We use a panel of Italian manufacturing firms for the period 2001-2014 to analyse the distribution of firm size, and then test for the validity of Gibrat’s law using unit root tests. Although Gibrat’s Law is rejected and the estimates suggest that small firms grow faster than larger ones, we do not observe a significant change in the average size of companies at the end of the period under investigation. Also, by using a long-run Transition Probability Matrix, we verify that the steady-state distribution of firm size remains stable. The higher propensity to grow shown by smaller firms is confined to the size class in which the firm is established. We further investigate the relationship between the rate of growth in a firm’s size conditional on specific firm and industry characteristics. Export intensity plays a significant role in affecting the size growth rate together with industry characteristics related to technological levels. Finally, we estimate the probability that a firm increases in size relative to the mean size prevailing in its own size class over a 14-year interval. This approach enables us to highlight those factors that affect this probability, thereby enabling us to underline how Gibrat’s Law tests, although important, require complementary analysis to ascertain whether a firm’s propensity to increase in size is a long run effect and thus a significant modification of the distribution of company size or only implies a marginal increase in size within a reference size class.
|Publisher||Vita e Pensiero|
|Number of pages||29|
|Publication status||Published - 2018|
- Gibrat's law
- firm size distribution