We use cookies to ensure that we give you the best experience on our website. You can change your cookie settings at any time. Otherwise, we'll assume you're OK to continue.

Department of Philosophy


Publication details for Prof Nancy Cartwright

Cartwright, N. (1995). Causal Structures in Econometrics Models. In On the Reliability of Economic Models. Little, D. Münster: Kluwer Academic Publishers. 42: 63-89.

Author(s) from Durham


There has recently been a renewal of interest among economists, especially econometricians, in deep versus shallow parameters, autonomous versus confluent equations, and fundamental versus phenomenological laws. What is at stake in these concerns is not the truth of various findings of economics but rather their range of reliability. The worry is that the parameters that we can measure are shallow, while the laws that we can establish are, in the terms of one of the founders of econometrics, Ragnar Frish, confluent, or as philosophers would say phenomenological. Only deep parameters and fundamental or autonomous laws are sure to hold come what may. (I take it that in econometrics, “Come what may” means, “Hopefully over the next one to five years, through a range of relatively minor policy changes we might envisage making.”) The idea is that we can measure only shallow parameters and confluent or reduced-form equations, whereas what we need for stability are autonomous or fundamental laws determined by deep parameters. So (if that’s the case) the empirical methods of economics are no good at finding out what we need to know in order to make predictions about what will happen in changing circumstances. The very possibility of forecasting is threatened and with it the hope for rational policy formulation.