like roles.Seen in this way the individual is anagency – an agency of himself, who can deploy several actor types In view of thefact that there are many different behavioural types we must assume that they allmanifest themselves, i.e we must define the actor in such a way that he is notdetermined by a definition but can activate different behaviours in accordance withthe situation, context etc (in our example: rational decision or institutional obser-vation of the rules Here the institution is not the framework within which the actorcontinues to make rational choices; as a rule follower he no longer decidesrationally) If such a spectrum of behavioural possibilities exists in the economy,
we can no longer assume that the actors dogmatically keep to a form of behaviour as
Trang 13if it were a rule This means, for example, that rational choice is only onebehavioural possibility among others Instead of assuming that all actors behaverationally it is on the contrary necessary to demonstrate the conditions under whichactors are capable of rational behaviour The recent theory of mechanism designresponds precisely to this circumstance by conceivingdesigns in which the actorscan act rationally once again (Williams2012) But who designs these markets?
“The rational actor model includes no principles entailing the communality ofbeliefs across individuals For this reason, the complex Nash equilibria that arise
in modelling the coordination of behavior in groups do not emerge spontaneouslyfrom the interaction of rational agents Rather, they require a higher-level correlat-ing device, a choreographer” (Gintis2009: 248) Gintis, who is thoroughly au faitwith the problem of the diversified actor’s world, reanimates Walras’ auctioneer,but unlike the engineer Walras he identifies him in the “common priors”: in socialnorms But regardless of what is successful in the theory, who implements it at thepractical level?
This means that the theory can no longer determine or define in its models how
an actor should behave as he knows a repertoire of behavioural possibilities and canimplement them as the circumstances require “Even with a commonality of beliefsand a social norm choreographing a correlated equilibrium, self-regarding individ-uals do not have incentives to play correlated equilibria” (Gintis2009: 249; see alsoBrodbeck 2002) But as the theories which speak of adaptative behaviour (seeGigerenzer and Selten2002) have long been aware, this means that the behaviouralrepertoire to which actors can have recourse in different situations is also applied(This does not mean that all actors make use of it: many behave conventionally in
an extremely routine fashion But they can make use of it This dispositive cannot
be ignored See Brodbeck’s distinction between “habit” and “creativity” Brodbeck
2002)
Economics had identified a series of restriction situations which shape differentmanners of action- all of them as rational action under constraints (e.g moralhazard, economics of contest, hidden knowledge, information asymmetry, credit-ability of threats etc (Hirshleifer and Riley2002; see also – applying game theory –Gintis2009)) In the individual analysis the situation or constellation is alwaysclear, but how about the market process? How many suchconstraints (or in Gintis’modus: how many games) exist at the same time? Which of them do the actorsrecognize and which not? How do they then assess the variousconstraints to whichthey respond with a form of behaviour or not?
The advantage of classical rationality theory was that one could assume thatwhen all actors behave rationally all of the others know that they are behavingrationally Thus a balance of expectations was established which no longer applieswhen (a) all of the actors behave differently – differently and/or with differingdegrees of rationality – and when (b) all of the actors are subject to differentrestrictions We can describe this clearly and distinctly in the analysis of individualcases, but in the market process it is unclear what aspects are perceived, ignored oravoided We have to do with an ambiguity of knowledge/lack of knowledge Alltransactions entered into under these conditions are no longer automatically optimal
4 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 14because the answer to the question as to what the actors recognized or ignoreddepends on knowing what they could have recognized better Akerlof introducedthis topic into economics as information asymmetry (Nachane2006) but a funda-mental problem lies concealed behind it: how can the actors in each case know whatsituation or constellation they are confronted with? In a landscape full of hetero-geneous possibilities and situational assessments the place from which the actoracts is no longer determined (andper definitionem cannot be determined).The analysis of the actual behaviour of the actors, which we owe to empirical orbehavioural economics, is insufficient when it only identifies specific patterns(fairness, reciprocity, altruism etc.) and lists of heuristics, because it remains anopen question under what conditions the agents activated which particular actorbehaviour The fact, for example, that actors act fairly, i.e do not only set theirsights on their own results, but also take those of others into account, is notpersistent behaviour, but is sometimes activated and sometimes not (see Priddat
1996) Under what conditions? Are we dealing here with oscillating cognitiveattentiveness or does a particular form of behaviour first result from the constella-tion within which it takes place, for example, by observation of the behaviour ofothers whom one then simply follows? (Then the actors behave fairly, not out ofconviction (belief system), however, but by copying other behaviour according torules in a social context from which they do not wish to be excluded as sociallydiscredited persons (social positioning)) Economic behaviour can thus no longer bedefined as persistent, as a quasi-natural property of subjects but first crystallizes itstypes according to the social and economic constellation in each case
Herbert Gintis, who analyses four incompatible basic models of human iour inThe Bond of Reason, calls for a “unified model of choice that eliminatesthese incompatibilities and that can be specialized in different ways to meet theheterogeneous needs of the various disciplines” (Gintis2009: 249) His proposalamounts to a unifying theory (“unifying the behavioral sciences”); the “bonds ofreason” are for him “forms of sociality” To this extent Gintis follows the results ofexperimental economic behaviour research, as he expects that “the actors arepredisposed to obey social norms even when it is costly to do” (dito) It is equivalent
behav-to a general theory of social norms as the institutional infrastructure of markets,which provideshared mental models or shared beliefs in order to re-establish thepreviously merely assumed rationality as a system infrastructure – i.e as thecondition for the possibility of rational action in which all know that all arebehaving rationally (Arrow argues along the same lines (Arrow1979; cf Priddat
2000a,b)) What was presumed to lie in the nature of man in the classical rationality theory turns out today to be a condition for the “coherence of beliefs”which has to be culturally or institutionally created But how? How does a societyproduce its culture, norms, institutions? On this topic we only have evolutionaryconcepts, i.e processes of emergence which develop historically but can scarcely
balance-be produced, guided or regulated
We are discussing on the one hand a social phenomenon: that decisions are notsolely decisions on (economic) alternatives, but are often at the same time decisions
on social positioning And we are discussing on the other hand epistemic problems:
Trang 15what do the actors know and what do they not know? How do people act when theybehave rationally and at the same time fail to recognize theconstraint to which theyare subjected? And what if they cannot distinguish between reality and possibili-ties? If we continue to insist on the postulate of individual rationality we overstrainthe cognitive competence of people who are in principle ordinary actors: we burdenthem with epistemic demands with which they cannot on average cope What do theactors know about what they can do (and can know)? (On the identity theme as aneconomic theme see Akerlof and Kranton2010; Herrmann-Pillath2011b).This is a paradoxical situation: i.e the more differentiated the manner in whichthe economy analytically differentiates the individual specific behaviour/constraint
or behaviour/situation relationships, the more complex the epistemic catalogue ofdemands on the actors in principle becomes Even when they are in some wayinformed they cannot estimate which information is relevant and, above all, whichinformation refers to whichconstraints or constellations The definition of eachspecific behavioural type within each of the specific behaviour/constraint concep-tions is then systematically underdetermined in regard to the actual flexibility of thebehavioural possibilities The analytical differentiation of economics, which someeconomists regard as the beginning of a “more realistic economics” distinguishesbetween the model worlds excellently without reflecting on what this means for theactors, who as a result operate within more and more unclear market processes andbecome unsure in view of the increase in situational diversification, the multipli-cation of constraints and the problems of interpretation What might be clear anddistinct for theorists is often only noise – informational and epistemic noise – for theactors, especially as they do not in any case know these theories Who among thereal economic actors has actually been trained in economics? (see Priddat2012a)?The reception of theory by the actors themselves has not been worked out in theeconomic theories “If the theory itself can enter into the studied reality, if an actorcan acquire knowledge of anexplanation of actions (of a theory) he always has thefree choice between basing his actions on this explanation of action or not doing
so Hence the description of his behaviour is undetermined .The theory cannotmodel its own implementation by an actor If a theory explains an action, then thisexplanation is useful; but if a useful theory is implemented by the actor the actionsituation explained by the theory thenchanges Consequently the theory – contrary
to its premises – does not explain the action.” (Brodbeck2002: 356f.)
Let us simply assume in a simplifying fashion, for example, that the many actortypes with their respective rationalities, which economics has in the meantimedeveloped, are all valid at the same time-for example completely (Debreu) orlimitedly rational actors (Simon), actors constrained by meta-preferences (Sen) orsearching for new options (Hayek) etc We cannot simply assume that the actorsfollow a single homogeneous specific rationality, but rather heterogeneous ratio-nalities (see Wolf2006) Of course one always models the rationality the modelneeds to function efficiently But do we think precisely enough? If there are manyactor rationalities – and the literature is always producing new ones (e.g Gilboa
et al.2008; Gintis2009or also Ayache2010) -, how can we assume that the actorsbehave in accordance with the model we have designed? (see Priddat1998)
6 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 16We cannot exclude the possibility that the actors we modelde facto follow othertheories than those we model To put it the other way round: Are we in a position todesign models which contain actors who follow various theories and, among others,not the theory the model itself prefers? What if in a more classical micro-economicsHayek actors are included? Or imagining Shackle actors? Or the simpleboundedrationality actors of Simon? Or meta-preferential Sen actors? And what does itmean when we model actors as rational, but they in practice follow individual,diversely distributed heuristics or “ersatz-economics” as Klamer terms the variouseveryday theories of economic actors (Klamer1987; see Priddat2012a)?
