Frank Hindriks — False Models as Explanatory Engines

Abstract. Many models in economics are very unrealistic. At the same time, economists put a  lot of effort in making their models more realistic. I argue that in many cases, including the  Modigliani-Miller irrelevance theorem investigated in this paper, the purpose of this process  of concretization is explanatory. When evaluated in combination with their assumptions, a  highly unrealistic model may well be true. The purpose of relaxing an unrealistic assumption,  then, need not be to move from a false model to a true one. Instead, it may be providing an  explanation of some phenomenon by invoking the factor that figures in the assumption. This  idea is developed in terms of the contrastive account of explanation. It is argued that  economists use highly unrealistic assumptions in order to determine a contrast that is worth  explaining. The process of concretization also motivates new explanatory questions. A high  degree of explanatory power, then, may well be due to a high number of unrealistic  assumptions. Thus, highly unrealistic models can be powerful explanatory engines. 

False Models as Explanatory Engines
Frank Hindriks
Forthcoming in Philosophy of the Social Sciences