A New Methodology for the History of Economic Thought

1. Introduction

Many leading economists are concerned about the future of the history of economic thought. Even though all recognize the importance of this discipline, they express pessimistic predictions concerning its future. Indeed these views reflect a strong reality in the academic world as the decreasing opportunities to pursue graduate studies in economic history clearly show it. In spite of the fact that economic sciences do not let much room for history of economic thought anymore, many young scholars choose to pursue a career in this field. Is there a future for them? Is history of economic thought a useless, old-fashioned discipline with no interest for current economic developments?

If we follow up the assumption that an observation and an understanding of the past are necessary for current economic theory, history of economic thought is indispensable in order to elaborate solutions to modern economic issues. That is how we envisage the future of history of economic thought as a powerful process of interdisciplinary analysis.

An overview of the “methodology” we use in our research should show how history of economic thought can contribute to provide relevant answers to current economic issues, by linking history of economic thought and current economic theory. Our methodology stresses two main features to that end: first, one has to articulate as rigorously as possible the positive and the normative economics; second, one has to prove the relevance of our results thanks to modern analytical tools.

Our paper is organized as follows: a second section outlines our methodology. Two examples of studies which are based on it are given in a third section. A fourth section concludes.


2. Describing the suggested methodology

Our methodology consists in carrying out analytical and applied developments closely related to historical and conceptual analysis. Our assumption is that our methodology is relevant for any modern economic issue and could be applied in order to propose new perspectives and solutions to deal with it. To be more specific, for any concept with whom economists deal, one must go through three stages: in the first place, the concept must be defined very precisely thanks to history of economic thought; in other words, the concept is examined in reference to the works of different authors. Hence, the first step of our methodology is devoted to a thorough analysis of the founding texts of this given concept in order to determine the value judgments it provoked. Secondly, the concept is applied to one or many fields of economic theory in order to analyze concrete problems it raises and to attempt to suggest new solutions for them. Indeed the second step of our methodology endeavors to propose solutions to a given set of problems, which the economic theory faces in different fields. The objective of this part is to articulate as rigorously as possible the positive and the normative analysis in order to show that the positive solutions proposed for a paradox or a problem are necessarily dependent on the value judgments, which govern the selected conceptual apparatus. And in our third stage, we test the proposed solutions resorting to contemporary economic instruments: for instance, applied econometrics or experimental economics. This last step enables us to check the validity of our process and, in particular, the relevance of the solutions we designed.

It must be noted that our methodology is connected to these applied in modern theories of justice. According to Fleurbaey (1996), a theory of justice – which combines philosophical and economic theories – is developed in three stages: 1) the selection of a philosophical theory of justice which advocates some moral values; 2) one or many axioms can thus be stemmed from them: this second step corresponds to an interpretative work for which one has to obtain a conformity between the moral values and the elaborated axioms; 3) in the third step which pertains to logical analysis, one has to examine the compatibility of the axioms. As regards our methodology the second step – applying a concept to one or many fields of economic theory – is similar to the second and third stages of Fleurbaey's. But our first step is different from his. For our part, we apply at the chosen concept in economic theory the methods used in history of economic thought, i.e., the comparative review of literature and text analysis. And our methodology introduces a third dimension: to test the relevance of the solutions we obtained so as to establish their pertinence.

Thus, our methodology is interdisciplinary by nature since it involves moral and political philosophy, history of economic thought, modern economic analysis and aims to contribute to current developments of economic analysis.

The first main characteristic of our methodology consists in combining a conceptual reflection, which resorts to the developments of economic thought, and an analytical process applying modern tools of economic theory. Therefore, the fundamental elements which reveal the real structure of the concept we cope with are stressed: the value and fact judgments it implies and the combination of normative and positive economics. We consider that it is necessary to link explicitly a moral value to its analytical treatment. One must overcome the classical distinction between normative and positive economics elaborated by Keynes (1890) and attempt to combine normative and positive judgments. This is strongly emphasized as well by Roemer (1996, 1998), Sen (1999) or Maniquet (1999).

The second main characteristic of our methodology consists in testing the results brought out by the theoretical framework we have used by the means of experimental economics and cliometrics. This analysis takes place in the third stage of our methodology which aims at examining the results' relevancy by appreciating the conclusions' validity.

