The Architecture of Globalization: A Network Approach to International Economic Integration. Raja Kali and Javier Reyes1 Department of Economics Sam M. Walton College of Business University of Arkansas Fayetteville, AR 72701 Abstract We combine data on international trade linkages with network methods to examine the global trading system as an interdependent complex network. We map the topology of the international trade network and suggest new network based measures of international economic integration, at both a global system-wide level and a local country-level. We develop network based measures that incorporate not only the volume of trade but also the influence that a country has on the international trading system. These measures incorporate the structure and function of the network and may provide a more meaningful approach to globalization than current measures based on trade volumes. We find that in terms of participation and influence in the network, global trade is hierarchical with a core-periphery structure at higher levels of trade, though integration of smaller countries into the network increased considerably over the 1990’s. The network is strongly “balkanized” according to geography of trading partners but not as strongly by income or legal origin. Using these new measures we find that a country’s position in the network has substantial implications for economic growth. We therefore suggest that a network approach to international economic integration has potential for useful applications in international business, finance and development. Keywords: globalization, economic integration, networks, international trade 1 Email: rkali@walton.uark.edu, jreyes@walton.uark.edu We are grateful to Jon Johnson, Fabio Mendez and Anand Swamy for helpful discussions. We thank seminar participants at Harvard Business School, the University of Arkansas, Williams College, IGIDR-Mumbai and MWIEG Fall 2005 for their comments. Viktoria Riiman provided outstanding reseach assistance. I. Introduction. While popular usage of the term “globalization” provokes strong and polarizing opinions across the world, such sentiments are usually associated with the effects, real or perceived, of what economists refer to as international economic integration. The increase in international economic integration that has characterized the last half-century has been associated with the spectacular economic performance and move out of poverty for large parts of the world (Sachs and Warner, 1995), but also with the increase in the volatility of country-level performance, reflected in several recent episodes of economic and financial “crises” (Forbes 2001). There is also a growing perception that the process of globalization has accelerated over the last decade and that the benefits and costs of increasing economic integration have not been evenly distributed across the world (Stiglitz, 2002; Bhagwati, 2004). Despite a sharp increase in interest on these issues, discussions are often handicapped by the dearth of appropriate measures of international economic integration. Most studies of international economic integration or globalization in the economics literature focus on the volume of trade (exports and/or imports as a fraction of total trade) between countries, or define “trade integration” as the sum of exports and imports divided by GDP (see for example Rodrik, 2000, IMF World Economic Outlook, 2002). While these indicators2 have been useful, the literature recognizes their shortcomings (which we describe in more detail below). Nevertheless, they are still widely used for studying international economic integration, primarily for lack of better alternatives. Recent advances in the study of networks (Albert and Barabasi, 2002; Newman, 2003) have placed elegant and powerful tools at our disposal, enabling us to suggest alternative measures of international economic integration (henceforth IEI) that turn from a sole focus on individual country trade levels to a consideration of the pattern of linkages that tie together countries around the world as a whole. In this paper we combine a network approach with data on international trade linkages in order to examine the global trading system as an interdependent complex network3. A network approach enables us to derive statistics that describe the structure and evolution of global trade in ways that existing measures do not capture, such as the number of actual and potential trading partners, the structure of regional trading and the influence of individual countries and groups of countries for the whole network and for specific regions. We use this change in perspective toward IEI to suggest new measures of integration that provide insights into global trade that have been overlooked by the literature. 2 Other measures based on volumes such as gross private flows to GDP, and total trade to merchandise value added also fall into this category. 3 Complex networks are large scale graphs that are composed of so many nodes and links that they cannot be meaningfully visualized and analyzed using standard graph theory. Recent advances in network research now enable us to analyze such graphs in terms of their statistical properties. Albert and Barabasi (2002) and Newman (2003) are excellent surveys of these methods. 2 With this objective, we first map the topology of the international trade network with a view to understanding its structure and properties. Armed with such an understanding, we then suggest new measures of IEI, at both a “local”, country-level, and a “global”, system-wide level, that incorporate the structure and function of the network. We use these measures to parse IEI along a number of different lines: geography, income and legal origin. This enables an examination of whether global trade has become more integrated or “balkanized” along these dimensions. We suggest network-based measures that capture not only the volume of trade but also the “influence” that a country may have on the international trading system. We have data on the network of international trade linkages at two points in time, 1992 and 1998, and are able to construct these measures for both years and examine how the network and thus “globalization” has evolved over the 1990’s. Since trade levels vary considerably from country to country and there could be some debate over what constitutes “consequential” levels of trade, we construct the network for different trade level thresholds4. We find that at low levels of trade, the global trading network has become much more integrated, while at higher levels of trade it has not changed much. At low levels of trade, the global trade network is quite decentralized and homogenous but at higher levels of trade the network looks much more hierarchical and heterogeneous, with a core-periphery structure. We also find that there is a high level of multilateralism in global trade and this has not changed much between 1992 and 19985. As an application, and to demonstrate the potential of the network approach to IEI, we use our measures of network importance in a cross-country growth regression and find they are all statistically and economically significant, have the expected signs and raise the explanatory power of the regression above that obtained using only volume based measures current in the literature. Using one of our measures of local integration, degree centrality, a measure of how centrally located a country is in the network6, we find that an improvement in the centrality ranking by 10 units at the two percent trade-link threshold increases the average growth rate of per capita GDP by 1.1 percentage points.7 A country’s position in the network can thus have substantial implications for development outcomes. The paper is organized as follows. Section II describes the data and definitions that we use to organize the trade-link data. Section III applies concepts from network analysis to understand properties of 4 We describe this procedure in more detail in section II. While we believe this is the first exercise to explicitly chart the topology of the international trade network and suggest the use of this topology for the understanding of economic integration, we are by no means the first to use network ideas in international business and economics. An excellent introduction to this literature is Rauch and Casella (2001) and the critique by Zuckerman (2003). Systems- or network-based measures of globalization have, to the best of our knowledge, not been used in economics before, but there is antecedent in the sociology literature. A paper by Smith and White (1992) uses international trade flow data to consider the change in the structure of the international division of labor with the goal of understanding patterns and cycles of hegemony in the world-system. The focus of this work is thus quite different from ours. 6 We describe various network measures in more detail below. 7 This is judged to be a substantial effect by the standards of the literature. For example, Yanikkaya (2003) finds that an increase of 10% in the total trade to GDP ratio would increase the average growth rate of per capita GDP by 0.18%. 