|Type||Thesis or Dissertation - Master's Thesis|
|Title||Firm Size, Age and Growth in South Africa|
The relationship between a firm’s size, age and proportional growth rate is examined using multiple samples of South African firm-level data from the early to mid-2000s. The foundation of this study is Gibrat’s Law of Proportionate Effect
(Gibrat, 1931), which states that a firm’s proportional growth rate is independent of its absolute size at the start of a given period. It is assumed that firm growth follows a random walk and, therefore, should not be affected by firm size. An implication of Gibrat’s Law of Proportionate Effect is that the firm size distribution is lognormal. However, based on both empirical and theoretical literature, this theory of firm growth has fallen out of favour and been replaced by the proposal that there is an inverse relationship between a firm’s proportional growth rate and both its size and age. Two questions are evaluated in this research using the samples of South African firms. The first is whether the firm size distribution is lognormal. If this is not the case then Gibrat’s Law of Proportionate Effect can be rejected. However, this approach cannot confirm that Gibrat’s theory is valid and will, therefore, be referred to in this paper as a partial test. It was shown that the log firm size distribution was not normal, but rather right-skewed with a Pareto distribution characterising the upper tail. Consequently, Gibrat’s Law of Proportionate Effect was rejected for the datasets of South African firms. This evidence is largely observational and does not explicitly assess the relationship between proportional growth rates and firm size. Therefore, the second question is whether Gibrat’s Law of Proportionate Effect holds. This was investigated by testing conditions derived from Gibrat’s Law of Proportionate Effect, the results of which can lead to either the rejection or acceptance of this proposition. This study extends Gibrat’s research in order to determine the relationship between firm age and proportional growth. Statistical methods, such as Ordinary Least Squares regressions, considering only firms that survived the period under consideration, were used. The results revealed that Gibrat’s Law of Proportionate Effect was invalid and there was a systematic tendency for the smaller, younger South African firms in the datasets to grow proportionally faster than the larger, older firms. This finding supports the view that firm growth is not entirely random.
|»||South Africa - Cape Town RSC Levy Firm Panel Data 2000-2006|