Let us focus here on the use of cliometrics as a process of empirical validation. Cliometrics, that is quantitative economic history, relies on historical econometrics of time series analysis so as to, in the last instance, establish causality relationships between economic variables (Diebolt (2005), Diebolt, Jaoul and San Martino (2005)). Thus, cliometric analysis is undertaken in order to demonstrate if empirical results corroborate the theoretical results we previously proposed. We note that, when empirical validation relies on cliometrics, the historical analysis plays a predominant part in the whole analytical process since it involves history of economic thought and quantitative economic history.


3. How to apply our methodology? Two examples

We develop two examples of studies – finished or current –, which are based on our methodology in order to illustrate our discussion. The first study attempts to define and to apply a concept of freedom, whereas the second one deals with education.


3.1. Defining and applying a concept of freedom

We first propose an example of the methodology we advocate, by putting forward the study of Igersheim (2004), which attempts to bring an answer to the following questions: which concept of freedom should be used in a theory of distributive justice or, more simply, when a redistribution policy is decided? How, in the particular context of social choice theory, public actions should be determined in order to respect, to protect and to guarantee to the members of a society this concept of freedom? Thus, this study pursues two goals: first, to define an operational and synthetic concept of freedom by examining theories defended by different authors. Second, it applies this concept to a particular context: social choice theory.

Thus, in a first part, elaborating upon the distinction of Constant (1819) between freedom of the ancients and freedom of the moderns, we show that this opposition should be overcome ( aufgehoben ): the necessity to overcome this opposition is strongly underlined by Rawls (1971, 1993, 2001) after Hegel (1821). We then argue that the notion of reconciliation between freedom of the ancients and freedom of the moderns is able in our concept of freedom to reflect the pair freedom of the ancients / freedom of the moderns. But modern bourgeois and ancient citizen cannot be perfect synonyms. The modern individual should try to reconcile these two notions and develop his subjectivity: reconciliation becomes a permanent quest. Therefore, in Hegel and Rawls, “modern” freedom is treated at a high level of abstraction and takes place only in a dimension called ancients / moderns.

We then explore the concrete meaning of “modern” freedom by examining the possibility conditions of the reconciliation. In Rawlsian theory some drawbacks appear: indeed, even if Rawls integrates the economic dimension and mentions formal and real freedoms with his difference principle, the latter remains his major argument. The economic dimension is in fact underestimated in justice as fairness. Moreover the problem of the repartition of physical and intellectual capacities and of the entailed inequalities is explicitly postponed by Rawls. In order to progress in defining the effectivity conditions of freedom another pair of freedoms apt to deal with the economic dimension and that of “talents / handicaps” is considered: the pair negative freedom / positive freedom. We examine Sen's contribution on this issue and the evolution of Sen's thought until the “capability approach” (1985, 1992). Sen stresses the necessity to define an objective measure of well-being which takes into account human diversity in economic and “talents / handicaps” dimensions. But one has to consider individual responsibility as well and Sen's capability does not. Still the concept of capability which reflects the pair negative freedom / positive freedom is integrated in our synthetic concept of freedom. However it will be complete only by integrating the notion of individual responsibility.

Individual responsibility, which is treated by Roemer who introduces original components such as the formalization of effort, is the third and last element of the concept of freedom. First, we study the Rawlsian treatment of responsibility, which is relatively crude: this notion is developed more subtly by his successors, particularly by Roemer (1996, 1998).

Finally, the three components of our synthetic concept of freedom are reconciliation, capability and individual responsibility. This concept of freedom is then used in the second part of the study in order to find some possibility results to impossibility theorems elaborated in the social choice theoretical framework. In particular, Sen's liberal paradox (1970a,b) is examined. The impossibility of a Paretian liberal is the first endeavor to introduce and analyze individual rights and freedoms in the social choice theoretical framework largely defined by Arrow (1951). For this reason the liberal paradox raises the question of a satisfying and implementable definition of individual rights in this framework.