Regardless of whether it takes a strong or a weak form (Kirchga¨ssner 2008),rational behaviour means that the actors know or believe that they can decide on thebest or the better alternative (under the conditions of competition between alterativeoffers) Rationality under constraint means still being able to decide rationallyapart from the fact that the range of choice is restricted in accordance with a specificcriterion (in the case of bounded rationality, for example, by limiting frames orcognitive constraints) Both rationalities are characterized by weighing up judg-ments, even though these are often minimal In contrast, heuristic behaviouralmodels only simulate rationality: one behaves either on the basis of experientialrules or intuitively Here one no longer weighs up what would be the betteralternative but decides decisionistically The actors know nothing about alternatives
or ignore them in order to remain capable of taking decisions This is also true of allroutinized decisions which always regularly copy older decision patterns (cognitiverelief from strain without further weighing up) The latter form of behaviour can nolonger be called rational; but nor is it irrational or random, but designed to enableconnectivity:the fact that a decision is taken at all is more important than decidingefficiently If there are many behavioural possibilities, we are dealing with substi-tution processes which have to be taken into account Economic theories that fail to
do so can no longer guarantee that the optima of the system are achieved, whichwould otherwise be possible
If we now have sufficient reason to doubt that all actors behave rationally, twoanswers to this situation are possible (1) One extends the concept of rationality toinclude all possible behaviour (for example by including fairness in the utilityfunction) But then one has the problem of needing to identify which interpretation
of rationality is valid in each particular case The expectation balance that all actorsbehave rationally can no longer be maintained because one has not yet answered thequestion as to how and in what form the behaviour was rational (2) One positsrationality – in its stronger or weaker form (Kirchga¨ssner 2008) – as onebehavioural possibility among others
In both cases it is no longer clear how far and whether the other actors behaverationally with the consequence of mutual assessability: the shared beliefs arelacking Thus we are in the peculiar situation that we have a list of variouseconomic behaviour models which, however, no longer constitute a uniform econ-omy The models have the advantage of always being able to define which specificforms of behaviour are effective under the specific constellation of conditions Inthis they are valuable from a partial analytical perspective But which models
Trang 17develop constellations in which all (or many) of these different types interact at thesame time? How does one deal with different perspectives within a model? Andhow does one deal with the fact that the different actor perspectives which arepresent do not also perceive the diversity of their perspectives?
Vice versa, empirical research suggests that economic actors operate in marketsaccording to other, personal or weaker heuristics, which not only cannot besynthesized to homogeneous rationality types, but often change in diverse distri-butions and adjust themselves in an adaptative or opportunistic fashion But thismeans that the economic actors – outside their model constitutions – have alreadybeen operating agilely in the economy for a long time (or in a behaviourallyconservative conventional or rule oriented manner), have already completed theseadaptive processes and, on the one hand, observe themselves within them and, onthe other hand, communicate about them If we add that most of the economicactors in the market know neither the economic models nor the type differentiations
in them, i.e do not know at all that or how they should behave rationally (Priddat
2012a), then we must look around in order to see which interactions or ings take place which cannot rest upon individual cognitive processing alone Afterall, in spite of all the heterogeneity, the markets function
understand-What can be unfolded intellectually and intelligently in the economic modelspresents thereal actors with a problem: are they in a position to identify what thetheory/the models configure for them? The game theory models, for example, are inthemselves analytically unambiguous With common knowledge assumptions itseems to be guaranteed that all the players understand the moves of all the others.Thus it is taken for granted that they know what game they are playing This is notthe case inreal economies Do the actors know what game they are in? Or even thatthey are involved in a game at all? What is evident for the theory as a result of itsself-definition is not automatically valid for the actors who make use of it Theyhave an epistemological problem: Where am I? Who am I? What game is beingplayed here?
The actors must interpret the situation themselves What the models determine
by definition is for the actors de facto a hermeneutical problem: they must interpretthe constellation in which they find themselves If it is true, as behaviouraleconomics emphasizes, that the actors bring their own interpretation of the eco-nomic situation with them, we can neither assume naı¨verational choice nor specificconstraints, which of course first have to be identified Here the version proposed byGigerenzer/Selten fits better, namely that in cases of uncertainty the actors fall back
on rules of thumb, on intuitive rules deriving from their experience of situations inthe past, which they seamlessly apply to the present (Gigerenzer and Selten2002;Gigerenzer and Gaissmaier 2011; see also Heiner 1983) Thus they are at leastcapable of taking action, although it is ultimately imprecise, misoriented orsub-efficient
Perhaps the tacit rationality behind all these processes is that one interprets theworld as well as one can, but mostly according to rules, i.e not with a specificrational assessment of the potential of the situation Following the rule meansfailing to achieve the best possible efficiency in the individual opportunistic
8 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 18interpretation of the situation, because it is a matter of observing the rule instead offinding the best alternative Following the rules does not define the best alternatives,but only the fact that all participants can operate in conformity with the rules andthus avoid collisions: effectiveness instead of efficiency If this is the case, we aredealing on average with sub-efficient actions which at least permit us to act ordecide at all, but not in an optimal way The economics of this behaviour would notthen be to achieve efficiency alone, but also connectivity If this is to be regarded asvalid then economics would be designed for effectiveness instead of efficiency.Let us go back to the multiple actors, whose ensemble (variation pool) we havecalled an agent According to this notion the individual possesses a transformationcompetence He must adapt (learning, adaptation) or activate specific actor types(in a kind of role change) in another way in new situations or contexts But thishighly diversified population of actors also gives rise to a problem How can theactors with their various rationalities interact in the same market? There is nounified foundation, no basic structure of reality in which all individuals can operate
in equal measure It was believed that balance models had found such a foundation
in the rationality concept of the individual, as the use of rationality guarantees, atleast by definition, that all can expect that all the others behave rationally (so tospeak as the institutional infrastructure of shared beliefs of shared beliefs in which
by means of the balance of expectations predictions on actions can be undertaken)
If we start from a concept of the individual in which each individual/agentconsists of several actors, then correspondingly different views of the world result,other segments of what is observed, and the observations themselves, even differ.Friedrich A von Hayek, who analysed the markets as arenas of distributed knowl-edge, still assumes a uniform information process If one starts from different actorpopulations this is no longer possible If we include the plurality of actors in ourcalculations, here can be no guarantee – on account of the variety of individualsinvolved – that the market and its signals (prices) will be homogeneouslyinterpreted or that coherent expectations can be developed We are then dealingwith system-endogenous risks
1.2 The Modern “I”: In a Language Game
The neuroscientist Gerald M Edelmann has shown that the brain of the individualperson has developed in the course of evolution in a reciprocal relationship withother brains This networking arises – precisely in regard to categorical classifica-tions and allocation – through the medium of language (see Herrmann-Pillath
2010) Carsten Hermann-Pillath takes up this evolutionary biological conception
in order to depict actors in markets as communicatively networked actorpopulations Reliabilities only developed through shared word meanings, variousnarrative perspectives and other kinds of communication (Herrmann-Pillath2010)
To this extent rationality is founded on communication itself This again is theprecondition for the ability of multiple individuals to act together in the market
Trang 19field.Individual rationality is not decisive but the rationality of the tion) system Consequently rationality is not a fixed parameter (in the way it is used
(communica-in balance theory), but is cont(communica-inuously modified and developed through cation We have already presented this fact in regard to transaction: taking indi-vidual rational decisions does not result in an economic action, as onlytransactionally successful linked rationalities or decisions of at least two actorslead to an economic result The “methodological individualism” which insists onrational individuals as the basic unit of economics is possibly a misleading con-ception, as a closer look reveals that we can only accept transactions exclusively asbasic units or basic events Transactions are structurally “consensual events” inwhich two actors have to reach reciprocal agreement The facts that we areaccustomed to transactions as highly routinized processes must not lead us to forgetthat they are in principlebargaining designed to achieve communicative consensus.This does not mean that personal individual interests disappear in the transactionsbut rather that a consensus must be found beyond the differences in interest.The individuality of the agents is a kind of readiness (precommitment), whichconsists in continuously negotiating one’s own self-image in communication withothers and balancing out the different motivations in the process The US Americanphilosopher Richard Rorty calls the modern “I” a “network of wishes and convic-tions” which is continuously woven anew In this communication the self-image,what is individual about the individual, is socially embedded.Hence what “I” want
communi-is always founded on what “we” want The economcommuni-ist Robert Sudgen and thephilosopher Martin Hollis have analysed this for economics as team preference: fordecision problems the “we” provides a different framework than the “I” The “I” isalways already embedded in the field of language, in which meanings, differentcontexts, symbolic representations and possibilities for interpretation lie (Kabalak
et al.2008) This means that in their decisions individuals are from the outset communicatively influenced (see Frith and Singer2008, but also Gintis2009) andcan thus modify their actor’s role without dropping out of the process individual-istically “Every exchange act is socially and communicatively embedded”(Brodbeck2002: 359; Wohlgemuth2008)
socio-This embedding has, however, at least two dimensions: on the one hand thecultural substructure of all individual actions which is available to the economy as acultural resource base via socialization and cultural institutions and modulates itsbehavioural possibilities In this sense the economy is always aneconomy withinculture (see Priddat2000b), without economic theory being in a position to state theinput it uses as a matter of course for the market design (the ceteris paribusassumption is posited in classical fashion, and is thus outside theoretical reasoning).Institutional economics, or at least that of D.C North, is an incipient attempt toanalyse institutions more precisely as the cultural infrastructure of the economy(Priddat 2004b) “Institution” is then the name for a persistence of socializationpatterns in a coherent population as a “unification of perceptions” and “sharedbeliefs” (Priddat 2003) But it would be too one-sided to bring only this time-bridging cultural inheritance into play The second dimension involves the ongoingcommunications in society, which continuously thematize society and its economy
10 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 20and bring in variants Communication is not a sociological special field whichaffects the economy only marginally (disqualified until recently by game theorists
as “cheap talk” (seen differently in Wernerfelt2008)) In communications ings are generated which provide society, and in like manner the economy, with anorientation for its actions They are the operative part of culture, which they irritate,question and partly reorientate by constantly raising new issues
mean-The fact that communication takes place constantly in the economy is not initself a finding for economics, although it is astounding that all transactions that donot take a routine course demand detailed communication between the participantsand with their networks, which are ultimately decisive for the conclusion of deals(up todue diligence in the case of mergers, but also with larger contracts of allkinds, form investments to the purchase of houses and cars) The point I wish tomake here is closely linked with what has been thematized above: the knowledgewhich actors lack in the manifold assessment of behaviour, situation and context,and which cannot be provided simply by the acquisition of information, as theinformation needs to be weighted, judged and assessed,is generated by communi-cation What then matters is not primarily the maximization of individual benefit,but the discovery of reasons for entering into transactions which one can justify tooneself and others (in the sense of social and epistemic legitimacy (Kettner2011;Kettner/Sauerland2012; also von Aaken2002))
Communications differ from information as they involve repeated furtherqueries until a meaningful and sensible result has been achieved If, as is oftenthe case in economics, units of information are seen as signalling acts involvingmore or less clear contents, it then seems that information automatically means
“better knowledge” In contrast, communication is a process of querying themeaning or relevance of individual pieces of information and assessing them inrelation to other information What is at issue, therefore, is the ascertainment of thereasons for the acceptance of information In the process of communication infor-mation is changed, modified, rejected, analogized, completed or substituted Theresult of information is emergent, the product of a process As opposed to the effects
of information, the influence of communication on decisions cannot, consequently,
be predicted Above all, communications are semantic processes of ascertainmentfor the participants What is valid? What can we accept as relevant? What can weexclude?
