Analyse & Kritik

Journal of Philosophy and Social Theory

Philosophical and Methodologial Issues in Economics

2004 (26) Issue 2
Guest-Editors: Mark S. Peacock / Michael Schefczyk


The 'dismal science of economics', as it was once called, has a mixed reputation. Some praise its clarity and elegance whilst others bewail its futility; others laud the precision of its mathematical form whereas others still descry the source of its irrelevance and unrealism in just this form. Many feel that precision and mathematisation are bought at a price too high, namely unrealistic assumptions, empty models with little or no explanatory power, unreliable predictions and a general state of affairs in which the theoretical world of economics has all but parted company from its 'real' counterpart.

When, in the winter of 2003, we issued our call for papers concerning philosophical and methodological issues in economics, we hoped to solicit contributions which would look at the methodology of economics in a nuanced way, taking neither a merely apologetic/affirmative nor a critical/dismissive stance on the matter. Of the many issues which, we deemed, are worthy of reflection, we were interested in shedding light on the epistemic goals of economics - what and how economics seeks to explain, describe or understand. This topic is necessarily related to the criteria for the adequate explanation, description or understanding of a phenomenon. Is adequacy in these areas an empirical matter or are, as is often held to be the case, the empirical shortcomings of economics indecisive when it comes to judging the adequacy of economic theory? Should economic theory emulate the methodology of other theories, e.g., those of the natural sciences, or would such comparisons flounder on the nature of the subject matter of economics the peculiarity of which makes such emulation inadvisable? We are satisfied that the four contributions to this symposium, by philosophers and economists, tackle crucial methodological topics of the `dismal science' and that they will provide stimulus to further debate.

Olaf L. Müller examines in Autodetermination in Microeconomics the scientific status and empirical content of the theory of demand (as presented in the microeconomics textbook). Is the theory of demand tautological or does in have empirical content? Müller argues that, viewed holistically, the theory of demand does indeed have empirical content and can hence be tested according to empirical criteria. Yet that content, he argues, is plainly false. In order to make sense of this from a philosophical point of view, Müller draws on the concept of 'autodetermination', recently introduced into the philosophy of physics by Oliver Timmer. According to this concept, a science specifies its own domain of application and hence empirical findings without this domain do not constitute negative evidence which may lead to the rejection of the theory; reality can be selectively adapted to theory rather than vice versa.

Julian Reiss continues the empirical theme in his Evidence-Based Economics. He outlines what would be a sound empirical basis for economics and discusses this basis with respect to three methodological problems - those of measurement, induction and idealisation - to which methodologists have not been lavish with their attention. Using Friedman and Schwartz's work on money, Reiss carefully examines the three aspects of evidence cited; he asks what may be rightly called adequate measurement, what role induction may play in economics and how it relates to causal relationships and, finally, whether and when idealisations are acceptable.

In Perfect and Bounded Rationality, Werner Güth and Hartmut Kliemt offer an outline of a general theory of 'games and boundedly rational economic behaviour'. The authors expect this theory to have greater explanatory power than standard approaches to economics based on the notion of maximisation. Although this explanatory power may come at the expense of a less general theory of rationality, the authors see no merits in generality if the latter is concomitant with irrelevance. Güth and Kliemt suggest strategies for future research into decision-making procedures of real individuals, not those which populate the models of most economic models. Their work raises the question why economists pursue models based on maximisation if, as Güth and Kliemt argue, they are so unsuccessful in explaining much of economic behaviour. Empirical success does not seem to be the criterion for adopting this approach.

The contribution of Till Grüne, The Problems of Testing Preference Axioms with Revealed Preference Theory, presents an analysis of revealed preference theory. It argues that testing choice data for a violation of certain axioms of the theory says little about the validity of those axioms because 'testing' axioms and interpreting test results is far more complicated than one often supposes. This applies alike to experimental and non-experimental work on choice behaviour. Although experimental evidence can be used to judge elements of economic theory, protagonists of such work would do well to consider the issues to which Grüne draws attention when they use experimental evidence as the basis of theory evaluation.

