L’hypothèse d’homogénéité dans l’analyse économique : une analyse historique et méthodologique

The homogeneity assumption posits that within a particular market, all goods sold are considered interchangeable (i.e., perfectly substitutable) by economic agents. Adopting this assumption permits the exclusion of product differentiation from market analysis, narrowing the focus solely to price and quantity adjustment mechanisms. This paper aims to reexamine the historical and methodological underpinnings of this assumption. First, this paper will explore the significance of the homogeneity assumption within the framework of "pure" market models, such as pure monopoly, Bertrand and Cournot oligopolies, and perfect competition. These models have been instrumental in shaping economic thought up until the early 20th century. The paper will particularly delve into the reasons behind the adoption of this assumption, largely recognized as a means of simplification (Abbott, 1953). Following this, the paper will shift focus to the emergence of differentiation theory in the 1930s, marked significantly by Hotelling's model (1929), which signaled a departure from the homogeneity assumption. This section will examine the arguments presented by proponents of differentiation theory for moving away from the homogeneity assumption, such as overcoming Bertrand's and Diamond's paradoxes and avoiding the proliferation of markets. Additionally, it will confront the challenges introduced by rejecting the homogeneity assumption, including the complexities of defining market boundaries.