3. Definitions, algorithms and the dataset
In order to make an empirical test of these propositions we first need a clear-cut
definition of both business groups and industrial districts. We will begin by discussing the
definition and the algorithm used to identify business groups. Then we will discuss the
definition and algorithms used to identify industrial districts.
3.1 Business groups: definition and statistical algorithm
As already mentioned, a business group is a set of companies legally distinct but
belonging to the same owner(s). From the empirical point of view groups have been
identified through control linkages between pairs of legal units. The operational guidelines
indicated at European level for the identification of control for statistical purposes are the
followings (EUROSTAT, 2003): i) a legal unit directly owns at least 50% plus one of the
voting rights in another legal unit; ii) a legal unit owns an equal share of voting rights, with
respect to other units, in another legal unit; iii) a legal unit owns, by means of other
subsidiaries, at least 50% plus one of the voting rights in another legal unit; iv) a legal unit
fully consolidates the balance sheet of another legal unit and no other one consolidates the
same legal unit. The algorithm used to empirically identify business groups applies these
operational guidelines to the shareholdings matrix thus deriving a control matrix.
The Italian Statistical Business Register on Business Groups (Archivio sui gruppi
d’impresa) is built-up by integrating three different statistical sources: i.e., (i) the archive of
all shareholders of non-listed companies; (ii) the archive of all shareholders of listed
companies; (iii) the archive of firms’ consolidated accounts. These sources are sufficient to
guarantee a complete coverage of the shareholders’ structure of all limited companies.
Elementary information on the shareholdings coming from these three sources is
integrated and chains of direct and indirect control are re-constructed by applying an
algorithm. The resulting structures are the statistical business groups, reflecting concepts
and definitions mentioned above.
More intriguing information
1. Happiness in Eastern Europe2. Modeling industrial location decisions in U.S. counties
3. Draft of paper published in:
4. Non-causality in Bivariate Binary Panel Data
5. Aktive Klienten - Aktive Politik? (Wie) Läßt sich dauerhafte Unabhängigkeit von Sozialhilfe erreichen? Ein Literaturbericht
6. A novel selective 11b-hydroxysteroid dehydrogenase type 1 inhibitor prevents human adipogenesis
7. Strengthening civil society from the outside? Donor driven consultation and participation processes in Poverty Reduction Strategies (PRSP): the Bolivian case
8. The name is absent
9. Nurses' retention and hospital characteristics in New South Wales, CHERE Discussion Paper No 52
10. The name is absent