Aristotle University Of Thessaloniki (AUTH)
Πολύπλοκα συστήματα και δίκτυα στην οικονομία
Abstract
dc:descriptionComplex networks and systems play an important role in the economy and refer to interconnected firms, industries or countries, and the relationships and interactions among them. The present doctoral dissertation investigates empirically the Global Value Chains or GVCs as directed weighted networks with countries as nodes and value added flows as edges. This methodology allows the research and visualization of the trade interconnection of many countries involved in international trade. The examination of GVCs as networks offers insight into the structure of global trade, the relationships and patterns of value creation and trade among countries, their role in different economic sectors and their integration in GVCs. Each node represents a country. The importance of a node is related to value added inputs and outputs, but also to its direct and indirect interconnection with the other countries in the network. The edges represent value added flows among countries as a percentage of global flows of value added. These value added flows represent the transfer of intermediate goods, components and services among countries in the production process. The edges capture the connections and dependencies among countries in terms of their contribution to value creation and their reliance on inputs/outputs from other countries. Edges have direction and weight. The direction indicates whether the flow of value added is imported or exported. The weight represents the share of value added flow among countries, providing a measure of the strength of the traded value added. The analysis of GVCs as networks allows the examination of various network properties such as centrality, clustering, connectivity, and resilience. It helps to identify key nodes that play a significant role in the network, to understand the patterns of value-added flows and to assess the vulnerability or resilience of the network structure. The study of GVCs as networks lets policymakers, such as politicians, economists, researchers, to understand the dynamics of global trade, identify opportunities for improvement and diversification, the effects of a slowdown or a shock, and to design policies aimed at enhancing relative position and competitiveness within GVCs.The empirical research focuses on the examination of GVCs using Leontief multipliers, entropy measures, and specific statistical network analysis indicators. Leontief multipliers, also known as input-output multipliers, are used to assess the interdependencies among countries in a sectoral network. Entropy is used as a measure of the trade flows diversification of each country. The statistical network analysis utilizes eigen centrality to assess the influence of each country, degree centrality to evaluate the relative position of each country in GVC networks, and the degree centralization to measure the concentration of trade flows within GVC networks.Applying the above mentioned measures, it was investigated the competitive dynamics within GVCs over a period of fifteen years (2000-2014), the structure of the sectoral GVCs networks, the dominant countries, and their level of trade diversification. Employing Granger Causality, it was examined the direction of causality between a country’s relative position in GVCs and its import/export diversification. Moreover, it was studied the transmission of economic shocks through the GVCs and the impact of a shock such as the “trade war” between the US and China. Leontief multipliers were used to estimate the impact of changes in demand on the output of each country in each sectoral GVC network due to a shock. To explore the relationship between the multiplier and each country’ s participation in GVCs, a panel regression analysis was conducted. The multiplier was set as the independent variable and the relative position of each country as importer, as exporter, imports diversification and exports diversification were considered as the dependent variables. Finally, it was performed a sectoral study of the GVC networks involved in the production of personal protective equipment, the demand of which increased excessively due to the pandemic “COVID-19”, namely chemicals, rubber/plastics, and textiles. In the sectoral networks, countries with higher share of value added exports than imports or “value-added surplus” were identified. Their flow diversification was assessed using entropy measures, and their egonets were analyzed. The empirical findings highlight the important but different role of specific economies in the GVC networks, namely China, the US and Germany. China participates in GVCs as the “global factory”, the US as the “brand creator” and Germany as the “trade centre” of the GVC networks. Countries with high degree centrality and therefore high dominance in GVC networks also present high eigen centrality and therefore increased influence. Consequently, a slowdown or a shock in these economies is expected to have a great impact on the GVC networks. In the context of GVC integration, a country seeking to establish or enhance its involvement in GVCs should adopt a strategy of cooperation with a few in non-polarized GVC networks. In addition, a country with a dominant position in a sectoral GVC network does not necessarily have high trade diversification, but when it has it is difficult to lose its dominance. Developing economies that adopt diversification strategies could improve their dominance in sectoral GVC networks.The Granger causality between a country’s position in GVCs and its trade diversification depends on the assumption of homogeneity or heterogeneity across sectors and across countries. The hypothesis of homogeneity across sectors and across countries suggests that import diversification causes the relative position of a country as an importer in the GVCs. The hypothesis of homogeneity across sectors and heterogeneity across countries leads to mixed results and weak evidence of causality from export diversification to relative position as an exporter in the GVCs. The hypothesis of heterogeneity across sectors and heterogeneity across countries suggests that relative position as importer/ exporter causes trade diversification of imports/exports in some cases, while in other cases a bi-directional causality.The cross country and sector panel regression analysis indicate a positive statistically significant relationship between multiplier with relative position as importer and export diversification and a negative relationship between multiplier with relative position as exporter and import diversification. Consequently, during periods of protectionism a country is expected to reduce its imports flows and/or shorten its customers’ network and/or increase its export flows and/or expand its suppliers’ network. Our empirical findings from the network analysis combined with the multiplier values suggest that the “trade war” will be less harmful for the US than for China. The US has a larger surplus of value-added flows than China and achieves much better diversification of its imports and exports, which is reflected in its smaller average multiplier than China’ s. In addition, the US imposed high import tariffs in sectors where China was its largest value added exporter. The inverse is not applicable, suggesting that the US designed the tariff policy to directly affect Chinese exports to the US and produce a significant adverse impact on the Chinese economy (due to higher multipliers). However, the evolution of the degree centrality values suggests that the US participation in the GVCs may have reached at a maturity stage, while China’s participation accelerates.The sectoral analysis of GVC networks involved in the production of personal protective equipment revealed that the US has the highest value added surplus in “chemicals”, Japan in “rubber and plastics” and China in “textiles”. The US has the most diversified flows in “chemicals”, Germany in “rubber and plastics” and Italy in “textiles”. Finally, the egonets analysis revealed that the USA is a key supplier in all the three sectoral GVC networks.
Degree
thesis:*- Grantor dc:publisher
- Aristotle University Of Thessaloniki (AUTH)
- Year dc:date
- 2025
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Angelidis, Georgios
- Αγγελίδης, Γεώργιος
Subjects
dc:subject × 14Rights
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/60561