Gibrat’s Law and Firm Growth Determinants

Since the argument of Gibrat (1931) that firm growth (increasing total employment) is independently decided upon the previous size of firm (the number of employees), much subsequent research tried to reconfirm his argument. A large body of research supports his argument and most of other research disagrees with him. Furthermore, papers which have been recently published use alternative size measures, such as sale, net asset, or productivity and often deny Gibrat’s law; and they also often segregate data with some classification criteria, such as industry, country or firm size. In this paper, we try to find important firm growth determinants and test whether Gibrat’s law is still held or not. We use three different size measures, total asset, the amount of sale, and net income to test a firm’s growth. We, especially, include net income as a new firm’s size measure, as increasing net income is one of the most important matters for firms to achieve. We believe that net income can be a good firm size measure. We also classify our sample based on industries, manufacturing and service firms, and age, young and old firms; and try to investigate whether different industries or age affect firm growth or not. In this paper, we find that there are different important determinants for firm growth in line with different measurements for a firm’s size. For example, in terms of the magnitude of coefficients, R&D expenses and ROA are important determinants for total asset growth; R&D expenses and Investment are important for sale growth; and finally, debt level is important for the growth of earnings. Namely, R&D expenditures increase total assets and the amount of sales for firms. It also shows the importance of capital structure for net income growth. Above all, we cannot forget to emphasise the roles of industry and the age of firms. Arguably, the industry in which a firm operates and its age are important firm growth determinants. For instance, industry dummy variables and a firm’s age in our results show a consistently significant association with a firm’s growth across our three different firm size measures. In addition, using a simple comparison of a firm’s growth after segregating data into quartile in accordance with prior firm size, we also find that small sized firms in the previous year have a fast growth in the next year across all three size measures as seen with regression estimators. In other words, the firms with a small total asset, a small amount of sale, or a small net income in previous years would grow faster in the coming year; and this effect lasts such a long period of time. In other words, firms in the first quartile of total assets or the amount of sales, or the second quartile of net income in the previous years show the fastest growth. The overall conclusion in this paper is that Gibrat’s law does not hold with our data; and therefore, each firm has important firm growth determinants in line with its characteristics.