Thus we are dealing with processes of knowledge generation which proceedneither individually nor in a controlled fashion (the illusion of advertising and often
of the media), especially as they fulfil another function: they place the focus onmeanings which put an end to the epistemic uncertainty which arises in the actors
on account of the diversity of the situations and behavioural potential (seeSiegenthaler1993: 94ffr., Kabalak2009: Chap 3; Wohlgemuth2008) What actorsare not in a position to assess rationally, their increasing cognitive overload, iscancelled out by social and specifically economic forms of communication Bymeans of communication they constantly find points of orientation,frames and keythemes.Within such a communicatively generated focus of meaning the still diversebehaviours can relate to one another without being attuned to a shared rationality
Trang 21(as Arrow proposes, as aprocess of communication of the constraints on action inthe sense of acoherence of beliefs (Arrow1979; Priddat2000a,b)) Communica-tion takes over what the economic theory of balance can no longer guarantee amongthe highly diversified population of actors – the closure of the heterogeneous field
of actors by means of a temporary consensus among the participants We thus have
a kind of balance of expectations arising from – temporarily valid – sharedinterpretations of the situation (similarly Herrmann-Pillath 2010) This again isfunctionally similar to the shared mental models used by D.C North to explaininstitutions, but it is now directly applied to the markets For if communications cangenerate shared action perspectives by means of semantic focussing, we are thendealing with a new type of institutional regulation (see also von Aaken2002) which
is more dynamic than the institutions with their time-bridging rule stabilities onwhich we usually set our sights
1.3 Embedded Individuals: Networks
Let us clear up one aspect: consumption moderated by networks The socialnetworks in which individuals participate are communicative arenas which regulatethe social position of the economic actors What is communicated in the arenas inthe way of themes, meanings and tendencies creates frameworks (mostly onlytemporarily valid) within which the individuals can relate to one another in spite
of their differences In this way the individuals set up the context within which theycan cooperate (apart from the specific institutional arrangements for communica-tion such astrading rooms (see Beunza and Stark2004))
In the markets – but above all parallel to them in networks and the media – newmeanings/new semantics are constantly thematized (and old ones are also con-firmed) Consequently markets can institutionalize communications Routines andhabits thus arise Or new meanings with new followers of the decisions arise Thismakes the chaotic structure of markets clear At any time ramifications can developwhich lead to other evaluations of goods than had been expected up to that point.Other new attractors focussing on meaning repeatedly take shape In order tounderstand this we must not only analyse the communication processes of society(networks) and markets more thoroughly, but, above all, we must include commu-nication as an important factor in the economy (Kabalak 2009: Chap 3) Herecooperation with sociology and, above all, with sociologists who adopt a network-theoretical approach is worthwhile (White2008a,b; White and Godart2007a,b).Markets are not oriented on individuals who have recourse to criteria of ratio-nality in each of their decisions; most transactions are based on habitual routines.Behaviour tends to follow a collective rationality (in the sense of communicativelyshapedshared mental models or, more precisely, conceptual schemes generated bycommunication (Turner2008; Ehrig2009; Ehrig and Kauffmann2010)) But at thesame time individuals, because they are individuals – generate deviations, inter-pretations of their own But they seldom have the confidence to do so alone (the
12 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 22individual lacks the confidence to be individual), and instead involve themselves innetwork communications which open up new horizons or confirm old ones forthem The knowledge/information which the economy usually assigns to its actors
as a matter-of-course form of access, is not only shaped within the market itself butparallel to it in the networks which communicate on all the life situations of theparticipants and put opinions/meanings into focus This applies to the markets aswell For the decisions taken there have been communicated in advance in networks(family, relations, friends, acquaintances, colleagues, scenes, associations etc.) Thefoci or frames which are formed there (linguistic communities, belief structures(White2008b; White and Godart 2007a, b)) orient the actors for their markets.Nobody will buy anything which is regarded in the networks as degoutant or associally discrediting
Many goods are not simply goods but are equipped with recognition indiceswhich can determine the position of the buyer in the social positioning matrix Inthis case the benefit is not only economically but also socially determined Rationalchoice ignores this dimension Or it is redefined by defining benefit as (a) benefit
1 of the good itself and (b) benefit 2 of the good in the symbolic space of society.However, the relationship between the two benefits can scarcely be mediatedwithout communication
The modern network culture of society with its heterogeneity, its cluster tions, its idiosyncratic linking of histories, milieus, people and organizations cannot
forma-be ignored in economic theory Such communications open up a space in which thesoundness and certitude of rational decisions is no longer valid It is no longer aquestion of taking decisions in rational worlds but of learning the dynamics ofchange – of adapting criteria to new givens, recognizing linking factors andestablishing new connections (see Nassehi 2004) Above all, these should notonly be economicsui generis but also symbolical It is obvious that we can nolonger have recourse in a standard fashion to normed rational behaviour, but mustinstead deal with a flock of more or less adapted individuals who communicate withone another about what they no longer understand or know and in the process ofcommunication arrive at results which replace what had hitherto been assumed to
be rational Rationality, if it can still be characterized as such, becomes a socialform of intelligence sustained by communicative processes
Classical rational choice rests upon a notion of relatively quietly developingmarkets, whereby individual preferences represented the knowledge and experience
of past acts of choice in order to project them into the future The supply tives remained relatively consistent and only the prices varied; in this situation itcan be asserted that the prices contain all the information In today’s hypermoderntimes we are dealing with more intensified supply dynamics in which prices,qualities, innovations and variations oscillate Precisely when one is confrontedwith new offers previous experience and old preferences are of no use: all newalternatives are neutral in terms of experience
alterna-What is new is attractive but untested Hence it is constantly necessary to decidewhether to choose something old, guided by experience, or something new andunknown The apparently safe decision itself becomes unsafe What is new can only
Trang 23be praised – in the sense of an economic rhetoric (economics of persuasion), which,among other things, bears the name advertising The market is a forum for perma-nent attempts to convince (Wohlgemuth 2008: 63), a “forum for persuasions”(Palmer 1991: 304) Rhetoric is the fitting concept because the actors must beconvinced to buy something they previously had not preferred Advertising ispreference-changing intervention It is, therefore, obvious that we can no longermaintain the assumption of economics that preferences are invariant.
