Global Models and Global Mechanisms III: Towards a Framework of Global Mechanisms for Global Models

Millendorfer, H. (1975). Global Models and Global Mechanisms III: Towards a Framework of Global Mechanisms for Global Models. IIASA Working Paper. IIASA, Laxenburg, Austria: WP-75-147

Preview

Text WP-75-147.pdf Download (900kB) | Preview

Abstract

As stated in "Global Models and Global Mechanisms I," the situation of social science -- described for the economic sector in H. Leontieff's amazing paper, "Theoretical Assumptions and Nonobserved Facts," -- is reflected in the state-of-the-art of world models. An impressive, intellectual effort and highly sophisticated mathematical methods are applied to handle complex sets of some hundred or thousand equations, while the structure of the whole model is mainly based on "theoretical assumptions and unobserved facts." The corresponding, poor empirical evidence is not even improved by the use of some ten thousand or hundred thousand statistical data structured according to nonverified assumptions.

For example, at the Third IIASA Global Modeling Conference on Food and Agriculture, decisive questions remained unresolved, i.e., diminishing returns of fertilizer in agricultural production; human capital as an important source of increasing productivity; the ecological capacity of rivers for agro-chemical pollutions. The reason for this being a lack of empirical observations, or even a lack of knowledge of existing empirical observations (e.g. observation of the U.S. Agricultural Department on diminishing returns of fertilizers). The deficiencies in the Models caused by the lack of behavioral equations, which should be sufficiently based on empirical observations, cannot be offset even by the highest sophisticated theoretical methods if the model builder claims that the model reflects the part of reality relevant for the question which is asked and if the decision makers should apply this to their questions. This is no way to solve their problems without an intensive effort in empirical investigations for interaction with theoretical considerations.

Applied systems analysis also needs the hard empirical work. The learning process in an iterative empirical--theoretical research-strategy leading to a step-by-step approximation between model and reality is described in Part I of "Global Models and Global Mechanisms: Methodological Considerations." The results of an application of this methodology to the question of long term economic development is presented in Part II.

The following paper outlines how the methodology can be applied to other questions and how these questions are connected to the concept of the General Production Function as a basis for interdisciplinary research, leading to a framework of consistent global mechanisms which can be used as empirically tested behavioral equations for global models.

Actions (login required)

View Item