It is claimed that our concept of freedom is able to carry out this role. In order to integrate it into social choice theory, its three main elements are considered: first, the reconciliation between the ancients and the moderns, i.e., maintaining the tension between the State and individuals. But the second pair negative freedom / positive freedom, which is represented by the capability, cannot be considered as such: indeed, the economic dimension is not present into the social choice theoretical framework. In order to take it into account, we just consider the notion of protecting individual rights and freedoms, in other words, the necessity for society to guarantee individual rights. Finally, we consider individual responsibility. We thus invoke these three elements of our concept of freedom in order to elaborate some possibility results for Sen's liberal paradox. We notably develop a Preference Modification Mechanism, which is based on them: it selects and launders individual preferences consistent with the values of society. This mechanism is defined by two rules, which are justified by the operational concept of freedom, i.e., respect of individual rights.

A third part could be added to this study as our methodology emphasized it: the Preference Modification Mechanism could be tested thanks to experimental economics, which aims at checking theoretical predictions against real individual choices. This third part would be in line with the studies concerning the individual cooperative behavior in public good games. They show that individuals, contrary to theoretical predictions, do not play Nash but overcontribute to public good (see, among others, Davis and Holt (1993) or Ledyard (1995) for a survey). As regards our Preference Modification Mechanism a repeated public good game could be elaborated in which participants could choose how to modify their initial preferences: either towards more individual freedoms or towards a renouncement of their individual sovereignty.


3.2. Defining and applying a concept of education

The methodology we support is experimented in a current study dealing with the relationships between justice, education and economic progress (Le Chapelain, 2006). Regarding our work, the problem is the following: how could a fair education system influence the economic development and what could we learn from it as far as public policy is concerned? Our work has three objectives so as to answer that question: defining a concept of justice for the education system relying on historical texts analysis, applying the highlighted concept through a theoretical model and validating the conclusions thanks to cliometrics.

In order to distinguish precisely what we mean by a fair educative system, the first part relies on the history of economic thought itself based on texts analysis. Our study brings out the comparison of Condorcet's writings (1791) about public instruction with Rawls' Theory of Justice (1971). Relying on both – a wish for fairness and a concern for elitism together with a demand for fairness and efficiency –, we demonstrate that Condorcet's and Rawls' views may be linked through the theme of talent. It is indeed thanks to equal chances offered to individuals that one may contemplate elitism and equality at the same time. Thus, on the basis of our analysis, we can specify our concept of justice as the equal opportunity for all individuals to develop their natural talents. We can also stress from our study some other complementary orientations. We can notice that Condorcet's thought advocates the use of methods, which are called today affirmative action methods. It is interesting to see that such currently debated ideas, which concern the concept of equal opportunity in the education system, were already mentioned and even, as in Condorcet's case, defended in previous texts. Therefore we can ask ourselves what are the best means to favor the concept of justice we support.

After having defined our concept of justice, a second part of our work consists in applying this concept to the formalized framework of a theoretical model. But the analysis of relationships between human capital and economic evolution generally takes place in endogenous growth models. Hence our study is carried on in this framework. However we emphasize that the originality of our work is based on the contribution of the concept of justice in the analysis of the connection between education and growth. The attempt to formalize an operational concept of justice can thus be appreciated.

In the last instance, results brought out are submitted to an empirical validation relying on cliometric analysis. Indeed, by using current quantitative methods in the study of historical evolutions, we attempt to determine if the experience attests the validity of causal relationships between growth, education and equity. It is particularly interesting to notice that in the process of empirical validation based on cliometric analysis the appeal to historical analysis becomes prevailing in the methodology we use. Two steps of our analytical methodology are based on historical sources (whether they deal with history of economic thought or quantitative economic history) to solve a problem as important as that of the education system's equity. Some may consider it as a paradox. But, on the contrary, we think that the convincing results of our methodology show the lack of clear-sightedness of those who see no point in resorting to an analysis based on economic history.


4. Conclusion

Thus, our methodology with its three stages – definition of a concept, application, empirical validation – resorts to many current economic tools and links strongly history of economic thought with contemporaneous research in economic theory.

History of economic thought should be seen in this perspective: by including an historic questioning, relying on text analysis and quantitative methods, as a part and parcel of a multidisciplinary analytical process, which is meant to answer a given issue.


Herrade IGERSHEIM & Charlotte LE CHAPELAIN, for AFC.
February 2006.