Because the new offers on the market cannot be accompanied by experience, aspace is opened up for imaginations and fictions (Beckert2011): one envisageswhat the new alternative could mean Nowadays goods are not simply goods, butgoods + S S is the symbol for sign The sign points to symbolic surplus value, to astory which embeds the purchases of the goods in a world of its own that the buyerenters unisono (Priddat2006; cf Hutter’s distinction between “price value” and
“praise value” (Hutter2011)) The notion which fills the gap in experience can becommunicatively generated and determined The purchase of an Apple computer,for example, is the purchase of a rather modest technology, which is, however, socharged symbolically that one pays the (higher) price for membership of a com-munity of enthusiastic self-willed people But the benefit is then defined by socialparticipation in a group generated by the strategy of the Apple Company, which hasthus created its owncommunity Here goods are not tailored to meet the needs ofclients, but new needs are created: Apple dictates what counts, without marketresearch Instead the company sets the standards This is not market serving butmarket making (or market shaping as my colleague Mark Giesler calls it).What Apple offers is a symbolically charged product Apple renders visible what
it means to obtain surplus value from symbolic space All brands undertake similarattempts with varying degrees of success in binding customers The communication
of the customers with one another is the decisive factor In saying this I only wish tosuggest that the decisions which ultimately lead to transactions are subject toinfluences which theory cannot ignore Advertising is not information but theexercise of influence – it aims to achieve a change in preferences Strictly speaking
it attempts to prevent the actors from taking independent rational decisions Itwishes to impregnate the situation with an external meaning which shapes thedecision in advance: we have called this aspect the economics of persuasion.Possibly the economic system is more discursively constructed than economictheory will allow
1.4 New Dynamics
Communication (through networks) constantly reproduces its own frameworkconditions Markets communicate consumption alternatives (a) by advertising and(b) by the evaluation of the consumers themselves “Dialogicity and the memory orstorage capacity of the mass medium internet .can fundamentally change theeconomy” (Kabalak2009: 378) As a result the markets become less predictable
14 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 24In a world in which what is possible is taken just as seriously as what is real we can
no longer model rationally acting people as if they decide on given alternatives onthe basis of perceptions (the mere fact that they also decide on notions about otherworlds into which possible consumption transposes them shows that we are not justdealing with factual givens) Hypermodern actors are incomplete individuals (seePriddat2005; Nassehi2004) who no longer only wish to consume optimally butalso to change, transform or complete themselves in the process This is beingnewly debated in economics as the problem of identity (Priddat1996; Akerlof andKranton2010; Herrmann-Pillath2011b; but also Nassehi2004)
In the world of networks theblogs and clusters, for example, create their ownevaluation alternatives After all, communication does not merely involve gatheringinformation on already existing alternatives but also the creation of new possibil-ities The public evaluation of consumption is a medium of communicative prefer-ence shaping.Innovation no longer lies in the hands of the enterprises alone; it hasalso reached the consumers Not only the fact that there are so many diversified “I”s
in the economy is remarkable, but also the way in which these “I”s, who rarily share a common interest, come together to form groups, projects andcom-munities – like shoals which stabilize themselves and then fall apart again or take updifferent developmental paths The markets are becoming fast, inconstant and self-irritating We scarcely have any practice in acting within them and our traditionaltheories are founded on social preconditions which are no longer valid Thesociology of economics is notup to date The uncertainty which the actors are innot only in regard to future events (to date economics has discusseduncertaintyonly in this context) has long come to include the ambivalence involved inevaluating the world of supply – an epistemic uncertainty which they overcomenot only with the help of expectational probabilities but also by the formation ofcommunicatively organized communities of practice (Wenger 1998) Just as inmarketing it is not only the response to advertising that is analysed but also theway the response is generated in social peer communication and networks inde-pendently of advertising, so too in the economy it is necessary to ask about thosecommunicative arenas the economic actors operate in and the solutions theypractice which have hitherto not been foreseen in economic theory These commu-nications are the new generators of knowledge which, in the increasing diversity ofbehavioural possibilities and situations, give the decision-takers the certainty theyneed to take a decision Here the aspect of social and epistemic legitimacy must beweighed up just as much as the question of optimality Rational choice, i.e theindependent individual cognitive assessment of the best options, then becomes onepossible line of action among others, without enjoying the prominence that eco-nomics, on average, still attributes to it Nobody can claim to know which of thenew alternatives is the best – neither among themselves nor in relation to personalexperience The criterion for choice is their attractiveness (their story, their imag-inative value, an affective inclusion) and their communicative legitimation Com-municative legitimation clarifies the fact that one is not standing alone in taking adecision, when others also decide to buy something Communication generatesshared beliefs, but only as a temporary consensus
Trang 25tempo-From a functional standpoint communication takes on the role of the systemicclose-down of the equilibria in the market But we can longer assert that thesecommunicatively closed equilibria are optimal, for the meanings that arise asshared mental models and orientate the actors in their uncertainty can no longer
be fully determined economically We are then dealing with multiple equilibria(Gintis’ correlated equilibria), which are effective rather than efficient: theirfunction is to generate opportunities to make connections
If we can assume that there are many behavioural dispositions in the economywhich are not all, and not even in the majority of cases, rationally based, economictheory must then start from the assumption that connections are continuouslyproduced but that they are not optimal Their “pedagogical programme”, if I mayput it this way, aims atrational education: the theory shows in ideal analyticalpictures what could be achieved in the optimal case It shows the analytical roomfor possibilities, but it must at the same time take into account the practical potential
of the actors – and their actual practice when acting, which is much more stronglyaffected by communication than economic theory has hitherto been ready toappreciate We must in fact be content with demonstrating the functioning of themarket in which the actors have difficulty in making decisions For example, theerrors which bank customers make in communicating with their bank advisors – totake up one dimension (Priddat2012c) – are based on a “ludo-narrative dissonance”(Hocking 2007) Ludo-narrative dissonance is the effect occurring in computergames when the game and the narrative fall apart Transferred to the economy itmeans that the game of the system which the economists describe is disconnectedfrom the narrative which is the basis for the decisions taken by the actors in realeconomic processes For the bank customers this means that the bank advisors sellthem securities although the risks involved are concealed from them, and, lackingthe necessary knowledge, the bank customers must trust the advisors withoutactually being able to trust them The one side sells what the bank requires it tosell; the other side hopes to be given advice (i.e to be advised not to buy suchsecurities) This marginal example reveals that communication must occur throughthird parties, who provide assessments that the other two, on account of theirpositions in the system, cannot themselves generate
Inpractical terms the economy has long developed into a triangular system inwhich the bilateral actor/actor relationships in the transaction have beensupplemented by a communication which functions as athird party enforcer (seeBarzel2001) Barzel develops aneconomic theory of the state –in the relationshipcitizen/citizen/state Here we are only interested in the transfer to the relationshipactor/actor/communication (Priddat2012b) In this triadic structure the heteroge-neity of the actors and the diversity in their behaviour is no longer a problem ofrelational rationalities because the relational differentials are communicativelyadjusted What the economy can no longer provide as the idea of a closed system
of balances is clarified by the accompanying communications in each case Theknowledge that the actors no longer have is made accessible to them communica-tively in each new case In their communities of practice the actors clarify what theyimagine they can decide That they bring in various manifoldly different heuristics
16 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 26ceases to be a problem, when they agree communicatively on the goals for action towhich they attribute significance To this end one does not need to be a homeoeconomicus Homo sapiens sapiens is fully sufficient After all, the orderingfunction does not rest upon the individual and his rational varnish, but on theactor/communication/system nexus It is easy to see what research still has toachieve.
References
Akerlof, G.A., and R Kranton 2010 Identity economics: How our identities shape our work, wages, and well-being Princeton: Princeton University Press.
Aoki, M 2001 Towards a comparative institutional analysis Cambridge: MIT Press.
Arrow, K.J 1979 Rational discourse and conflicts in value & judgement In Recent approaches to social sciences, ed Hans K Betz, 3–32 Calgary: The University of Calgary.
Ayache, E 2010 The blank swan: The end of probability Chichester: Wiley.
Barzel, Y 2001 A theory of the state: Economic rights, legal rights, and the scope of the state Cambridge: Cambridge University Press.
Beckert, J 2011 The transcending power of goods: Imaginative value in the economy In The worth of goods Valuation & pricing in the economy, ed J Beckert and P Aspers, 106–130 Oxford: Oxford University Press.
Beunza, D., and D Stark 2004 Tools of the trade: The socio-technology of arbitrage in a Wall Street trading room Industrial and Corporate Change 13(2): 369–400.
Binmore, K 2011 Natural justice New York: Oxford University Press.
Brodbeck, K.-H 2002 Wirtschaft als kreativer Prozeß In O¨konomik und Sozialwissenschaft,
ed W O ¨ etsch and St Panther, 353–387 Marburg: Metropolis.
Coates, J 2012 The hour between dog and wolf London: Harper Collins Publishers.
Ehrig, T 2009 Reasoning about novelty in markets: Concepts and models Dissertation at the faculty of Economics and Management at University Witten/Herdecke.
Ehrig, T., and L Kauffmann 2010 The formation of expectations for novel opportunity, working paper Leipzig: Max Planck Institute for Mathematics in the Sciences, University of Illinois at Chicago, 2 November 2010.
Elster, J 1987 The multiple self Cambridge: Cambridge University Press.
Falk, A 2003 Homo Oeconomicus versus Homo Reciprocans: Ansa¨tze fu¨r ein neues politisches Leitbild? Perspektiven der Wirtschaftspolitik 4(1): 141–172.
wirtschaft-Fehr, E., and U Fischbacher 2004 Social norms and human cooperation Trends in Cognitive Sciences 8(4): 185–190.