5 3 the network. We first provide an overview of the topology of the network and then delve deeper into the data and propose measures of local and global economic integration. Section IV is our application to economic growth. Section V summarizes our findings and suggests further applications of these measures. II. Definitions and Data. The first step in our approach is to identify the fundamental building blocks of the network and their specific properties. A network is a set of points, called nodes or vertices, with connections between them, called links or edges. In our context, each country is considered to be a node of the network. Since international trade is usually measured using the monetary value of exports and imports between countries, trading relationships are analogous to valued links in a network, and these vary from country to country. In order to chart the structure of the network we wish to take into account the magnitude of these relationships but not specifically their exact value. We do this by considering a network link between two countries to be present if the trade level between them is above a certain threshold. Specifically, we define a trade-link between country i and country j to be present if the value of exports from country i to country j as a proportion of country i’s total exports is greater than or equal to a given magnitude. Since exports of country i to country j are in effect imports of j from i we are able to construct both export and import networks in order to understand IEI from both sides. Moreover, since trade levels vary considerably from country to country and there could be some debate over what constitutes “consequential” levels of trade, we construct the network for different trade level thresholds, which we explain below. Examining how the structure of the network changes as the trade threshold used to define the presence of links varies also enables us to understand the sensitivity of various topological characteristics of the network to differing trade magnitudes. Constructing the network for different thresholds enables us to incorporate both magnitudes and network features into our analysis. Using thresholds enables us to avoid working directly with valued-directed links even though implicitly these thresholds embody the values of the trade links in our data. The data used for our international trade network was extracted from the COMTRADE Database of the United Nations8. We use the US dollar value of exports and imports of all commodities between 189 countries for 1992 and between 192 countries in 19989. Countries are the nodes of the network and a link 8 United Nations database STIC 1. A list of countries is included in Table 1A of the Data Appendix. It should be noted that even though our trade network is extensive, it is not all-inclusive. The United Nations database includes more than 230 countries/areas, plus some NES (not elsewhere specified) areas. Although we compute the total exports and total imports from the all-inclusive raw database, in our trade network analysis we only include countries. In other words we ignore regions and NES figures. Additionally some countries, like Guadeloupe, Martinique, Reunion, and others, are excluded from our analysis because there are some inconsistencies in the data reported by these countries. 9 4 between them represents trading relationships among these countries. We study import and export relations separately and therefore we have a directed graph where country A can export to country B without having country B exporting to country A. Initially we organize the data in matrix form, letting columns represent exporting countries and rows denote importing countries, and we analyze the flow of payments instead of the flow of goods. This means that exporting countries will be recipients of payments for their exports, while importing countries will be sources of payments for their imports. This methodology allows us to analyze the influence of importing countries on exporting ones as influential buyers. We use the share of exports of country A to country B out of the total exports of country A and construct binary matrices for different magnitudes of trade. If country A’s exports to country B, out of the total exports of country A are greater or equal to a given threshold, then the link B→A is present10. Our primary trade-link definition thus measures export dependency. As an illustration, Table 1 is the binary matrix for the first 10 countries in our sample when we use the exports dependency ratio described and a trade-link threshold of 0% for 1992. For example, the link between Algeria (source of payments) and Albania (recipient of payments) exists and the cell entry (source = Algeria, Receiver = Albania) is 1, denoting that imports of Algeria from country Albania are greater than zero11. [Insert Table 1 here] It is also important to note that the number of countries across the two years considered is not constant. There are a total of 194 countries included in the analysis but only 189 existed in 1992 and 192 in 1998. The 1992 list excludes the Czech Republic, Eritrea, and Ethiopia, while the 1998 excludes Czechoslovakia and Former Ethiopia. All the network indicators computed take this into consideration. Alternatively, and as a robustness check, we let the columns of the trade matrix denote importing countries and rows represent exporting ones and we analyze the flow of goods. In this case, we use the share of imports of country A from country B out of the total imports of country A, which is a measure of import dependency of country A on country B, to construct binary matrices for different magnitudes of trade. This approach allows us to analyze the influence of exporting countries as influential suppliers. We find that the results for both dependency measures are very similar, and in the following sections focus on the results obtained using export dependency. We expand on the results obtained with the alternative import dependency measure to define trade-links in Section IV, which applies the network indicators to economic growth. 10 11 The directed edge goes from B to A because B is the source of payment and A is the recipient of this payment. Note that this also means that the exports from Albania to Algeria are also greater than zero. 5 III. A Network Approach to Measuring International Economic Integration. III.I Network Overview Just as nodes and links are the basic components of any network, node degree is the basic component of complex network analysis. The degree is the number of links connected to a given node. For directed networks we have two different measures, in-degree and out-degree. The first one deals with inbound links, in other words how many times a specific node acts as a receiver. The second one deals with the outbound links, counting how many times a specific node acts as a source. These two measurements provide an initial overview of network structure. We can locate highly connected nodes, referred to sometimes as hubs, and by looking at in- and out- degree measures separately, identify potentially influential receiver and source countries. It is also possible to obtain an overall idea of how homogeneous the network is. In a homogeneous network, flows are not dominated by a small group of nodes, implying that there should be no dominant nodes. Since our data are on the dollar value of exports and imports, and we initially construct the network using exporting countries as recipients of payments for their exports (in-degree) and importing countries as sources of payments for their imports (out-degree), henceforth we use the more intuitive terms export-degree and import-degree instead of in-degree and out-degree respectively. Using our binary matrix representation of the network [as in Table 1], the export-degree of country i is calculated by summing up the links that are present in column i, while the import-degree for country j corresponds to the summation of the links present in row j. We construct the network and associated network measures for several different values of the tradelink threshold12. The zero percent threshold indicates the mere existence of trade among two countries and in this sense it is the least restrictive threshold. It simply acknowledges the presence of positive trade. We choose the one and two percent thresholds because eighty three percent of the trade shares in 1992 (eighty seven percent in 1998) are between zero and one percent, and this number increases to eighty nine percent when the range between zero and two percent is considered for the 1992 data (and to ninety two in 1998)13. These thresholds embody higher levels of trade. The export and import degree of country i denote the number 12 The export and import degree results for all countries in the 1992 and 1998 trade networks at the 0, 0.5, 1 and 2 percent thresholds are not reported here for matters of space but are available upon request. 13 The reader should be alerted to the possibility that if trade flows over existing links increase substantially over time the current approach could be problematic. Consider the case where the number of links is constant across time but the flow has increased in such a way that the number of links that meet the threshold of two percent remains constant but a large portion of these move from a range of two percent to a level of four percent. In this case our measures would not capture these changes in trade flows. The percentages discussed in the text show that this is not the case as the number (percentage) of links below one and two percent increase between 1992 and 1998. This suggests that in the data, the number of trade shares above one and two percent are falling even though volume is rising. And this trend holds even for higher thresholds, like five and ten percent. 