Economic Complex Networks

Abstract

This thesis aims to contribute to the understanding of the structure, composition, properties, functions and dynamics of economic systems. The research output results from joining advances in the scientific disciplines of systems theory, complex networks, Input-Output analysis and economic activity classification systems. The relevance of this research primarily lies in the insight and overview that arose from bridging these disciplines. The central theme in this research is the complex network of economic sectors in which the transactions between sectors determine the dynamic structure of an economic network. A sector network is an abstract aggregate of sectors. When observed in more detail, the network structure of an economic system becomes visible between sub-sectors and their constituent parts (revealing organisations and individuals at most detailed level). In this thesis, this hierarchy is also referred to as a holarchy. Sectors can be analysed from various perspectives such as activities (functions), products (services and/or goods), organisations (enterprises), occupations (crafts of individuals) and production means (tools). For instance, financing is an activity, a loan is a product, a bank is an organisation and a banker is an occupation. Sector names such as finance sector or banking sector are commonly used which illustrates the difficulties in establishing a set of non-ambiguous sector names and definitions. In daily practice the above mentioned perspectives tend to be mixed up and (overlapping) thematic aggregates are defined as sectors (e.g. creative sector, technology sector), complicating the production of statistical compilations. This mixed perspective approach ignores classification principles such as completeness, summability, exclusiveness and categorising from one perspective, based on the criterion of homogeneity. The perspective of activities or functions emphasised in this thesis, is also expressed in the design of a novel sector model and sector network model. The functional analysis of sector related activities (sourced from the UN economic activity classification system (ISIC) and a set of standards and models specific to the telecom related sector) shows that only 3% of the inventoried functions of a sector was found unique on average. This finding is in line with propositions from system theory. Prominently included in this thesis are the representation of economic systems as complex networks and the accompanying analysis of German and Dutch Input-Output data. This analysis of the monetary flows in the German and Dutch economic networks shows a full-mesh structure at sector level, thus each sector can directly share produced (unique) value with every other sector. Supported by findings from literature review, a contemporary sector network appears to consist of twenty complementary economic sectors. When descending the hierarchy below the sector level, the link density decreases, thus not all sub-sectors are directly connected. Observations from the analysis of the Input-Output data (1987-2007) show an increasing network clustering coefficient and a decreasing value of the interaction ratio of the sector network over time. Furthermore organisation fragmentation is observed from Dutch statistics about the composition of enterprises (2001-2012). Possibly these indicators directly relate to the increasing scale of outsour-cing activities and the governmental policy of privatisation and the dismantling of monopolies in order to stimulate competition. This observation and interpretation show that network analysis generates interesting questions and hypotheses, which could enrich economic science. The above mentioned economic phenomena and the recent economic crises, raise the question if the sector network can be safely exposed to market forces. This thesis pleads for a more intensive orchestration of the sector network as a whole and the functioning of our vital sectors in particular.