Resumen
The purpose of this empirical study is to investigate whether the growth process of firms is best explained essentially by a random process as envisaged by Gibrats law, or by identifiable systematic influences such as growth persistence and firm size. A dynamic random coefficients model is applied to data on 260 Canadian firms classified into four groups according to firm size. Gibrats law of proportionate effect is not supported by the empirical results. Specifically, the findings indicate that smaller firms grow faster than larger ones in all cases. However, they also show that the effect of the disadvantage of size on growth is somewhat different for each group.