6 of trading partners to which i exports to (i.e., i depends on for revenues) and from which i imports from (i.e., depend on i for revenues) respectively, that are active in the binary trade matrix for any given threshold14. As the trade-link threshold is increased, the export and import degree distributions change, providing insight into the structure of international trade. For relatively low levels of trade the degree distribution for both imports and exports are similar. Most of the nodes have a relatively high export and import degree which means that most of the countries have a large number of trade partners for both exports and imports. But as the trade-link threshold is increased, the distribution of export-degree and import-degree changes dramatically. The export-degree (number of countries exported to) falls considerably for all the nodes while the import-degree (number of countries importing from) remains constant only for a very small group of nodes (the G-7 appear in this group) and falls substantially for the others. The export-degree change tells us that at higher levels, all the countries (G-7 included) export to a relatively small number of partners – which turns out to be by and large the same set of countries, the G-7 plus Spain, Belgium and the Netherlands. These countries account for almost fifty percent of world imports. The interpretation for the change in import degree is that for higher levels of trade, a small block of influential countries import from most of the other 179 - 182 countries, while the rest only import from a small number of countries. The asymmetric change of the import and export degree distributions imply that from the imports (source of payments) perspective the network is quite skewed, but from the exports (receiver of payments) perspective it is quite evenly distributed. The mean of the export-degree distribution is 9 countries at the 2%, for 1992 and 1998, trade-link threshold and 13 and 15 countries at the 1% threshold in 1992 and 1998, respectively. This pattern of inequality in the degree distribution can be visualized by computing Lorenz curves and Gini coefficients. Figure 1 presents the Lorenz curves plots for the 1% and 2% thresholds and the Gini coefficients derived from the deviation of the forty five degree line from each of the Lorenz curves. These plots and numbers reveal that the 37 most connected countries (20% of the total countries) account for almost 80% of the outbound links in 1992 and 75% in 1998 at the 1% trade-link threshold. These numbers are almost completely reversed for the inbound links, where the 37 (20% of the total countries) most connected countries 14 From an import degree perspective the maximum degree for all threshold levels is equal to the number of countries included in the analysis minus one. For the export degree, this still holds for the zero and 0.5 percent threshold. But given the criteria used to determine the presence of a link, the maximum export degree changes as the threshold increases. For the 1 percent threshold the maximum export degree is 100 since the percentage of exports to all trading partners cannot add up to be more than 100. Similarly logic explains why at the 2 percent threshold the upper bound for export degree is 50. For the 0.5 percent threshold the upper bound is 200, but this is not an issue for the analysis since it is more than the number of trading partners in our dataset (189 in 1992 and 192 in 1998). 7 account for only 30% of all inbound links, in 1992 and 199815. Similar results are obtained from the analysis of the 2% trade-link threshold. The 80/20 finding has special significance in the study of networks as it reflects the existence of a Pareto distribution, as opposed to a random network where the distribution of node degree is random. This kind of distribution is also often referred to as a power-law (exponential) distribution as the number of nodes with degree k, N(k) follows a power law, i.e., N(k)~k-γ where γ is the degree exponent. Power laws mathematically formulate the fact that in many networks the majority of nodes have only a few links and that these nodes coexist with a few big hubs, nodes with an anomalously high number of links. In contrast, for a random network, the peak of the distribution implies that the majority of nodes have the same number of links. Therefore a random network has a characteristic scale in its node connectivity, embodied in the average node and fixed by the peak of the degree distribution. In contrast, the absence of a peak in a power-law distribution implies that there is no such thing as characteristic node. In other words, there is no intrinsic scale in a powerlaw network. Such networks are therefore referred to as being scale-free16. The international trade network is thus scale-free at higher levels of trade. This is especially interesting as it implies that it does not make much sense to speak of a “typical” country in terms of the number of trading partners. III.2. Measures of Integration We now introduce more detailed measures of global and local integration. A. Global Integration Measures Centrality Degree analysis suggests that the international trade network has a core-periphery configuration from an imports perspective with the industrialized countries as the center of gravity. In this type of system the countries at the core are the most influential nodes since shocks to the core will affect the whole network. Another way to examine this feature of the network is through the notion of centrality. In many complex networks, centrality is used as a measure of power and influence. According to Wasserman and Faust (1994), central actors (nodes) must be the most active because they have the most ties to other actors (nodes). For our trade network, we can compute node centrality and network centrality. Node centrality measures how central a given node is with respect to the others while network centrality measures how centralized the network is with respect to a perfectly centralized network. Here we present the results on network centrality; we address individual node centrality in the section on local 15 Perfect equality, in this case perfect symmetry, would correspond to 37 countries (around twenty percent) accounting for twenty percent of the in or outbound links. 16 A startling discovery from recent research on complex networks is that almost all complex networks in nature are scale-free (see Albert and Barabasi, 2002). 8 measures of integration. We focus here and in the local measures section only on import-degree centrality indices. There are two reasons for this. First, we already know from the node degree analysis that the exportdegree distribution is very homogenous. Therefore not much information would be added by analyzing the differences in the export-degree centrality. Second, and more importantly, we are interested in understanding which countries are influential importing countries in the international trade network. In order to analyze the centrality of the international trade network from an imports perspective, we compare it to a perfectly centralized network of the same size. A perfectly centralized network is one in which only one node sends/receives to/from the other vertices. This is called a star network (the most unequal possible network)17. Freeman (1979) proposes the following expression as a centralization index: ∑ [C = max ∑ [C g CI d max i =1 − C D ( ni ) g i =1 max ] − C D ( ni ) ] (2) d and Cmax represent the actual maximum degree centrality observed in the data for an where C max individual node and the theoretical maximum degree centrality for an individual node in a network with g countries, and CD(ni) denotes the degree centrality of node i18 . The denominator in expression (2) is the summation for the star network, and equals (g-1)(g-2) where g denotes the number of nodes in the network. The degree centrality of an individual node can be simply represented by its degree d (ni ) but a more standard way is to normalize the individual node centrality in the following fashion, C D ( ni ) = d ( ni ) g −1 (2.1) The Centralization Index, C I , thus measures the degree of variability in the degrees of nodes in the network as a percentage of that in the star network of the same size. The way in which the star-like configuration of the import-degree international trade network evolves between 1992 and 1998 provides information regarding the proportion of countries that have moved toward or away from the center of gravity. With the increasing volume of international trade observed during the nineties and the opening of countries like China and former Soviet-bloc countries, it is conceivable that the international trade network has been becoming less of a star-like network. This would result in a lower level of influence for the G-7 countries and the emergence of a number of other influential countries that previously belonged to the periphery. 17 In a star network, all nodes but one have an export/import degree of one except for the central node which has an export/import degree equal to the number of nodes in the network minus one. 18 A more in depth discussion of degree centrality for an individual node is in the section of Local Measures of Integration. Note also that there are two other measures of centrality, Betweenness (Freeman 1977) and Closeness (Sabidussi, 1966). We use the current measure on account of its simplicity as compared to the others and because it seems better suited to the notion of the “core” of a network in the context of international trade than the others. 9 Table 2 presents the results for the import-degree network centralization index19. This index shows us that for the lowest trade-link threshold of 0%, the network centralization index for import-degree is around 56% for the period of 1992 and 42% for 1998. In other words, the network is not very centralized. As we move to higher thresholds, such as 0.5%, 1%, and 2%, we observe dramatic changes that are in line with those obtained from the node degree distribution analysis. As the threshold increases, the imports network becomes extremely centralized. That is, a small group of countries are destinations for the bulk of imports which come from (i.e., are exports from) a large number of countries in the network. We could refer the former group of countries as the core and the latter group as the periphery. The comparisons of these indices across time imply that the core-periphery structure did not change noticeably over the nineties and a relatively small number of countries still constitute the core of the network, a core that is likely to exercise an enormous amount of influence on the periphery. We provide specific measures of influence later in the section. Network Density Node degree and centrality analyses are useful because they allow us to identify the presence or absence of a center of gravity for the network and give us an overview of the structure and configuration of the network as a whole. But these indicators do not directly address how integrated the network as a whole really is. One way to start examining the extent of global integration of the network is to measure the proportion of all possible links (trading relationships) that are actually present in the network. This ratio is called network density. The maximum number of edges for a network is determined by E max = E D max g ( g − 1) 2 = g ( g − 1) where Emax and E D max denote the maximum number of edges/links for an undirected and a directed graph, respectively and g is the number of nodes. The density of a directed network, like the international trade network, is simply the ratio of the links actually present to the maximum possible, E D max . ∆D = L g ( g − 1) (3) where L stands for the number of links present in the network. The density calculations for the international trade network are also presented in Table 2. It is important to keep in mind that as the threshold increases, the maximum number of potential links decreases. For example, when the threshold used is one percent, the maximum number of countries to which a given 19 Calculated using UCINET software package, specifically Freeman’s degree centrality measures routine. 