Epistemological analysis of model building and functions of models in economic science

Žitko, Mislav. (2017). Epistemological analysis of model building and functions of models in economic science. PhD Thesis. Filozofski fakultet u Zagrebu, Department of Philosophy.
(Poslijediplomski doktorski studij filozofije) [mentor Mikulić, Borislav].

Preview

PDF (Croatian) Download (2MB) | Preview

Abstract

The point of entry chosen for the analysis of economic models is the long-standing discussion on the issue of realism in science. Anti-realism has been the dominant position in the first half of the 20th century, as the philosophy of logical positivism reigned supreme. Realism, on the other hand, became more prominent in the second part of the century. In recent times, new theoretical platforms such as constructive empiricism made their appearance and somewhat change the terms of the debate. But the main points of contestations have nevertheless remained the same. The key questions that have emerged in the philosophy of science revolve around the issues of unobservable entities, causality and explanation. These are the issues that have been debated most completely and carefully in the context of philosophy of natural sciences, particulary in the philosophy of physics. Hence, the first task of this thesis is set in terms of revisiting and redefining the ongoing debate in order to make it more pertinent to the field of social sciences. The implicit presupposition is that the epistemological status of the object of inquiry in the respective fields of natural and social sciences is rather different inasmuch as the latter has to take into account human agency and the its inherent reflexivity. Once the nature of the debate between realism and anti-realism is brought to the fore in the manner suitable to the requirements of research in the field of social sciences, the discussion proceeds to address the issue of modeling practices. Given the fact that there are many different ways to build models and to think about the functionality of models, we explore model building as an epistemic genre and seek to examine the level of compatibility between economic modeling and scientific realism. This question is quite important since it is clear that models do not simply describe the world, they rather idealize, simplify or isolate a particular feature of the world. In other words, models are epistemic tools that by definition disfigure and misconstrue the part of the world they investigate or, to put it more accurately, the researcher who uses models does not get a full, flawless and precise picture of the world, but rather a tainted one. Still, models, as it is argued in this thesis, can and often do play an important role in the process of obtaining truthful insights about the world. Furthermore, models can play this role only of the background of true or approximately true theory. The argument developed here rests in the fact that models do not stand alone nor can they be seen as a completely autonomous instruments. They are first and foremost instruments devised and created for the purpose of specifying a particular aspect of theory, so that we could bring the theoretical assumptions and the material world more closer together. We show how this is done on the level of theoretical and empirical models in economics. On the theoretical level the operating procedures of neoclassical economics are brought to scrutiny. We review the history of theoretical modelling in micro- and macroeconomics and show how the process of naturalization of modelling in economics has been related to the quatification and mathematization of the discipline. Special care is given to the issue of microfoundation and we review the reasons for preeminence of microeconomics in neoclassical theory. On the level of empirical modelling the central focus is put on econometric models and the problem of capturing the causal relationships between various phenomena of interest. We survey the debate on the relevance of econometric modelling that began with the exchange between Keynes and Tinbergen and proceed to see whether econometrics can enable us to meet the aspirations of scientific realism. In the last part of the thesis the more general aspects of knowledge production in field of economics are analyzed given the fact that the problems of ideological and political influence have been present in relation to the economic discipline from the very beginning. Through the examination of the place of the dichotomy between facts and values, as well as through the inquiry of the role of ideology a distinction between orthodox and heterodox economic discourse is put forward with a view to affirm a progressive scientific research program based on scientific realism.

Actions (login required)

View Item