DOI 10.1007/s11151-005-5053-z
Review of Industrial Organization (2005) 27:303–328 © Springer 2005
Technology and the Size Distribution of Firms:
Evidence from Dutch Manufacturing
ORIETTA MARSILI
Rotterdam School of Management, Erasmus University, Burg. Oudlaan 50, 3000 DR
Rotterdam, The Netherlands, E-mail: omarsili@rsm.nl
Abstract. Empirical studies have shown that the size distribution of firms can be described
as a Pareto distribution. However, these studies have focused on large firms and aggregate
statistics. Little attention has been placed on the role of technology in shaping firm size
distributions. Using a comprehensive dataset of manufacturing firms and the Community
Innovation Survey from the Netherlands, the paper investigates the relationship between
firm size and technology. It shows that technological factors shape the distribution of firm
size, suggesting that the Pareto law is not an invariant property and that technology can
constrain the “self-organising” character of industrial economies.
Key words: firm size, Gibrat’s law, innovation, Pareto distribution.
JEL Classifications: L11, L60, O33.
I. Introduction
The Pareto law is a well-known property of the size distribution of firms.
It says that the frequency of firms in a population above a certain size is
inversely proportional to the firm size. In logarithms, this relationship can
be represented graphically as a straight line.
A number of studies have tested empirically this hypothesis and for-
mulated models able to generate Pareto-like distributions (Steindl, 1965;
Ijiri and Simon, 1977). These studies have been extremely influential in
industrial economics and elsewhere. For example, there has been a renewed
interest in the Pareto distribution across many different disciplines. This
interest focuses on the properties of the Pareto distribution, as a power
law able to describe the organisation of different scientific and social sys-
tems (Krugman, 1996; Bak, 1997). If distributed according to a power law,
the structure of the industrial system would depend only on the interac-
tion between its components and not on external factors or the individual
behaviour. In addition, a similar structural form would be observed at
different levels of aggregation, such as countries and sectors. Finally, this