10 D country can export is one hundred, therefore the maximum number of possible links, E max , is determined by g⋅100 instead of g⋅ (g-1). The results in Table 2 show, as expected given the previous results regarding a core/periphery structure, that network density drops as the threshold is increased. The results across years allow us to compare changes in economic integration. For the period of 1992 – 1998, network density increased by 35% at the 0% threshold and 15% at the 0.5 % threshold. At higher thresholds, density increased as well but by a smaller margin; 9% and 4% respectively for the 1% and 2% thresholds. Once again, the implication is that international economic integration measured this way increased much more at lower levels of trade. Clustering The 1990’s have been a booming era for international trade agreements like the NAFTA, MERCOSUR, and the EU. In light of these preferential trade arrangements, an interesting question is the extent to which trading partners of a particular country are also linked to each other. This corresponds to the analysis of the proportion of multilateral trade relationships relative to bilateral ones. In a more globalized world the share of multilateral relations relative to bilateral ones should be higher than in a more balkanized world. In terms of network topology, the extent of multilateralism can been seen through the property of network transitivity, sometimes also called clustering. In many networks it is found that if node A is connected to node B and node B to node C, then there is a heightened probability that node A will also be connected to node C. Clustering thus measures the probability that “the partner of my partner is also my partner” and provides insight into what is referred to as the neighborhood structure of the network. Transitivity in network topology means the presence of a heightened number of triangles in the network – sets of three nodes each of which is connected to each of the others. This can be quantified by defining a clustering coefficient, C, (Watts and Strogatz, 1998), which is the mean probability that two neighbors of a given node are also neighbors of each other and can be expressed as the proportion of triples that form a triangle out of all the triples present in the network20. 3 × Number of triangles C= 0 ≤ C ≤1 number of connected triples 20 For example a complete triple (triangle) would be A→B, A→C and B→C and/or C→B, and connected triple can be just A→B, A→C. The factor of 3 accounts for the fact that each triangle contributes to three triples and ensures that 0≤C≤1. See Newman (2003). 11 where a “connected triple” means a single node with links running to an unordered pair of others. In effect, C measures the fraction of triples that have their third link filled in to complete the triangle. In terms of the international trade network, C is the mean probability that two countries that are linked to the same third country are also linked to each other. Note that since our trade-link definition is directional, C is computed on the basis of these directional links. Thus a triangle with links A to B, B to C, and C to A is different from a triangle with A to C, B to A and B to C. The results for the international trade network presented in Table 2 show that the clustering coefficient is 0.41 at the 2% threshold for the 1998 network and is very high at all thresholds. Moreover, the clustering coefficient has remained practically constant between 1992 and 1998. The implication of this is that both the number of complete triangles and triples increased proportionally. This suggests that the extent of multilateralism has remained fairly high across the time period of our data. Assortative Mixing We examine country specific characteristics to investigate the existence of trade patterns driven by similarities between countries. In network terminology, the presence of such patterns is referred to as assortative mixing and community structure (Newman, 2003). If countries that share similar characteristics trade more between themselves than with countries that do not, then it can be concluded that the international trade network is an assortative network and that there is a definite pattern of preferential attachment. The specific characteristics that we use to partition the data are income level, region and legal origin21. Such patterns seem particularly relevant given current globalization debates and allow us to view IEI from a number of different angles. For example, if high income countries trade with other high income countries twice as much today relative to previous years, and less with low income countries, we could say that the network as a whole is becoming more “balkanized” rather than more “globalized” along the income dimension. If more trade occurs between instead of within groups, then this could be considered evidence of a more economically integrated system. While the rationale for examining assortativity in the data along income and geographical region are fairly obvious, the rationale for using legal origin is the idea, emphasized by Rodrik (2000) and others, that transaction costs associated with contractual enforcement owing to differences in legal systems can be a major impediment to trade. Legal origin (La Porta et al. 1998, Shleifer and Glaeser, 2002) has been found to exert an important impact on many developmental outcomes. Newman (2003) shows that assortative mixing can be quantified by the following assortativity coefficient, 21 The countries are grouped according to the World Bank classification of income, the WTO classification for Regions and Legal Origin. 12 ∑ e −∑ a b r= 1− ∑ a b i ii i = i i i Tr (e) − e 2 1 − e2 i i (4) where e is the matrix containing the elements eij, which is defined to be the fraction of links in a network that connect a vertex of type i (i.e. region 1) to one of type j (i.e. region 2), e means the sum of all elements of the matrix e. ai and bi are the fraction of each type of end of a link that is attached to nodes of type i. If r = 0, then we conclude that there is no assortative mixing. If r =1, the network is said to be perfectly assortative, and if the network is disassortative then r is negative and its value is determined by, rmin = − ∑ab 1− ∑ a b i i i i , i i which will generally lie in the range of − 1 ≤ r < 0 . [Table 3 here] The results for the assortativity coefficient obtained for the variously partitioned data sets, presented in Table 3, show evidence of relatively high assortativity in international trade from a regional perspective. For this partition of the network the assortativity coefficient, r, is positive and has increased over time, between 1992 and 1998. This implies that over the nineties trading relationships have been predominantly established or strengthened between countries of the same region. In particular, a closer look at the data shows that the trading relationships within African countries and within CES countries increased significantly, as well as the trading activities of these two groups with the rest of the regions22. Additionally, Table 3 also presents the assortativity coefficients based on income and on legal origin partitions of the network. These numbers suggest that preferential attachment within countries of the same group, but the degree of assortativity is not as strong as in the case of regional partitions and the mixing patterns have not changed significantly during the nineties. Degree Correlation Assortative mixing on the basis of a scalar characteristic such as node degree is known as degree correlation. This measure determines whether there is preferential attachment between high-degree nodes and low-degree nodes, or if there is preferential attachment between low and high degree nodes, referred to as disassortative mixing. Newman (2003) shows that it is possible to compute the degree correlation coefficient simply by calculating the Pearson correlation coefficient of the degrees at either ends of a link. This 22 Density changes are not presented here for reasons of space, but are available upon request from the authors. 