Frith, Ch.D., and T Singer 2008 The role of social cognition in decision making Philosophical Transactions of the Royal Society 363(1511): 3875–3886.
Fruehwald, E.S 2009 Reciprocal altruisms as the basis of contract, Research Paper No 08/09, University of Louisville Law Review, Hofstra School of Law.
Gigerenzer, G., and W Gaissmaier 2011 Heuristic decision making Annual Review of ogy 62: 451–482.
Psychol-Gigerenzer, G., and R Selten 2002 Bounded rationality: The adaptive toolbox Cambridge, MA: MIT Press.
Gilboa, J., A.W Postlewait, and D Schmeidler 2008 Probability and uncertainty in economic modelling Journal of Economic Perspectives 22(3): 173–188.
Gintis, H 2009 The bounds of reason Game theory and the unification of the behaviorial sciences Princeton: Princeton University Press.
Trang 27Gourge´, K 2001 O¨konomie und Psychoanalyse: Perspektiven einer Psychoanalytischen O¨konomie Frankfurt am Main: Campus Verlag.
Heiner, R.A 1983 The origin of predictable behaviour American Economic Review 73(4): 560–595 Herrmann-Pillath, C 2010 A neurolinguistic approach to performativity in economics Journal of Economic Methodology, 17(3): 241–260, penultimate version: Frankfurt School working paper series no 123 SSRN: http://ssrn.com/abstract ¼1414167
Herrmann-Pillath, C 2011a Institutions, distributed cognition and agency: Rule-following as performative action Journal of Economic Methodology, penultimate version: Frankfurt School Working paper series no 157, SSRN: http://ssrn.com/abstract ¼1745437
Herrmann-Pillath, C 2011b Economics of identity and creativity St Lucia: University of Queensland Press.
Herrmann-Pillath, C 2012 Towards an externalist neuroeconomics: Dual selves, signs, and choice Journal of Neuroscience, Psychology, and Economics 5(1): 38–61.
Hirshleifer, J., and J.G Riley 2002 The analytics of uncertainty and information Cambridge: Cambridge University Press.
Hocking, C 2007 Ludonarrative Dissonance in Bioshock http://clicknothing.typepad.com/click_ nothing/2007/10/ludonarrative-d.html , Blogentry Accessed 7 Oct 2007.
Hutter, M 2011 Infinte surprises: On the stabilization of value in the creative industries In The worth of goods Valuation & pricing in the economy, ed J Beckert and P Aspers, 201–222 Oxford: Oxford University Press.
Kabalak, A 2009 Institutionelle Spiele Marburg: Metropolis.
Kabalak, A., and B.P Priddat 2010 Emotion, Gefu¨hl und Entscheidung Elemente einer Affektenlogik der O ¨ konomie In WENKE2 – Wege zum nachhaltigen Konsum, ed Irene Antoni-Komar et al., 132–159 Marburg: Metropolis.
Kabalak, A., B.P Priddat, and E Smirnova (eds.) 2008 O¨konomie, Sprache, Kommunikation Neuere Einsichten zur O¨konomie Marburg: Metropolis.
Karpik, L 2011 Mehr Wert: Die O¨konomie des Einzigartigen Frankfurt am Main: Campus Verlag.
Kettner, M 2011 Gute Gr €unde f€ur und in Konzeptionen o¨konomischer Rationalit€at, Beitrag fu¨r die Proceedings des Kongresses der Deutschen Gesellschaft fu¨r Philosophie, Mu¨nchen 2011 Manuskript.
Kettner, M., and D Sauerland 2012 Weder homo oeconomicus noch animal spirits – Gr €unde statt
Pr €aferenzen Vortrag auf der Tagung ‘Menschenbild und O¨konomie’ vom 28.–30.6.2012 an der Universita¨t Witten/Herdecke, unvero¨ffentlichtes Manuskript.
Kirchga¨ssner, G 2008 Homo oeconomicus: Das o¨konomische Modell individuellen Verhaltens und seine Anwendung in den Wirtschafts- und Sozialwissenschaften Tu¨bingen: Mohr Siebeck Klamer, A 1987 As if economists and their subjects were rational In The rhetoric of the human sciences, ed J Nelson, A Megill, and D.M McCloskey, 163–192 Madison: University of Wisconsin Press.
Nachane, D.M 2006 Economics of asymetric information New Delhi: Deep & Deep Publications.
Nassehi, A 2004 ICH-Identita¨t paradox In Das individualisierte Ich in der modernen schaft, ed G Nollmann and H Strasser, 29–44 Frankfurt am Main: Campus.
Gesell-North, D.C 1992 Institutionen, institutioneller Wandel und Wirtschaftsleistung Tu¨bingen: Mohr Siebeck.
North, D.C 2010 Understanding the process of economic change Princeton: Princeton sity Press.
Univer-Palmer, T 1991 The hermeneutical view of freedom In Economics and hermeneutics,
ed D Lavoie, 299–318 London: Routledge.
Priddat, B.P 1996 Moralischer Konsum U ¨ ber das Verha¨ltnis von Rationalita¨t, Pra¨ferenzen und Personen In O¨konomie und Moral Beitr €age zur Theorie o¨konomischer Rationalit€at, ed K.-
R Lohmann and B.P Priddat Mu¨nchen: Oldenbourg/Scienta Nova.
Priddat, B.P 1998 Nichto¨konomische o¨konomische Theorie Vivienne Forresters Buch ‘Der Terror der O ¨ konomie’ als anregende Lektu¨re zum Problem der Kontextspezifita¨t von rational choices Homo Oeconomicus 1998, Nr XV (2): 141–169.
18 1 The New Population of Economics: Multiple, Fair, Ignorant and Emotional
Trang 28Priddat, B.P 2000a “Communication of the constraints on action” K.J Arrow u¨ber Kommunikation Zeitschrift f €ur Nationalo¨konomie und Statistik, Bd 220, H 4, Juli Priddat, B.P 2000b Beyond equilibrium Kultur als Hintergrund/Vordergrund der O ¨ konomie In Kapitalismus, Krisen, Kultur, ed B.P Priddat Marburg: Metropolis.
Priddat, B.P 2003 Kultur und O ¨ konomie In Ph €anomen Kultur Perspektiven und Aufgaben der Kulturwissenschaften, ed K.E Mu¨ller Bielefeld: Transcript.
Priddat, B.P 2004a Strukturierter Individualismus Institutionen als o¨konomische Theorie burg: Metropolis.
Mar-Priddat, B.P 2004b Historische Methode und moderne O ¨ konomie U¨ber das Methodische in der Historischen Schule und das Historische in der Neuen Institutioneno¨konomik In Wirtschafts- geschichte als Kulturgeschichte, Dimensionen eines Perspektivwechsels, ed H Berghoff and
J Vogel, 99–118 Frankfurt am Main: Campus.
Priddat, B.P 2005 Unvollst €andige Akteure Komplexer werdende O¨konomie Wiesbaden: VS-Verlag.
Priddat, B.P 2006 Moral als Kontext von Gu¨tern In Ethik des Konsums, ed P Koslowski and B.P Priddat, 2–22 Mu¨nchen: Fink.
Priddat, B.P 2007 Neuroo¨konomie Neue Theorien zu Konsum, Marketing und emotionalem Verhalten in der O¨konomie Marburg: Metropolis.
Priddat, B.P 2012a Polyphone Ma¨rkte: Ma¨rkte als gemischte Sprachspiele In Die Sprache O¨ und andere, O¨konomie – Narration – Kontingenz, ed G Meins Paderborn: Fink.
Priddat, B.P 2012b Akteure, Vertr €age, Netzwerke Der kooperative Modus der O¨konomie Marburg: Metropolis.
Priddat, B.P 2012c Vertrauen, Wissen, Nichtwissen – Bank-Kunden-Beziehungen in der Finanzkrise In Entfesselte Finanzm €arkte Soziologische Analysen des modernen Kapitalismus,
ed K Kraemer and S Nessel, 263–279 Frankfurt am Main: Campus.
Shafy, S 2012 Crash der Alpha-Ma¨nnchen Der Spiegel Nr 27/2012, S 111
Siegenthaler, H 1993 Regelvertrauen, Prosperit €at und Krisen Tu¨bingen: Mohr (Siebeck) Turner, M 2008 Frame blending In Frames, corpora, and knowledge representation, ed Rema Rossini Favretti, 13–32 Bologna: Bononia University Press.
von Aaken, A 2002 Deliberative Institutioneno¨konomik oder: argumentiert der homo oeconomicus? In O¨konomik und Sozialwissenschaft, ed W O ¨ etsch and St Panther, 201–
White, H.C 2008b Meaning emerges in relational dynamics Manuskript fu¨r den Workshop
‘Relational Sociology – Transatlantic Impulses’, Berlin 25.89 2008 (unpublished).
White, H.C., and F.C Godart 2007a Ma¨rkte als soziale Formationen In M €arkte als soziale Strukturen, ed J Beckert, R Diaz-Bone, and H Ganßmann, 197–215 Frankfurt am Main: Campus.
White, H.C., and F.C Godart 2007b Stories from identity and control Sociologica N 3/2007.
Trang 29Chapter 2
Mutual and Self-Enforcing Agreements.
