Preface to ‘Economic Models as Credible Worlds or as Isolating Tools?’

Abstract

Economics is often criticized for its modelling practices. Typical charges include claims that modelling assumptions are unrealistic, that modelling is overly formalistic, that models are hardly ever tested and that model results do not say anything about the real world. Given these shortcomings, critics argue, economics is not fit to predict or explain, or to give useful policy advice. In fact, some claim on these grounds that economics is not a science at all. Many of these criticisms were put forward in the context of the so-called realism of assumptions debate, which centred on various attempts to either establish the realism of theoretical assumptions in economics, or to justify their non-realism. Interest in this debate waned somewhat in the 1970s, as many economists adopted Milton Friedman’s ‘as-if’ approach to justify their modelling practices. However, the central issue remained unresolved. In particular, it became clear that the predictive success of economic models was very limited, and that it alone could not justify standard modelling practices in the way defenders of the ‘as-if’ had hoped. In more recent times, a perspectival shift has reinvigorated the debate. Instead of focussing on the assumptions themselves, recent discussions have focussed on the epistemology of economic models instead. This newer perspective generally acknowledges the autonomous role that models and modelling plays in scientific theorising. It often also acknowledges that economic models and modelling practices have characteristics that set them apart from other scientific models, in particular from physics. Two main positions can be distinguished in the newer debate. The isolationists regard modelling as a way to isolate causal factors or capacities of the real world. In analogy to laboratory experiments, models allow for theoretically removing the influence of some factors on others in a given situation. Such models then constitute