13 calculation should give a positive number for assortatively mixed networks and negative for disassortative ones. The results for the degree correlation coefficient, presented in Table 2, show that the international trade network is a disassortative network. High degree countries trade with low degree countries, and vice versa. In other words, countries with lots of trading partners trade with countries with few trade partners. This could be interpreted as yet another manifestation of the core-periphery structure of the global trade network. It is worth noting that these results should not be interpreted as a contradiction of our previous assortative mixing results. In this case there are no groupings of nodes according to some specific attribute. Degree correlation only records the node degree (number of trading partners) at both ends of each link and then calculates the correlation between both series. The disassortative mixing result should thus not be surprising from an economics perspective. International trade relations are not determined by the number of trading partners that each country has. They are based on structural or natural characteristics like natural resources and cultural, social, or geographical attributes that lead to comparative advantage. B. Local Integration Measures International trade to GDP ratios and individual country shares of international trade out of total world trade are two indicators that have frequently been used as measures of a country’s degree of openness. These measures do not take into consideration important features implicit in international trade linkages, like the number and importance of trading partners and the specific configuration of the international trade network. By not doing so they over or underestimate a country’s degree of economic integration and cannot be used to make arguments about the influence that a given country can exercise on others. Recent advances in complex network analysis offer a variety of tools that can be used to measure the degree of economic integration at the individual country level. Node Degree Centrality The number of in and out-bound links will ultimately determine the connectivity of an individual node, but there are different ways in which this connectivity can be measured. The simplest of these measures is Node Degree Centrality. Equation (2.1) shows how it is possible to calculate an index for node degree centrality. This index can show which countries are at the core or close to the core of the network. If a country is at the core of the network then its node degree centrality will be close to one. For a periphery country, this number will be close to zero, given that the number of international trade linkages is relatively small. 14 In Table A1, included in the data appendix, we report the import-degree centrality indices for the 0, 1 and 2 % thresholds for the years of 1992 and 1998 for all the countries in our sample23. Higher numbers indicate more central countries. For the same reasons as explained in the overall network centrality discussion, we present only import-degree centrality indices. As expected, the industrialized economies are part of the core of the network from an imports perspective, ranking in the top 20 for the different thresholds and periods considered. These numbers corroborate the finding that the centrality of the network has not changed significantly over the nineties since very few countries have dramatically increased their centrality indices. In essence, when the top twenty five countries from the 1992 data are compared with the top twenty five of 1998, very few changes are observed. Countries such as Brazil, South Korea, Indonesia, Malaysia, Mexico, the Russian Federation, Thailand, and Turkey are among the top thirty (some in the top fifteen) most central countries in the international trade network. This is especially noteworthy since these countries have been at the epicenter of several financial, currency and balance of payments crises and contagion episodes of the nineties. This is suggestive of the importance of international trade linkages for financial contagion (Forbes, 2001; Abeysinghe and Forbes, 2002). For comparison across methodologies, Table A1 in the data appendix presents the share of total international trade (imports plus exports) out of total world trade and the ratio of total trade to GDP for all the countries considered. In the interests of brevity we do not present country rankings according to these indices, but there are significant differences when country rankings obtained with these indicators are compared with those that are obtained when we rank countries according to node degree centrality. Node Influence or Importance Node degree centrality provides a preliminary approach to the identification of influential nodes. It is based on the number of countries that can be reached through direct links by an individual country. But it misses important features of the international trade network. The number of trading partners is a relevant statistic, but the specific characteristics of these trading partners may amplify or dampen the influence that a specific country has on others and on the whole network. One could say that it is not only the quantity of your partners that matter for influence, but also how influential they are in turn. If country A trades with country B and B trades with fifty other countries, then A exerts indirect influence on these fifty countries. In a prominent paper, Salancik (1986) argues that “Accurate assessments of the structural power of several interdependent parties are hampered by the fact that parties depend on one another indirectly as well as directly and that any one’s dependencies are not equally important for all parties.” He goes on to propose an index for dependency networks in which nodes are defined as more important if others nodes depend more 23 Calculated using UCINET software. 15 on them and if the other nodes depending on them are themselves important. Applying his index to our context, the importance of country i is a function of the dependence of other nodes on i and the importance of these other nodes. imp (i ) = ∑ dep (ij ) imp j + iv( i ) for all j ≠ i (6) j where imp ( i ) is the importance of country i , dep ( ij ) is the extent to which country i is depended upon by country j, and iv(i) denotes the intrinsic value of country i. Equation (6), which represents a system of i equations, determines that if a country is not depended upon by other countries, then this country will be unimportant. Also, if a country is depended upon only by unimportant countries, then it would also be considered unimportant. For the intrinsic value of country i we consider three alternatives, no specific value (Intrinsic Value, IV=1), the share of total trade of country i (exports plus imports) out of total world trade (with Intrinsic Value =Trade Share) and the ratio of the GDP per capita of country i with respect to that of the US (with Intrinsic Value =GDP ratio).24 Equation (6) can be rewritten in matrix form as follows, IMPi = [D]ij * IMPi + IVi (7) where [D]ij denotes the matrix of dependencies of each country j on each country i. For the international trade network exporting countries depend on the importing ones. Therefore the elements of [D]ij are the share of exports of country j to country i out of the total exports of country j. This is essentially the same matrix that has been used in the calculation of all the measures reported so far, but in this case there is no need for the threshold analysis. By solving the system of equations, denoted by equation (7), it is possible to determine the importance of an individual country relative to the 194 other countries included in the study 25. The importance indices thus computed take into consideration volumes of trade and the number and importance of all trading partners. [Table 4 here] Table 4 shows the results for the top thirty countries, according to importance in 1998, but the indices for all one hundred and ninety four countries are included in Table A1, located in the data appendix. Importance index measures for the three different approaches to “intrinsic” value of a country described 24 These trade shares were calculated using the same trade data used for the network indicators and for the countries where the GDP per capita was not available the intrinsic value was set equal to zero. 25 From matrix algebra, the solution to equation (7) requires the existence of [I-D]-1 for which the columns of D should not all sum to one. This requires that there be at least one member of the network who does not depend only on other network members for some of the resources transacted through the network. This condition is satisfied for our network because, as mentioned in footnote 9, our trade network is not all-inclusive but the total exports and total imports are computed from the all-inclusive database. Therefore we have some countries for which the trade network data does not capture all international trade (in average we capture 90 to 95% of all trade). The reader is referred to Salancik (1986) for a detailed discussion of the properties of the importance measure. 16 above are reported. It is worth noticing that country rankings according to importance are starkly different from those obtained when countries are ranked by the ratio of total trade to GDP. The correlation between importance index (with Intrinsic Value = GDP ratio) and the ratio of total trade to GDP is 0.14, and between the same importance index and GDP is 0.79. This suggests that the importance results are not solely driven by wealth effects; there are consequential network effects26. IV. Application to Economic Growth This section illustrates the usefulness of the local integration indicators discussed above by introducing them in a growth accounting exercise where the objective is to determine the effect of international economic integration, sometimes referred to as “openness”, on economic growth. Harrison (1996), Frankel and Romer (1999), Irwin and Trevio (2002), and Yanikkaya (2003), among others, have used different indicators and methodologies, based on volumes of trade, in order to examine the relationship between openness and growth. Most of these studies consider a long-run growth model where a country’s GDP or income per capita growth rate ( γ y ) is a function of initial GDP conditions (yI), physical capital (k), human capital (h), and a vector of control variables (Z) that represent country specific characteristics (degree of openness, geographical, and political characteristics). γ y = F ( y I , k , h; Z ) (8) Following Harrison (1996) and Yanikkaya (2003), we use data from the World Development Indicators of the World Bank to calculate GDP per capita growth rates. Initial GDP per capita levels are obtained from the