This thesis is an exercise in applied methodology. Ideas in the history and philosophy of science which have proved to be influential in the methodology of economics, and in shaping economists' self-image in this regard, are selected for closer analysis and criticism. The main ideas that are addressed are those of empiricism, with emphasis on the methodological falsificationism of Karl Popper and Imre Lakatos, and Laudan's problem solving model of scientific progress . The thesis examines the relationship between empirical evidence, in the form of both econometric test results and stylized facts, and the development of theories about exchange rates and the open economy. This analysis begins with Cassel's formulation of purchasing power parity theory in 1916, through the elasticities, absorption, and Mundell-Fleming models of exchange rates and the balance of payments, up to the present day monetary and asset market models. This is done with regard to the broad methodological issues examined earlier in the thesis. Some of the main empirical and methodological difficulties in testing such theories are addressed, with particular reference to the role played by the Duhem-Quine thesis and the ceteris paribus assumption. Although some of these difficulties may be regarded as a matter of degree compared to similar problems in the natural sciences, it is argued that this difference is significant for the workability of falsification in economics . Moreover, the presence of hypotheses about expectations in many economic theories appears to be a substantive difference such that the difficulties posed by the Duhem-Quine thesis apply with much greater force in a social science like economics. The main conclusions are that neither the Popperian nor Lakatosian versions of falsification are seriously practiced in the area of exchange rate economics and that, unlike the position taken by advocates of falsification such as Mark Blaug, it would be inappropriate and misguided to do so. A tentative case is made, with some reference to the theories surveyed in this thesis, for the possibly greater relevance of Laudan' s more pragmatic problem solving model for the methodology of economics, particularly as regards his analysis and emphasis on conceptual problem solving in the progress of knowledge.
| Identifer | oai:union.ndltd.org:netd.ac.za/oai:union.ndltd.org:rhodes/vital:1012 |
| Date | January 1997 |
| Creators | Hodge, Duncan |
| Publisher | Rhodes University, Faculty of Commerce, Economics |
| Source Sets | South African National ETD Portal |
| Language | English |
| Detected Language | English |
| Type | Thesis, Doctoral, PhD |
| Format | 261 leaves, pdf |
| Rights | Hodge, Duncan |
Page generated in 0.0019 seconds