Causality in Economics
Stephen F. LeRoy · 2006
Formal analysis of causal relations using graphical methods has become increasingly popular in the natural and social sciences, but has been used much less in economics. The reason may be that heretofore graphical methods have been based on models that are structural in the sense that the equality symbol is taken to denote causation directly. We argue that current economic models are not structural in this sense. This paper proposes a formal analysis of causality that applies to economic models and develops its properties. Among other results, it is shown that Granger causality can be connected to causality only under very strong assumptions. Also, it is shown that graphical methods are not likely to Þnd much application in determining causal orderings as deÞned in this paper in economic models. Formal analysis of causal relations using graphical methods has become increasingly popular in the natural and social sciences. Pearl [22] is a good sample of this work. Graphical methods are a development of analytical techniques that originated in economic theory–speciÞcally, they take as their basis structural economic models as deÞned by the Cowles group half a century ago. The standard reference on causation is Simon [26]. The term “structural” was was given several distinct, though related, meanings by the Cowles economists, as has frequently been observed. At a minimum, the term refers to the distinction between the structural form and the reduced form of a model. In the structural form each internal variable is expressed as a function of some other internal variables and some external variables, whereas the reduced form referred to the solution of a model, in which each internal variable is expressed as a function of the external variables alone. As the term implies, the structural form was viewed as more fundamental than the reduced form. It was seen as containing information not