Contracts as the Basic Institution
of Economics: Network Knowledge Instead
of Rational Choice
O E Williamson sees institution economics as moving “from choice to contract”.This is more than a supplement to previous economics; in its focus on contracts itrepresents a break with rational choice economics: “economics (moves) in thedirection of being ascience of contract, as against a science of choice” (Williamson
2002: 172)
In his justification for this step O E Williamson refers to J M Buchanan:
“thinking contractually in the public ordering domain leads into a focus on the rule
of the game Constitutional economics issues are posed (Buchanan and Tullock
1962; Brennan and Buchanan2000)” (Williamson2002: 172) Williamson takesBuchanan’sconstitutional economics as a basic pattern for a more general case,which, in complementary fashion, he callsprivate ordering What this means is thatthe individualrational choices assume the existence of a market order which they
do not themselves determine in the neo-classical case Whereas Buchanan – in thecontext of hispublic choice theory – wishes to leave the determination of the rules
of the political game of distribution to the political citizen as far as the publicdomain is concerned, this is transferred by Williamson to the sphere of the market:
“Whatever the rule of the game, the lens of contract is also usefully brought tobear on the play of the game This latter is what I refer to asprivate ordering, whichentails efforts by the immediate parties to a transaction to align incentives and tocraft governance structures that are better attuned to their exchange needs Theobject of such self-help efforts is to realize better the ‘mutuality of advantage fromvoluntary exchange (that is) the most fundamental of all understandings ineconomics’ (Buchanan2001: 29), due allowance being made for the mitigation ofcontractual hazard Strategic issues – to which the literatures on mechanism design,agency theory and transactions cost economics/incomplete contracting all have abearing – that had been ignored by neoclassical economists from 1870 to 1970, nowmake their appearance” (Williamson2002: 172)
In regard to the “mutuality of advantage from voluntary exchange” Williamsondifferentiates further between the more formalincentive alignment (mechanismsdesign, agency theory, the formal property rights theory) and the “governance ofongoing contractual relations (contractual implementation)” (Williamson 2002:
B.P Priddat, Communication and Economic Theory, Ethical Economy 47,
DOI 10.1007/978-3-319-06901-2_2, © Springer International Publishing Switzerland 2014 21
Trang 30173) Richter and Furubotn (2005: 161) extend this distinction to one between theagency-contract theory on the one hand and the self-enforcing agreements theoryand therelational-contract theory on the other In the two last cases the incentives
do not play as dominant a role as in the first case; but when neitherrational choicenorincentives can model the transaction, what is a contract or, to be more precise,what determines the contractual form of the transaction? And what is the signifi-cance of this for economics?
The transaction becomes the “basic unit of analysis” (Williamson 1985: 41).Thus a fundamental difference to neo-classical market theory is introduced almostincidentally What counts is no longer the efficiency of anonymousexchanges Therational choice of an individual actor is no longer decisive economically but thecooperation of at least two actors in a contractual arrangement The rationalchoice merely offers dispositions for transactional contracts; without the contractthe acts of choice of the two rational actors presented are economically irrelevant.Market only takes place when the goods are delivered and paid for; its basic unit isthe bilateral transaction
The neo-classicalexchange knows the contractual form only as a legal sheathwithout significance for the rational key decisions Law comes into play as a legalinstitution as a third party called upon in cases of conflict (on the one hand as alawbased institution (Greif2006: 222) on the other as athird party enforcer (Barzel
2002; on this point cf Brousseau2008and Priddat2010c)) In theexchange therational actors choose independently For the exchange the contract is only atechnical finale without significance for the actual process of exchange
In the exchange the final contractual form of the transaction is not a kind ofcooperation but rather the temporary coordination of two utility functions Basi-cally they do not come together; their utility curves merely touch one another hereand there; theirmutuality of advantage is configured in an extremely minimalistfashion To put it more precisely: in the technical finale of the exchange the tworational actors come together under the condition that they have alreadydecided previously, without negotiating or bargaining what they want toexchange with one another The exchange is based on reciprocal acceptance ofprevious individual decisions, which are only carried out in the transaction but notnegotiated (zero mutuality) The exchange is a quasi contractless transaction; theessential feature of a contract, its negotiational core, is neutralized
The innovation from institution economics is ultimately only a reminder of thepreconditions on which neo-classicalexchanges are based Williamson et al see thecontract as the socio-economic place which also regulates the preconditionsguaranteeing its validity In the Middle Ages and in the modern era it was generallyobvious that contracts require an instance guaranteeing their validity: “Thus thebargaining (‘exchange as a species of contract’) has three constituents which soonbecame distinguishable; the making of the agreement, the delivery one way, andthe delivery the other As soon as the distinction is made, the agreement itselfbecomes no more than a promise to deliver Trading is trading in promises; but it
is futile to trade in promises unless there is some reasonable assurance that thepromises will be kept But even in dealings between merchants there can be
Trang 31misunderstandings and there may be deceptions; and there will be the contingenciesfor which no provision has been made Disputes will therefore arise, and there must
be a means of settling them, in order that contracts should be reliable Legal (or atleast quasi-legal) institutions are therefore required” (Hicks 1969: 34f.) AvnerGreif shows the complex history of the “impersonal exchange” (Greif 2006).Yorim Barzel speaks generally ofthird party enforcement as an agency of stabili-zation (Barzel2002) Contracts are accordingly triadic structures in which the legalsystem takes on the role of providing an institutional guarantee in the case ofnon-agreement, i.e in the case of a failure of bargaining
2 the contract guaranteed by the legal system which regulates reciprocal conflicts
as an external third party; itsmutuality is configured triadically, as legal ation (1> mutuality > 0), and
medi-3 the mutuality which places a contract in a focus of things in common (emergentdyadic or bilateralmutuality: mutuality¼ 1); mutuality or mutual agreement areparticularly valid for relational contracts that leave important commitmentclauses open something that can only succeed because the parties enter intoexchange relationships (Richter and Furubotn 2003: 176; with reference toMacneil 1978; Scott 2003) “The stronger the relationality determines thecharacter of the contract, the weaker or more imprecise its legally binding effect
is and the stronger the role of convention or internal enforcement instruments.The element of self-enforcement becomes increasingly more important”(Richter and Furubotu2003: 169; see also Brousseau2008)
After a long phase of negation of the contractual nature of transactions in theeconomy, Commons, in 1924, is the first to speak again of the contractual con-struction of the economic relationships of transactions (in the context of a theory ofthe relations betweeneconomics & law (Commons1995[1924]; also Coase1937)).The transaction concept is explicitly not an exchange concept; it refers to “buying”and “selling”, both mediated monetarily Commons comes from the older Americantradition of institution economics, which, through Vebeln etc., draws on the Ger-man historical school The more recent transaction cost theory which in turn builds
on institution economics (Coase 1937; Williamson 1985; North 1990a, b) seeseconomic contracts as institutional forms whose costs differ (‘new institutionalism’(see Klaes2002); Brousseau2008; Garrouste and Saussier2008) According to thistheory the form of the contract plays a decisive part in the optimization
Trang 32“Optimization” is extended here: How does one avoid too high costs of tional agreements in a sustainable way?
transac-What soon establishes itself as transaction cost theory is in its formation marily conceived as an unfolding of the transaction as a process of contractualnegotiation:
pri-In order to carry out a market transaction it is necessary
• to discover who it is that one wishes to deal with,
• to inform people that one wishes to deal
• and on what terms,
• to conduct negotiations leading up to a bargain,
• to draw up the contract,
• to undertake the inspections needed to make sure that the terms of the contractare being observed
• and so on (Coase1960: 15)
Coase interprets the contractual process as different stages of communicationand negotiation, which are necessary for the realization of a contract For himtransaction costs are “costs of using the price mechanism” (Klaes2002: 3), i.e “thecost of discovering what the relevant prices are” (Coase1937: 390) or “the cost ofnegotiating and concluding a separate contract for each exchange transaction”(Coase1937: 391f.) The “costs of using the price mechanism” reveal a significantdifferentiation between the competitive prices and the price actually paid in thecontract (difference between marginal costs and prices) If contracts, the contractualform of the transaction, include the costs of negotiation in addition to the specificagreements laid down in the contract, the prices actually realized in the marketthrough the contracts are singular, as they not only reproduce the “the cost ofnegotiating and concluding a separate contract for each exchange transaction” butvice versa permit every contract to fix its own price Instead of the neo-classicalprice taking transactional contracts are economic arenas of price making
It is thus understandable that market theory is divided up through the newinstitution economics into:
1 a neo-classical sphere in which the buyer and the seller meet in spot markets “toexchange standardized goods at equilibrium prices” (Williamson 2002: 176),and
2 an institutional sphere in which negotiations generate their own prices, whichcan no longer be weighted as competitive prices because they are unique,i.e incomparable: every negotiation generates other prices (price¼ competitiveprice + transaction costs) We can speak of local equilibria (which are as multiple
as the number of negotiated contracts, similar to the multiple Nash equilibria(see Brousseau2008))
The market relevant prices can no longer be determined independently of theprocess of arriving at them, which only means that the kind of communicationinvolved in the transaction process co-determines a selection of the transactionobjects and ultimately which goods are open to transaction at what price In the
Trang 33transactional negotiations it is not only a question of finding the final price (thetransaction costs over and beyond the market or offering price, which is contractu-ally determined according to specific performances), but at the same time ofdefining the transaction object For the transaction costs are generated by thescope and the specifications of the performance and the guarantees (both securitiesand assurances) It only becomes clear in the course of the negotiations how thetransaction object is constituted or should be constituted, or possibly even what it is,etc In negotiations the object is also always negotiated which makes up the contract– with all the necessary open options, as a mutual outcome (with a commonlydefined object) must result.