The last article in this volume is not a contribution to the symposium. But Sonja Vogt and Jeroen Weesie examine in the tradition of an economic approach Social Support among Heterogeneous Partners. In their paper they derive several hypotheses on how dyadic social support is affected by the heterogeneity of the actors. They distinguish heterogeneity in respect to the likelihood of needing support, the benefits from support relative to the costs for providing support, and time preferences. Their analysis yield a series of theorems which shed new light on the research of similarity in social support theory.

Mark Peacock, Michael Schefczyk

Less →

Table of Contents

Title: Autodetermination in Microeconomics. A Methodological Case Study on the Theory of Demand
Author: Olaf L. Müller
Page: 319-345

Abstract: My philosophical case study concerns textbook presentations of the theory of demand. Does this theory contain anything more than just a collection of tautologies? In order to determine its empirical content, it must be viewed holistically. But then, the theory implies false factual claims. We can avoid this result by embracing the theory's normative character. The resulting consequences will be illuminated with the new autodetermination thesis recently proposed in the philosophy of physics by Oliver Timmer. Applying his ideas to the theory of demand reveals that the statements of this discipline simultaneously concern both values and acts.

Title: Evidence-Based Economics. Issues and Some Preliminary Answers
Author: Julian Reiss
Page: 346-363

Abstract: This paper presents an outline of a methodology of ?evidence-based economics?. The question whether an economic statement is evidence-based must be answered on three different levels. The first level concerns measurement: it asks whether claims made about economic quantities such as inflation, unemployment, growth or poverty are justified by the data and measurement procedures. The second level concerns induction: it asks whether claims made about the relations between economic quantities (such as ?number of babies born predicts growth?, ?change in money causes change in monetary income?, ?non-borrowed reserves can be used to control the interest rate?), are justified by the inference procedures. The third level concerns idealisation: it asks whether the quantities and relations selected are justified by the stated aim of the inquiry. The paper provides a discussion of these three types of investigation and of some solutions that have been offered.

Title: Perfect or Bounded Rationality? Some Facts, Speculations and Proposals
Author: Werner Güth / Hartmut Kliemt
Page: 364-381

Abstract: Simple game experiments of the reward allocation, dictator and ultimatum type are used to demonstrate that true explanations of social phenomena cannot conceivably be derived in terms of the perfect rationality concept underlying neo-classical economics. We explore in some depth, if speculatively, how experimental game theory might bring us closer to a new synthesis or at least the nucleus of a general theory of ’games and boundedly rational economic behavior’ with enhanced explanatory power.

Title: The Problems of Testing Preference Axioms with Revealed Preference Theory
Author: Till Grüne
Page: 382-397

Abstract: In economics, it has often been claimed that testing choice data for violation of certain axioms - particularly if the choice data is observed under laboratory conditions - allows conclusions about the validity of certain preference axioms and the neoclassical maximization hypothesis. In this paper I argue that these conclusions are unfounded. In particular, it is unclear what exactly is tested, and the interpretation of the test results are ambiguous. Further, there are plausible reasons why the postulated choice axioms should not hold. Last, these tests make implicit assumptions about beliefs that further blur the interpretations of the results. The tests therefore say little if anything about the validity of certain preference axioms or the maximization hypothesis.

Title: Social Support among Heterogeneous Partners
Author: Sonja Vogt / Jeroen Weesie
Page: 398-422

Abstract: This paper derives hypotheses on how dyadic social support is affected by heterogeneity of the actors. We distinguish heterogeneity with respect to three parameters. First, the likelihood of needing support; second, the benefits from support relative to the costs for providing support; and, third, time preferences. The hypotheses are based on a game theoretic analysis of an iterated Support Game. We predict that, given homogeneity in two of these parameters, the prospect for mutual support is optimal if actors are homogeneous with respect to the third parameter as well. Second, under heterogeneity with respect to two of the parameters, support is most likely if there is a specific heterogeneous distribution with respect to the other parameter that "compensates" for the original heterogeneity. Third, under weak conditions, the overall optimal condition for mutual support is full homogeneity of the actors.