Gibrat's law in the trucking industry

The distribution of firm size in the trucking industry is highly skewed, with a few firms accounting for a disproportionate share of transport. The distribution is also highly disperse so that firms have no characteristic size. This paper demonstrates that the distribution of firms sizes can be well approximated by the Zipf distribution. Such a distribution is consistent with a random growth process that is independent of firm size (Gibrat's law). We use data from the Federal Motor Carrier Safety Administration in 2016 to demonstrate that the Zipf distribution provides a fit that is as good or better than other competing distributions, strongly supporting the model of random growth.