It is only in the course of the negotiations that the room is opened up forpossibilities which partners can only concede to one another when they have thereciprocal trust needed to treat each other honestly –mutuality The market onlyprovides the dispositions which are first configured in the bargaining of thecontractual process This is the particular performance of the contracts: that they
do not exchange what the market offers but create their object in the process itself.Contractual transactions are emergent processes
Coase considers the various communicative activities from the standpoint of theminimization of their costs, i.e he wishes them to be organized efficiently Ulti-mately this observation serves his configuration of the theory of the firm (Coase
1937) This has in the meantime become the dominant reading
But this is only one of the possible interpretations of the process, which ignoresthe communicative selection of the final transaction condition, the epistemic oper-ation so to speak The communication of the various stages of searching,assessing and negotiating the contractual process is here seen onlyone-sidedly as a foregoing production process for the creation of the contract,whose performance lies in the efficiency of the process of creation, withouttaking into account that the other performance is the choice of what is signif-icant It is not only a matter of the guarantee or assurance of the fulfilment of thecontract which is to be included in the transaction process, but rather a question ofthe constitution of the transaction itself: its objects and dimensions, including theregulation of the procedure culminating in a trustworthy settlement
Let us reformulate Coase’s elements of the transaction process listed above:
• the search for and identification of potential transaction partners (informing; as areciprocal process: communication)
• signalling the personal desire for a transaction (communication),
• clarification of the conditions of the transaction (communication),
• bringing the negotiations to a close (communication)
• paying attention to the observation of the terms of the agreement on acceptance(interpretation of the agreed contractual process),
It becomes clear that four of the five processes listed are communications,particularly when one is aware that the first two processes are closely intertwined.The transaction contained in the contract is the result of foregoing communicationswhose course determines what is transacted In its original form the transaction
Trang 34cost theory includes a transaction process theory which conceals an undevelopedtheory of transactional communication.
Analytically two levels must be distinguished: the transactions and the tion costs Coase differentiates the transaction as a process in different functionalstates, which mark differing communicative and above all negotiative stages(Wernerfelt 2008) Negotiations are interactions: a mode of social relationshipwhich differs significantly from the neo-classical coordination of utility functionswithout interaction The transaction costs come into play with the interpretation ofspecific transaction relations as “bilateral dependency conditions” (Williamson
transac-2002: 176) Williamson distinguishes between:
1 generic transactions where ‘faceless buyers and sellers meet for an instant to exchange standardized goods at equilibrium prices’” (Ben-Porath 1980 : 4) and
2 exchanges where the identities of the parties matter, in that continuity of relation has significant cost consequences (Williamson 2002 : 176).
Williamson thus emphasizes that the partners in the contractual process get toknow one another – in contrast to the anonymity of the exchange – and that throughthis acquaintance a familiarity or relational specificity arises which generatestransactional continuities The parties continue to transact with one another in thefuture because they have invested in the relationship: “the key factor here iswhether the transaction in question is supported by investments in transaction-specific assets” (Williamson2002: 716)
The transaction costs arising from the emergence of the specific performancesand guarantees in the contractual negotiations are investment costs The invest-ments are investments in transaction-specific assets, i.e in arrangements forguarantees and specific advantages which can only come about through the nego-tiations themselves “Parties to transactions that are bilaterally dependent are
‘vulnerable’, in that buyers cannot easily turn to alternative sources of supply,while suppliers can redeploy the specialized assets to their next best use or user only
at loss of productive value As a result, value-preserving governance structures – toinfuse order, thereby to mitigate conflict and to realize mutual gain – are sought”(Williamson2002: 176)
In the contractual agreement the exchange is only secure when, uno actu, itestablishes a governance, which must be regarded as a kind of constitution of thetransaction (see above the methodical analogy to J B Buchanan’sconstitutionaleconomics) Williamson speaks of the “contract as framework” (Williamson2002:177; with reference to Llewelly1931: 36f.) Theconstitution of the contract is notadequately safeguarded by the legal form of the contract (The legal form onlysafeguards the compensation rights in the case of non-fulfilment, as a kind oflaw byexception), but first by economic stipulations, supervision and trust-building mea-sures (So also Van Aaaken 2007; Thu¨sing 2007; Hermalin 2008; Gomez 2008;Nicita and Pagano2008; Priddat2010c)
Avner Greif has traced the difficult process of the crystallization of the sonal exchange” in the Middle Ages, “the transition from reputation-based personalexchange to law-based institutions” (Greif2006: 222) Williamson is working on an
Trang 35“imper-approach transcending the law-based institution, which he calls “mutuality ofadvantage from voluntary exchange” (Williamson 2002: 172) – an economicconception of the emerging formation of the contract which is to be dependentneither on reputation mechanisms nor on legal safeguards, but on the trust whichdevelops during the negotiation (mutuality¼ 1, essential for the implicit and rela-tional contracts) It is in its mutuality that the contract first proves to be a contract,
by passing beyond a coordinative coincidence to a cooperative relationship That it
is put into effect “in the shadow of the law” (Gomez2008: 101) remains evident,but does not make up its economic content
The cooperative relationship of the contract involves more than trust and ness, as, after trust and fairness have been achieved in the contractual process, athird internal phase succeeds, in which results resulting from negotiation can beachieved that cannot be reached by the mere coordination of customary contractualprocesses Cooperation as “suitable interplay results in a greater complexity thanthat of each individual” (An der Heiden 2010: 9) Greater complexity bringsalternatives into play which cannot even be made visible by neutral coordination.There is a willingness to commit oneself to opportunities and options in a sphere ofmutual trust which remains undeveloped in the latent sphere of mistrust Thegreater complexity which can be achieved gives us an innovative potential inevery contractual process, which remains latent as long as it cannot be coopera-tively heightened or developed
fair-Themutuality of advantage from voluntary exchange requires different nances: a different one for the firm (as a nexus of contracts) than for contractswithin the market (which achievegovernance through institutions of trust, by againdrawing on forms of “reputation-based personal exchange” D M Kahan speaks of
gover-astrong reciprocity (Kahan2005; but see also Gintis et al.2005))
Williamson refers explicitly to the “classical contracting, according to whichdisputes are costlessly settled through courts by the award of money damages”(Williamson 2002: 177) The distinction made here between costless contracts(neo-classical economics or classical contracting theory) and costly contracts(transaction cost theory) refers to “disputes”, i.e to the communicative dimension
of the negotiations in contracts designed to clarify disputed points The observation
of the communication between the contracting partners shows that the legal antee is far too abstract for the specificity which the partners can negotiate about.What “under current rules (could) be brought to a court” could “be resolved instead
guar-by avoidance, self-help and the like That is because in ‘many instances theparticipants can devise more satisfactory solutions to their disputes than can pro-fessionals constrained to apply general rules on the basis of limited knowledge ofthe dispute’”(Williamson2002: 177)
It is interesting that the reconstruction of transaction cost theory assesses cally Common’s theme of linking economy and law more closely in the contract.Because of its lack of knowledge of the specificity of contracts the law cannot settledisputes about validity and interpretation in the way the partners can with oneanother when they define, observe and regulate the governance of the contractualprocess themselves (see Hermalin2008; Gomez2008; Katz2008) The fact that
Trang 36Williamson develops the governance structure as a form of management in firmsdoes not deprive it of its general validity for every contract, whether as a “nexus ofcontracts” (Aoki et al.1990) of a company or simply as a bilateral agreement in themarkets.
