Was Robert Gibrat right? A test based on the graphical model methodology

Using both regression analysis and an unsupervised graphical model approach (never applied before to this issue), we confirm the rejection of Gibrat's Law (stating that a firm's growth is independent of that firm's initial size) when our firm-level data are considered over the entire investigated period, while the opposite is true when we allow for market selection; indeed, the growth behavior of the surviving most efficient firms is in line with Gibrat's Law. This evidence reconciles early and current literature and may have interesting implications in terms of both theoretical research and policy suggestions regarding subsidies to small firms, which do not necessarily grow faster than their larger counterparts.Challenging Gibrat's Law, this study reveals that small Italian firms initially outpace larger ones in growth, but selection evens the field over time; this evidence calls for smarter and targeted policies. Indeed, our analysis challenges the widely accepted result that small firms grow faster than their larger counterparts, thus rejecting Gibrat's Law stating that a firm's growth is independent of that firm's initial size. Using a unique combination of regression analysis and innovative graphical models, we tracked newly founded Italian manufacturing firms over 11 years. We discovered that initially, smaller firms grow faster, rejecting Gibrat's Law. However, over time, as less efficient firms exit the market, the surviving, more efficient firms display growth patterns consistent with Gibrat's Law. This finding bridges the gap between previous and recent studies on firm growth. The study's key implication is for policy makers: support for young and small firms is crucial, but policy focus should shift to ensuring markets function efficiently, and aiding the most promising businesses. This approach could foster a more dynamic and robust economic environment, benefiting society by promoting sustainable business growth and stability.