Peculiarities of international financial network mapping

Abstract

Over the past two decades the international financial system has undergone significant changes. Due to the development of information and communication technologies as well as innovations in financial products, under the conditions of financial liberalization, in recent years the scale, scope and speed of capital flows circulating in the international financial system have increased significantly. International financial flows during the period of international financial crises of 2007–2008 have become not only the mechanism for the capital and financial power allocation, but also the source of systemic risks. Non-trivial nature of the interdependencies between individual countries in the international financial system creates the need to investigate the cross-border financial connectedness as a complex phenomenon, incorporating the value of international financial linkages as well as their structure. Despite the growing interest in the cross-border financial connectedness issues, its conceptualization and assessment remain complicated and unresolved. While analyzing the scientific literature, unanimous international financial network identification principles were not observed. The network analysis provides possibilities to identify international financial system as a network of nodes (i.e. financial institutions or countries) and their interlinkages (i.e. financial flows, contractual relations, and etc.) that are aggregated into specific topological structure. The development of network analysis tools allowed considering not only the existance of financial interlinkages, but also their directionality as well as their value. The first stage of a typical network analysis starts with the identification of researched network. The international financial network is suggested to be identified as a network of countries and their bilateral financial flows. As the international financial linkages have both value and direction, the most appropriate model for the international financial network should consider these two features. Therefore it is suggested to apply the weighted directional network model. As the inclusion of every additional feature of network links (value or direction) makes network analysis more complex, the network symmetry is usually checked before finalizing the empirical network model. The research methodology was applied for testing the proposed network mapping in the international portfolio investment market in 2001–2011. The international portfolio network was identified while identifying its nodes as separate countries and the network linkages as bilateral international portfolio investment flows. The identified nodes and their interlinkages were united in the weighted directed international portfolio network model. The data for the research was applied from the CPIS (i.e. coordinated portfolio investment survey) database. The research resulted in 11 network graphs for each year of the researched period covering international portfolio investment positions of 78 countries. This research was performed with the help of network analysis software Netminer 4.0 . The limitations of this software were complemented with custom created software in Visual Basic for analysis of some specific network indicators in need. The research results provided the following conclusions: The analysis of the identified portfolio investment network revealed that the intensity of the researched network dramatically increased in the period of 2001–2011 not only in the value of portfolio investments, but also in the number of financial links. The network density (the number of all possible links) in 2011 reached 65,7 % in comparison with 56,7 % of links that were realized in 2001. It confirmed that not only the value of international portfolio investment is important, but also the number of links it is distributed through. The international portfolio investment network was identified as asymmetric in the number of its linkages as well as their value. It means that the incoming linkages do not match the outgoing linkages neither in their number nor in their value, even though many researchers follow the assumption that the directionality of financial linkages does not provide additional information to its structural analysis. It is worth mentioning that the international portfolio investment network is less asymmetric in the number of links (76 % of links are reciprocated) than in their value (49 %).