The transaction argument against the high legal costs of the contractual form isclearly used: every emerging solution lowers the transaction costs, but also requiresinsight, reciprocity, trust and obligation (precisely what the incentive-orienteddepartments of institution economics did not include For a critical approach tothis point (see Williamson 2002: 188ff.)) Institution economics is clearlyendeavouring to arrive at an economic theory of the contract in order to overcomethe classical dominance of the juridical theory
Through its mode of the negotiating process the transaction becomes the nomic site of the production of contractual stability and security Or, in other words:what was previously regarded simply as anexchange, i.e as technical allocationalcoordination in markets, becomes a process in the new institution economics, whichco-establishes the conditions for its validity by means of contractual cooperation:
eco-“simple market exchange thus gives way to credible contracting, which includespenalties for premature termination, mechanisms for information disclosure andverification, specialized dispute settlement procedures and so on” (Williamson
“nexus of contracts” (Aoki et al.1990), the form of governance becomes decisivefor the capacity to enter into contracts in the market and for the amount of thetransaction costs ofcontracting
Trang 372.2 Contractual Processes
Williamson’s transaction/contract theory clearly refers to contracts between panies But the critical unfolding of the processes which make up a contract isgenerally valid: for every transaction in all markets What seems to be evident in thecase of contracts between firms remains concealed for all other contracts(Williamson assigns them, without further interpretation, to the neo-classicalexchange sphere He does not analyse their specific governance of the mutualagreement) (Fig.2.1)
com-The neo-classical sphere comprisesprima facie the processes ex ante 1 – 4 when
we enter it into the elaborated transaction/contract model Ex ante 1 – 4 canfunction without a contract; basically1 – 3 are processual implementations of theprocess, which Walras calls tatonnement for the formulation of his equilibriumeconomics – the processes of searching and comparing which rational actorsundertake before making their rational choice For the description of the marketprocess in terms of equilibrium economics only ex ante 4 matters, but with adecisive difference: here it is not a question ofnegotiation, but of the technicalimplementation previously taken individual rational decisions of the independentactors A process of cooperation, such as contracts are, is neither envisaged nornecessary
In this sense the processex ante 4 in the above model for the dimension of theneo-classical exchange is to be replaced by 4*: for negotiation we must inneo-classical terms substitute exchange (quasi-negotiation) The condition thatexchanges must be undertaken to competitive or equilibrium prices cannot besustained in the contractual process, as it is precisely concerned with the negotiation
of the price Consequently, according to this logic, contracts can be excluded asforms of negotiation This also means that expost 1-4 represent a post-processing ofneo-classical transactions, which no longer belong to the sphere ofrational choice
Fig 2.1 von Sayn-Wittgenstein 2009 : 10, with reference to Coase and Williamson
Trang 38coincidence, but arise when exchange conditions are not observed This calls for theintervention of a third instance: the legal system.
Contracts, however, “(are) dealing with issues of conflict, mutuality and order”(Williamson 2002: 191; A W Katz speaks of “coordination and commitment”(Katz2008)) Only when the three conditions are fulfilled are contracts complete inthe sense of Williamson’s theory On this point we must take a closer look at thecontractual process as a process Contracts are an instance which reflects the fairfulfilment of the transactions Contractual processes indicate that the result isnegotiated: an explicit interaction mode (cooperation instead of coordination).Contractual processes have their own procedural forms, their own path depen-dencies Of course one can always abandon thebargaining at any time and ventilateother offers from third parties(¼ex ante 1 – 4) To this degree the competitiveness
of the market has an effect But in the contractual process a phase turnaroundoccurs: partners condition each other reciprocally, gain trust, win new options in thecollaborative process (which can no longer be equalized by third party alternatives,unless there is mistrust) The trust defines the relative exclusiveness of the contrac-tual consensus (as a lock-in) One varies, but within the partnership and not incompetition with one another Once such a state has been achieved competitive-ness is transformed into cooperation Successful contractual processes aremutual agreements with an emerging exclusion of competitiveness
Cooperation is not merely a formal finale; it emerges during a process Allbargaining is potentially capable of lock-in Trust is the outcome of the negotiationand not an external precondition The same is true of fairness (in fulfilling thecontract (see Thu¨sing2007; Priddat2010b,c))
This has consequences for the rational choice basis of economics Rationalchoice is based on competitiveness and mistrust; but contracts are built on trust(i.e – the exclusion of mistrust in certain phases and hence the exclusion of thetactical exploitation of the partner) If mistrusting actors are assumed, who maxi-mize their own expectational values, trust without mechanisms of control cannot beexplained As games theory predicts, such actors secure for themselves the paymentwhich they can force through with the maximum application of their own means,without having to depend on the goodwill of the others As is well-known, rationalactors prefer guaranteed payments from uncooperative solutions to exploitablecooperations, even if successful cooperations would be more advantageous for allthe participants (The Prisoner’s Dilemma)
Trust in people goes hand in hand with the credibility of their promises The
“promise” has been an economic topic for a long time In the modern period,particularly in modern economics, the older romantic kind of trust in the uncondi-tional credibility of a promise (and hence faithfulness) has been replaced by othertime-bound relationships of trust, which resemble more a contract than an eternallyvalid bond (Priddat 2005,2010b) What is common to both kinds of trust is thatcertain behavioural alternatives are suppressed –either totally or temporarily Suchsuppression of what is latently there suggests the presence of emotional influences
In trusting relationships possible breaches of trust disappear into unmarked spheres
of behavioural space and are for the moment ignored, so that the possibilities of the
Trang 39marked side can be explored in peace If emotions steer this suppression, this hasnot so much to do with normativity as with the construction of social arenas, which
is impossible or at least improbable in the neo-classicaldefault case
Contracts (contracts characterized by negotiations) are auto-generative or gent events which do not exclude alternative evaluations (exogenous phase 1), butbreak off at a certain point (exogenous phase II) In phase II the negotiations areconducted with trust, i.e in the expectation of reciprocally coinciding benefits (thecollective result being estimated as higher than the individual result of bothparties¼ mutual agreement) This occurs from the moment the parties have acommon model of the possible contract at their disposal (shared mental model ofcontract gains or contractual frame), when the mutual advantages outweigh thebenefits of the alternatives (which remainunframed)
emer-To this end it is necessary to enter openly into thecontracting: it is only possible
to talk with one another, to allow oneself to be convinced when one’s position is notalready fixed (Oberwittler2010: C1: col.2) If one enters into the contracting withthe firm intention of winning, the room for manoeuvre of the other party is toonarrow; the result will not then be considered as fair, but as a compromise (for manywho are unable to deal with conflict the most comfortable but not the best solution).Two basic patterns can be distinguished: overreaching contracts which are unfairbut comfortable and fair contracts which however assume a symmetrical startingposition (see theHarvard negotiation concept (Fisher et al.2000))
Contracts are arenas of cooperation gains Themutuality achieved in phase II(mutuality¼ 1) allows the partners to abandon their original ideas of the benefit forthemselves on account of the shared positions achieved in the negotiations and togenerate new preferences or concepts of benefit (cf team preferences or teamutilities (Sudgen 2000, 2002, see also Kabalak and Priddat 2010)) Instead ofpre-stabilized rational choice coincidence we are now dealing with open roomfor negotiation, which must be characterized on the one hand as innovative and onthe other hand as cooperative rationality Williamson’s transposition from choice
to contract must be expanded: from contract to cooperation When it comes intophase II every contract has a cooperative finale, which pursues a different logicthan the logic of coordinative coincidence (Phase I)
2.3 From Choice to Contract (I) from Contract
to Cooperation (II)
What is at issue here is not general illusions of community but precisely theexclusion of competition in phase II of the formative process of the contract Atthe core of the market process, which we have become accustomed to considering
as competitivesui generis, the contract first becomes productive when it generates aform of mutuality which, as cooperation, stands in contrast to competition Just asthe companies as hierarchical organizations representexpressis verbis non-market
2.3 From Choice to Contract (I) from Contract to Cooperation (II) 31
Trang 40operators – Coase’s discovery in 1937 – the contractual forms of the customarymarket transactions are in the same way based on non-competitive structures of thefinal contractual cooperation, and primarily on processes of mutual appreciation.
At the core of the market process we are dealing with an emerging processualform, which provides full freedom to reach an agreement This means that it ispossible to redefine the competitive conditions under which contracts were firstnegotiated in accordance with their temporary status (Phase II) The contractualfreedom refers not only to the transactions but also to the conditions of the contract,
to the contractual form and its emerging obligations Instead of the neo-classicaloptimization of what is already given it is possible to create new givens in theongoing contractual process, which develop their own local equilibria They are,suigeneris, potentially innovative instances, which have a Hayekian quality: that theypursue their own discovery procedures in the market through which they canchange their starting conditions
It is important to note that in phase II of the bargaining and negotiating processthe discovery process is oriented on the reciprocal potentials of the partners and nolonger, or only marginally, on the market We are dealing with an exclusion of thirdparties in the course of the process (both competitive offers from third parties andsafeguarding clauses) This specific intervention of phase II is based on the trustdeveloped in the process, which functionally generates the exclusion of alternatives(on this function of trust asemotional pre-contracting (see Muramatsu and Hanoch
2005)) This is often not the result of deliberate decisions taken beforehand, but ofagglomerated processual outcomes which have reached a certain level of specific-ity, so that comparability with third party or external alternative offers no longerseems possible, as negotiations with these parties have not taken place or have come
to an end In phase II the process of contractual negotiations has crossed a thresholdand achieved an intrinsic value, a closed recursive form, which can longer bereadily mediated to third parties The trust-building role of “empathy” in thisprocess (Singer and Fehr2005) is revealed in some of the more recent researchoutcomes of neuro-economics, which see the cognitive aspects of decision-making
as being systematically linked with the affective aspects
But independently of “shared feelings” (Singer and Fehr2005) the contractualnegotiations reach shared foci which are unique in regard to their specific relation-ships (as other negotiations with other partners would create different interventions.Accordingly every phase II process is unique and only conditionally comparable(this bounded comparability could only be done away with again at the expense ofthe specificity; by bringing price or service offers of third parties into play onewould only sketch the naked starting point of possible negotiations) but not theagreements already generated or still to be generated in the course of the negotia-tions) A decision, for example, to break off phase II would devalue the in themeantime specifically accumulated negotiational gains The alternative negotia-tions would first have to be carried out without knowing whether one could achievethe same intervention In a different terminology: thesocial capital invested in themeantime in the process of negotiating the contract would be completely devalued
It turns out that the advantage of economics allowing comparisons to alternative