Penn Tables Mark 5.6. Life expectancy and telephone lines/1000 data, obtained from Easterly and Lu’s Global Development Network Growth Database27, are used as proxy variables for human and physical capital, respectively. Political regime and war deaths data is also obtained from Easterly and Yu. The geographical control variables included in the study are physical access to international waters and tropical climate, both obtained from the Sachs and Warner dataset28. For the degree of openness two types of variables have been considered in the literature. The first category includes indicators based on volumes of trade, like total trade (imports plus exports), the ratio of total trade (imports plus exports) to GDP, and total trade with OECD countries and non-OECD countries. The 26 Similar conclusions are reached with the centrality indicators. The correlation of the centrality index, at the one percent threshold, with the ratio of total trade to GDP is - 0.07 and 0.58 with GDP. 27 http://www.worldbank.org/research/growth/GDNdata.htm 28 Sachs and Warner data set is published on the Center for International Development Web site accessible from http://www.cid.harvard.edu/ 17 other category includes indicators based on trade restrictions, like tariffs, export duties and taxes on international trade in general. We use total trade to GDP ratio as the control variable for economic openness and compare these results to those obtained when we add the local integration measures, namely importance, maximum flow and degree centrality. Harrison (1996) and Yanikkaya (2002) estimate the following equation, γ y = β 0 + β1 y I + β 2 h + β 3 k + β 4Tropical + β 5Water (9) + β 6 Political + β 7War + β 8 Open + ε t and report a positive and strong relationship between trade shares in GDP and economic growth 29 . Specifically Yannikkaya (2000), through a panel regression analysis spanning over three decades (70’s, 80’s and 90’s), concludes that the coefficients (and their signs) for initial GDP conditions (-), human (+) and physical (+) capital, climate (-), and the total trade to GDP ratio (+) are strongly significant (at the one and five percent level) and robust, while those for the political regime (-), war deaths (-) and the physical access to international waters (-) are weakly significant (at the ten percent level). Due to limited data availability for the international trade network we only have network indicators for 1992 and 1998. Therefore we cannot follow Yanikkaya’s three period panel regression approach. We consider the data for 1987 to 1998 and divide the data into the periods 1987 - 1992 and 1993 - 1998. We average the variables for these two sample periods and perform a panel regression where the 1992 local integration indicators are used for the 1987 – 1992 sample and the 1998 indicators are used for the 1993 – 1998 sample. Our results are presented in Table 5. Column (1), which corresponds to the regression that uses the total trade to GDP ratio as the control variable for openness, shows that changing the panel regression from a three decade approach to the two sub-samples of 1987 – 1992 and 1993 – 1998 does not affect the results obtained by Yanikkaya. The coefficient for total trade to GDP ratio (+) is statistically significant while the other coefficients and their signs are also in line with his findings30. The rest of the columns in Table 5 show the results obtained when the IEI indicators are included in the analysis. These indicators incorporate network based measures of IEI for each country that embody more than just trade volumes. They capture a country’s relevance for the international trade network, whether it is at the center or the periphery of the trade network, 29 Yanikkaya (2002) uses the natural log of GDP as y and the natural log of life expectancy as h. The regressions in this I study, discussed below, use these transformations as well. 30 Our results show a positive sign for the control variable for access to international waters, while in Yannikkaya (2000) the sign is negative. This is explained by the definition of the variable. We use the proportion of land with access to international waters, while Yannikkaya uses the proportion of landlocked land. We did not include war deaths in our regression given that there is no data available for the late nineties. 18 and the magnitude of the direct and indirect effects it has on other countries. For the regressions we use the country rankings for each of the local integration indicators, where a lower number (higher ranking) denotes higher degree centrality and importance 31 . Therefore we expect negative signs for these variables in the regression results. As a country drops in the rankings, its relevance or its extent of IEI falls and therefore the advantages from trade and its positive effects on economic growth diminish accordingly. [Table 5 here] Column (2) presents the result for the econometric specification that includes the importance indicator with constant intrinsic value (i.e. with Intrinsic Value =1) as the IEI variable while excluding the total trade to GDP ratio. Columns (3) through (8) show the results obtained when other IEI indicators were used in the analysis while always including the total trade to GDP ratio. Columns (9) and (10) explore the possibility of the IEI indicators interacting with the level of physical and human capital. As a robustness check, columns (11) through (14) present the results obtained when we analyze the flow of goods and the import dependency measure, share of imports of country i from country j out of country i’s imports, is used to compute the network indicators instead of the export dependency measure, exports of i to country j out of the total exports of country i. The results of Table 5 show that the local integration indicators are statistically significant and have the expected negative sign. They posses explanatory power individually, when they are included as the sole control variable for economic integration32, and they add information to the economic growth regression when they are considered in conjunction with the total trade to GDP ratio. Moreover, the effect of higher centrality in the network is quite striking. For example, column (8) reports that an increase in the centrality ranking of 10 units at the two percent trade-link threshold increases the average growth rate of per capita GDP by 1.11 percentage points. A country’s position in the network can thus have substantial implications for economic growth. A more in-depth analysis of the results of Table 5 uncovers a possible relationship between the position of a country in the network and measures of physical and human capital that are included in the estimated equation. When the local integration indicators are introduced into the regression analysis, with and without the trade openness measure, the magnitude and the statistical significance of physical capital decreases while those for the level of human capital increase. Regarding geographical characteristics, climate 31 In the regression we use country rankings instead of the raw indices computed for two reasons. First the degree of fit of the regression is better with the ranking data. This leads to the second reason. In many cases the centrality or importance indices increase for a given country, but its ranking actually decreases. The fact that the ranking data gives a better degree of fit suggests that network effects are relative and not absolute, i.e. what matters is the relative position in the network. The general conclusions obtained with the rankings data hold when the regression analysis is carried out with the raw indices, but for matters of space these are not presented in the paper but are available upon request. 32 The coefficient for importance (IV=1) in column (2) is negative and statistically significant. This result holds for all the other local integration indicators used in the analysis, but individual results are not presented for matters of space. 19 and access to water, we find that their explanatory power in the regression is also diminished when the local measures of integration are included. Specifically, t-tests show that the human capital coefficients for some of the regressions that include the network measures of IEI are greater (statistically) than the one observed in regression (1). The coefficients on physical capital and the geographical variables (climate and access to water) are statistically significant in regression (1) but become statistically insignificant in a number of regressions that include the IEI indicators. These patterns suggest that a higher ranking in the centrality and importance indices diminishes the effects that country-specific characteristics (region, climate and technology) have on growth. A more integrated country is able to make up for the lack of good location and relevant technological improvements by being better connected in the network, i.e. physical capital and IEI are substitutes. And by being better connected, the positive effects of human capital on growth are amplified, i.e. human capital and IEI are complements. An interpretation of this result could be that human capital productivity is enhanced by international economic integration since a more integrated country offers greater growth opportunities to individuals. To test these conjectures more carefully we introduce the IEI indicators through an interaction term. We consider the case for importance (with Intrinsic Value = GDP ratio) and centrality, at the one percent threshold, and interact them with physical and human capital. However, the results, presented in columns (9) and (10), are not conclusive since the statistical significance of the coefficients is not consistent across the two regressions. The coefficient of the interaction term between human capital and the IEI indicators is only significant when the IEI indicator is the importance indicator, while for the interaction term between physical capital and IEI the coefficient is only significant when centrality is used as the IEI indicator. This suggests that the measured effect of international economic integration on economic growth may vary with the kind of IEI measure used. The importance of a country, which is a measure of how influential the country is, may interact differently with human and physical capital as compared to the manner in which the centrality of a country within the network interacts with physical and human capital. These differences in the interaction affects are intriguing and emphasize that the specific channels through which connectivity affects economic growth warrant deeper investigation. The overall conclusion that international economic integration matters for economic growth still continues to hold. As a robustness check, we compute all the network indicators (global and local) with the alternative approach of following the flow of goods instead of cash. The results are very similar to those obtained with the original approach (export dependency) and therefore are not discussed or reported on in detail. In this case the columns of the trade matrix denote importing countries, while rows correspond to exporting countries33. 