The Distribution of Firm Size in Africa's Manufacturing Sector and Its Implication for Industrial Policy

Abstract

This paper analyzes the distribution of Kenyan firms into four size categories: micro, small, medium and large size. Four questions are investigated. First, do small firms grow faster than large ones or is firm growth independent of size as stipulated by Gibrat's “Law of Proportionate Effect” Second, what is the steady state size of firms? Third, how long does it take to reach the steady state size? Fourth, does the use of bank credit affect a firm's growth and the process of firm size convergence? On the basis of ergodic probability distributions, we derive information on firms' steady state growth and convergence. Using data on Kenya's manufacturing sector, empirical results suggest that firm growth is associated with overall economic performance. Before the early 1990s, firms had a high growth potential. In equilibrium, 70 percent of surviving firms converged to large size. In contrast, growth and convergence in the 1990s reflected the economic crisis that hit the Kenyan economy: the steady state distribution of firm size was concentrated in the first two quartiles, with suggestive evidence that smaller firms recorded the highest rates of failure. Also, the use of credit increased the growth of surviving firms but appears to have contributed to precipitating firm failure. These results suggest that support policies such as special credit schemes are not a panacea for a firm's survival and growth.