Premium
A Statistical Illusion of Large‐firm Conformity to Gibrat's Law
Author(s) -
Giordano James N.
Publication year - 2014
Publication title -
managerial and decision economics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.288
H-Index - 51
eISSN - 1099-1468
pISSN - 0143-6570
DOI - 10.1002/mde.2615
Subject(s) - log normal distribution , conformity , variance (accounting) , illusion , econometrics , economics , statistics , distribution (mathematics) , mathematics , law , psychology , accounting , political science , mathematical analysis , neuroscience
When mean growth rates decline across large size classes of firms, the differences can become too small to achieve statistical significance, creating an illusion of proportionate size‐independent growth (Gibrat's law). Thus, a final confirmation of the law requires a lognormal size distribution with a secularly increasing variance (growing firm size inequality). An empirical test on the less‐than‐truckload trucking industry demonstrates the illusion. The growth rate mean and variance between the two largest classes are statistically indistinguishable, despite an actual decline in both. Their combined size distribution gives a poor fit to lognormal, however, so large‐firm conformity to the law is rejected. Copyright © 2013 John Wiley & Sons, Ltd.