33 In this case, from an export degree perspective the maximum degree for all threshold levels is equal to the number of countries included in the analysis minus one. For the import degree, this still holds for the zero and 0.5 percent 20 The network is again, extremely centralized with a network export degree (i.e out-degree) centrality, at the one percent threshold, of 87.25% in 1992 and 85.15% in 1998. Implying again a core-periphery structure, where the core once again includes the G-7 countries as the most influential nodes (suppliers) in the network. The similarities across the local measures of integration of both dependency measures can be observed in Table A1. This table reports the indices obtained for the importance indicator (with Intrinsic Value=GDP ratio) and the centrality indicator at the one percent threshold for both dependency measures. Moreover, the correlation between the columns is 0.98 and 0.96, respectively34. The regression results also match fairly well. Columns (11) through (14) present the results for the case where the network indicators are computed using import dependency. The numbers reported show that there are no noticeable differences from the results previously discussed. The network indicators increase the explanatory power of the regression and the complementarities between the network indicators and human capital still hold. V. Discussion We have attempted to chart the international trading system explicitly as a network and examine its structure and function from such a perspective. This has enabled us to obtain a clearer understanding of the structure of the global trading system and construct measures of international economic integration at both the global, system-wide level and at a local, country-level. While these metrics are implicitly based on the volume of international trade, they add new dimensions to the analysis of global integration that have not been previously considered and offer a new approach to describing local, country level integration into the global network. As a preliminary application we use our measures of network importance in a cross-country growth regression. Using these new measures we find evidence consistent with the hypothesis that a country’s position in the network has substantial implications for economic growth, but the specific channels through which connectivity affects economic growth requires deeper analysis. We believe that more detailed research into the relationship between human capital, physical capital, international economic integration and economic growth is warranted. The literature on financial contagion (Kaminsky and Reinhart, 2000, 2003; Forbes, 2001, Forbes and Rigobon, 2002) continues to puzzle over why many of the recent crises that began in relatively small economies had such global repercussions and why shocks originating in one economy spread to some markets, while markets in other countries were relatively unaffected. We find that ranking countries according to threshold. But given the criteria used to determine the presence of a link, the maximum import degree changes as the threshold increases. The reasoning for this follows from the arguments discussed in footnote 14. 34 The validity of the 1% and 2% thresholds for the import dependency measure is based on the same criteria used to justify these thresholds for the export dependency measures. 21 measures of “importance” to the network may provide insight into why and how financial crises are propagated than simple volume-based measures. For example, using 1992 data to construct the international trade network, we find that Thailand, a country which was the epicenter of the 1997-98 Asian financial crisis, ranks 22nd in terms of global trade share but 12th by our measure of network “importance”. In other words, network based measures identify several of the countries behind the financial crises and contagion of the 1990’s as highly influential countries, with a number of them even ranking above G-7 countries in terms of influence in the network. We thus believe that a network approach that is capable of incorporating the cascading of interdependent ripples that happens when a shock hits a specific part of the network will provide us with a deeper understanding of economic and financial contagion. It is also possible that such network-based measures may have real policy relevance in terms of identifying countries that are potentially vulnerable nodes for the entire network in case of economic and financial collapse. In a separate paper (Kali and Reyes, 2005) we examine this question in more detail by using these network measures of country-level and global integration as the backbone upon which to explore transmission mechanisms for international financial crises. In Kali and Reyes (2005) we use network-based measures of connectedness to explain stock market returns during recent episodes of financial crisis. We find that a crisis is amplified if the crisis epicenter country is better integrated into the trade network. However, target countries affected by such a shock are in turn better able to dissipate the impact if they are well integrated into the network. This arguably leads to a better understanding of why the Mexican, Asian and Russian financial crises were highly contagious, while the crises that originated in Venezuela and Argentina did not have such a virulent effect. In conclusion, we believe a network approach to international economic integration may have useful applications, both academic and policy, in several areas of international business, finance and development. 22 References Abeysinghe, T. and Forbes, K. (2002), “Trade Linkages and Output-Multiplier Effects: A Structural VAR Approach with a Focus on Asia.” Forthcoming in the Review of International Economics. Albert, A. and A. L. Barabasi, (2002), “Statistical Mechanics of Complex Networks”, Reviews of Modern Physics 74, 47. Bhagwati, J. (2004), In Defense of Globalization, Oxford University Press. Borgatti, S.P., Everett, M.G. and Freeman, L.C. 2002. Ucinet 6 for Windows. Harvard: Analytic Technologies. Feenstra, R. (1998), “Integration of Trade and Disintegration of Production in the Global Economy,” Journal of Economic Perspectives, 12(4): 31-50. Feenstra, R., Huang, D-S. and, G. Hamilton (2003), “A Market-Power Based Model of Business Groups,” Journal of Economic Behavior and Organization, 51, 2003: 459-485. Forbes, K. (2001), “Are Trade Linkages Important Determinants of Country Vulnerability to Crises,” In Sebastian Edwards and Jeffrey Frankel, eds., Preventing Currency Crises in Emerging Markets. University of Chicago Press, Chicago, pgs. 77-124. Forbes, K. and M. Chinn, (2004), “A Decomposition of Global Linkages in Financial Markets Over Time,” Review of Economics and Statistics, 86(3): 705-722. Forbes, K. and R. Rigobon, (2002), “No Contagion, Only Interdependence: Measuring Stock Market Comovements,” Journal of Finance, 57(5), pgs. 2223-2261. Freeman, L. C. (1979), “Centrality in Social Networks: Conceptual clarification.” Social Networks 1, 215239. Frankel, J. A. (1997), “Regional Trading Blocks in the World Trading System,” Washington DC: Institute of International Economics. Frankel, J.A. and D. Romer, (1999) “Does Trade Cause Growth?,” American Economic Review, 89(3) :379 – 399. Frankel, J. A., E. Stein and S. Wei, (1995), “Trading Blocs and the Americas: the Natural, the Unnatural and the Supernatural,” Journal of Development Economics, 47(1):61 – 95. Freeman, L. C. (1977) A set of measures of centrality based on betweenness. Sociometry 40, 35--41. Glaeser, E. L., Laibson, D. and B. Sacerdote (2002), “An Economic Approach to Social Capital,” Economic Journal 112(483): 437-58. Gomory, R. E. and T. C. Hu, (1964) “Synthesis of a communication network,” Journal of SIAM (Applied Math), 12, 348 . Granovetter, M. (1985), "Economic Action and Social Structure: The Problem of Embeddedness," American Journal of Sociology, 91 (3): 481-510. 