ABSTRACT
Since Gibrat (1931) the evolution of firms’ size distribution has been an important area in industrial organization (IO), an area where theoretical and empirical work has been crucially intertwined to reciprocally motivate the explanation of statistical regularities and the deep analysis of data. There are plenty of surveys on firms’ performance dynamics, firms’ size distributions and Gibrat's Law; Sutton (1997) and Caves (1998) are two influential cases in point. They are both mainly based on the empirical findings made available by larger datasets on firms in the last decades, with mildly different emphases in their attention to theoretical developments. Our aim in this brief survey is to provide a review with a focus on theoretical models and in particular on the details of how, over a long time span, older types of purely stochastic models were discarded in favour of introducing stochastic elements into standard maximizing models (Sutton, 1997, p. 42), featuring more or less important market imperfections. Although this may be of interest in itself, the motivation of our focus is driven by the view that this perspective can allow us to select from this literature, which is mainly positive in spirit, some broad (industrial) policy views and lessons.