23 Harrison, A. (1996) “Openness and Growth: A Time Series, Cross-Country Analysis for Developing Countries,” Journal of Development Economics, 48, 419 – 447. Hanneman, R.A. (2001) “Introduction to Social Network Methods,” Mimeo. University of California, Riverside. International Monetary Fund, (2002) “IMF World Economic Outlook: Trade and Financial Integration,” World Economic and Financial Surveys, September, 2002. Irwin, D. (1996), “The United States in a New World Economy? A Century’s Perspective,” American Economic Review, 86: 41-46, May 1996. Irwin, D. and M. Trevio (2002) “Does Trade Raise Income? Evidence from the Twentieth Century,” Journal of International Economics, 58, 1 – 18. Kali, R. and J. Reyes (2005b), “Financial Contagion on the International Trade Network,” Working Paper, University of Arkansas. Kaminsky, G. and C. Reinhart (2003), “The Center and the Periphery: The Globalization of Financial Shocks,” NBER Working Paper 9479. Kaminsky, G. and C. Reinhart (2000), “On Crisis, Contagion and Confusion,” Journal of International Economics, Vol. 51 No. 1, June 2000, 145-168. Krugman, P. (1995) “Growing World Trade: Causes and Consequences,” Brookings Papers on Economic Activity, No. 1, 327-62. La Porta, R., Lopez-de-Silanes, F., Shleifer, A. and Vishny, R. (1998), “Law and Finance,” Journal of Political Economy,” 106: 1113-55. Newman, M.E.J. (2003), “Mixing Patterns in Networks”, Phys. Rev. E 67, 026126. Newman, M. E. J. (2003), “The Structure and Function of Complex Networks”, SIAM Review 45, 167-256. Putnam, R. (2000), Bowling Alone: The Collapse and Revival of American Community, NY. Simon and Schuster. Rauch, J.E. and A.Casella (2001) (eds.), Networks and Markets, New York: Russell Sage Foundation, 2001. Rodrik, D. (1998), “Why do More Open Economies Have Bigger Governments?,” Journal of Political Economy, Vol. 106, No. 5, 1998. Rodrik, D. (2000), “How Far Will International Economic Integration Go?” Journal of Economic Perspectives, 14(1): 177-186. Rodrik, D., A. Submranain and F. Trebbi (2002), “Institutions Rule: The Primacy of Institutions Over Geography and Integration in Economic Development” Mimeo, Harvard. Sabidussi, G. (1966) The centrality index of a graph. Pyschometrika 31, 581--603. 24 Salancik, G. (1986), "An Index of Subgroup Influence in Dependency Networks,” Administrative Science Quarterly, vol. 31, 194-211 Sachs, J. and A. Warner, (1995), “Economic Reform and the Process of Global Integration,” Brookings Papers on Economic Activity, Vol. 1995, No. 1. 25th Anniversary Issue: 1-118. Shleifer, A. and E. Glaeser, (2002), “Legal Origins,” Quarterly Journal of Economics, 117(4): 1193-1191. Stiglitz, J. E. (2002), Globalization and Its Discontents, W.W. Norton & Company. Waserman, S. and C. Faust, (1994), Social Network Analysis, Cambridge University Press, 1994. Watts, D. J. and S. H. Strogatz. (1998) “Collective Dynamics of ‘Small-World’ Networks”, Nature 393:440442. Watts, D. J. (1999), Small Worlds: The Dynamics of Networks between Order and Randomness, Princeton University Press. World Bank (1998), “1998 World Development Indicators” World Bank. Yanikkaya, H. (2003), “Trade Openness and Economic Growth: A Cross-Country Empirical Investigation,” Journal of Development Economics, 72, 57 – 89. Zuckerman, E. (2003), “On Networks and Markets by Rauch and Casella, eds.” Journal of Economic Literature, 41: 545-565. 25 FIGURE 1: LORENZ CURVES AND GINI COEFFICIENTS FOR EXPORT AND IMPORT DEGREE DISTRIBUTIONS EXPORT DEGREE IMPORT DEGREE Lorent Curves of Imports Degree Distribution at the 1% threshold 1.00 0.90 Gini Coefficient: 0.80 for 1992: 0.22 0.70 for 1998: 0.21 0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.01 0.12 0.23 0.34 % of links % of links Lorenz Curves of Exports - Degree Distribution at the 1% threshold 1992 1998 45 degree 0.45 0.55 0.66 0.77 0.88 0.99 1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.01 Gini Coefficient: for 1992: 0.79 for 1998: 0.75 1992 1998 45 degree 0.12 0.23 0.34 % of countries % of links % of links 1992 1998 45 degree 0.45 0.55 0.55 0.66 0.77 0.88 0.99 Lorenz Curves of Imports Degree Distribution at the 2% threshold Lorenz Curves of Exports Degree Distribution at the 2% threshold 1.00 0.90 Gini Coefficient: 0.80 for 1992: 0.20 0.70 for 1998: 0.20 0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.01 0.12 0.23 0.34 0.45 % of countries 0.66 0.77 0.88 0.99 % of countries 1.00 0.90 0.80 0.70 0.60 0.50 0.40 0.30 0.20 0.10 0.00 0.01 Gini Coefficient: for 1992: 0.83 for 1998: 0.80 1992 1998 45 degree 0.12 0.23 0.34 0.45 0.55 % of countries 26 0.66 0.77 0.88 0.99 TABLE 1. PARTIAL BINARY MATRIX FOR ZERO PERCENT THRESHOLD IN 1992 I M P O R T E R S Afghanistan Albania Algeria Andorra Angola Barbuda Argentina Armenia Aruba Australia . . . Afghanistan 0 0 0 0 0 0 1 0 0 1 Albania 0 0 1 0 0 0 1 0 0 1 E X P O R T E R S Algeria Andorra Angola Antigua and Barbuda Argentina 0 0 0 0 0 0 0 0 0 1 0 0 1 0 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 Armenia 0 0 0 0 0 0 0 0 0 0 Aruba 0 0 0 0 0 0 0 0 0 0 Australia 1 1 1 0 1 1 1 0 0 0 . . . TABLE 2. SUMMARY RESULTS: NETWORK OVERVIEW Threshold 0.00% 1992 1998 Network Centralization (import degree) Network Density Clustering Coefficient (overall graph) Degree Correlation 56.79 41.92 0.78 -0.48 0.50% 1.00% 2.00% 1992 1998 1992 1998 1992 1998 42.55 82.38 84.23 81.70 82.94 77.03 78.25 56.62 9.54 10.97 13.21 14.40 17.89 18.65 0.77 0.54 0.50 0.49 0.46 0.45 0.41 -0.36 -0.21 -0.15 -0.17 -0.13 -0.12 -0.12 TABLE 3. ASSORTATIVE MIXING 1992 1998 0% 1% 2% Regional 0.067 0.244 0.248 0% 1% 2% 0.075 0.274 0.276 Income Legal Origin -0.041 0.012 0.057 0.150 0.064 0.169 -0.025 0.074 0.084 0.019 0.153 0.181 Notes: Higher values signify greater assortativity. Regional classification according to World Trade Organization. (North America, Latin America, Western Europe, C./E. Europe/Baltic States/CIS, Africa, Middle East, and Asia) Income classification according to World Bank. (High Income: OECD, High Income: Non-OECD, Upper middle Income, Lower middle Income, and Low Income) Legal Origin classification according to La Porta (1998). (British, French, Socialist, German, Scandinavian, and not classified) 27 TABLE 4. TOP THIRTY COUNTRIES ACCORDING TO THE 1998 IMPORTANCE INDEX (IV=1) Importance (IV=1) USA Germany Japan France United Kingdom Italy Belgium-Luxembourg Spain Russian Federation Netherlands Thailand India China Rep. of Korea Brazil Singapore Canada Portugal Australia Norway China, Hong Kong SAR Turkey Denmark Switzerland Saudi Arabia Austria Greece Sweden So. African Customs Union Poland 1992 1 2 10 3 4 5 6 7 33 9 12 14 8 11 24 16 18 13 20 31 22 21 15 19 17 25 54 42 28 23 1998 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 Importance (IV=TS) 1992 1 2 3 4 5 6 9 12 24 7 23 31 10 13 25 15 8 29 20 26 11 32 22 14 21 17 39 18 33 34 1998 1 2 3 4 5 6 10 11 20 9 24 27 8 14 21 15 7 34 22 28 13 31 25 16 32 19 41 17 40 30 Importance (IV=GDP ratio) 1992 1998 1 1 9 14 4 10 16 20 19 17 17 19 10 13 22 24 35 52 14 11 59 57 101 102 92 91 31 32 55 54 18 2 8 8 28 26 13 7 7 3 3 6 57 50 6 5 2 4 135 141 11 12 29 31 15 15 45 48 54 43 Total Trade to GDP Ratio 1992* 1998* 148 150 104 123 150 153 117 128 100 105 125 122 16 14 127 125 96 115 36 42 61 58 151 149 142 138 75 89 152 151 1 1 95 79 62 88 135 135 64 78 2 2 138 126 72 91 65 84 53 74 56 67 114 134 81 77 144 141 118 114 Notes: Countries ranked according to 1998 Importance index (IV=1). * For the ranking according to the Total Trade to GDP ratio, the data in the 1992 column is the average for the 1987 – 1992 period, while for the 1998 column the average is for the 1993 – 1998 time period. - Importance (IV=1) denotes the importance index was computed using a constant intrinsic value, set equal to one, for all countries, while Importance (IV=TS) and Importance (IV=GDP ratio) denote the importance indices computed using the share of total trade of country i (exports plus imports) out of total world trade and the GDP per capita ratio, respectively as intrinsic values. 28 TABLE 5. PER CAPITA GDP GROWTH RATE REGRESSION (1987 - 1992 and 1993 - 1998) Network Indicators Based on Export Dependency Ratio 1 Log (IGDP) -0.7570 ** 2 -0.7316 *** 3 -1.0009 ** 4 -1.2552 * 5 6 -0.9748 ** -0.9720 ** 7 -1.1453 * Network Indicators Based on Import Dependency 8 -0.9451 ** 9 -1.0335 * 10 -1.3743 * 11 12 13 14 -1.2155 ** -1.2309 * -1.0508 ** -1.4580 * -2.0592 -1.8171 -2.2941 -2.8034 -2.1294 -2.3448 -2.7069 -2.3332 -2.8070 -3.8767 -2.5630 -2.7966 -2.4236 -3.9397 Human Capital 1.4363 ** 1.8231 ** 2.1775 ** 2.7599 * 2.0968 ** 2.1659 ** 2.9382 * 2.6308 * 2.1160 * 3.8160 * 2.6785 * 2.6365 * 2.7908 * 3.8138 * Physical Capital 0.0120 * 2.0475 2.1937 0.0073 *** 2.4511 0.0093 ** 3.0119 0.0086 ** 2.2023 2.5960 0.0102 * 0.0086 ** 3.2410 3.0242 0.0049 0.0034 2.9383 0.0124 ** 5.0497 -0.0100 2.6422 0.0085 ** 3.4634 0.0094 *** 2.8735 0.0069 *** 4.7979 -0.0071 3.0078 1.9539 2.4225 2.2767 2.7196 2.2075 1.3274 0.8925 2.1560 -1.3246 2.1073 1.7615 1.8243 -0.9680 Regime -0.0917 -0.2763 -0.1884 -0.1790 -0.1345 -0.1674 -0.1136 -0.0938 -0.0528 0.0540 -0.1539 -0.0662 -0.1151 0.0268 Climate -0.7500 *** -0.2642 Access to Water -0.9530 -0.5854 -0.5518 -0.4081 -0.5104 -0.3504 -0.2917 -0.1445 0.1563 -0.4551 -0.3442 -0.3514 0.0762 -0.6557 -0.4669 -0.4871 -0.4841 -0.4423 -0.4082 -0.4760 -0.0433 -0.1399 -0.4561 -0.1785 -0.5035 -0.3094 -1.6382 -1.4091 -0.9844 -1.0593 -0.9664 -0.9468 -0.9208 -1.0582 -0.0863 -0.2902 -0.8890 -1.0947 -1.1095 -0.6299 1.3522 ** 1.3326 * 1.1844 ** 0.8438 1.0970 ** 1.2331 ** 1.1887 ** 1.3138 * 0.9006 1.0419 *** 1.0496 ** 0.9120 * 1.0182 ** 1.1871 ** 2.4179 Total Trade to GDP Ratio 2.6605 0.0102 ** 0.0095 ** 2.4777 Importance (IV=1) 2.3712 2.1378 -0.0084 *** -1.8324 1.6338 0.0118 * 2.8859 2.1158 2.5517 0.0089 ** 2.0268 0.0095 ** 2.1779 2.4429 0.0102 ** 2.5594 2.6928 0.0088 ** 2.2437 1.5541 0.0073 *** 1.6357 1.8883 0.0100 ** 2.4006 2.1263 0.0084 ** 1.9457 6.6483 0.0071 1.4160 1.9698 0.0091 ** 2.2555 2.1026 0.0096 ** 2.2447 -0.0133 * -2.6113 Importance (IV=TS) -0.0190 * -3.3306 Importance (IV=GDP ratio) -0.0115 ** -0.1288 ** -2.3916 -0.0162 ** -2.3744 Centrality 0% -2.0756 -0.0964 * -2.7795 -0.0242 * -3.1289 Centrality 1% -0.0795 * -0.2320 *** -4.5165 Centrality 2% -0.1413 * -1.6995 -3.3697 -0.21687 *** -1.6299 -0.1089 * -4.4856 Importance (IV=GDP ratio)*Human Capital 0.02947 ** Importance (IV=GDP ratio)*Physical Capital 0.02005 ** 2.1057 2.4742 -0.00005 -0.000002 -0.6661 -0.0487 Centrality 1%*Human Capital 0.0223 Centrality 1%*Physical Capital 0.0005 * 0.0242 0.6285 0.6985 0.0005 * 2.3348 R Squared Adj. R squared Number of observations 0.162 0.134 183 0.146 0.118 191 0.191 0.156 174 0.222 0.189 174 0.181 0.146 174 0.202 0.168 174 0.246 0.215 174 0.238 0.206 174 0.206 0.162 174 0.296 0.257 174 2.2243 0.173 0.138 174 0.185 0.141 174 0.210 0.177 174 Notes: t-statistics for the coefficients in italics. Rankings data for local IEI indicators was used in these regressions. *, **, and *** denote statistical significance at the 1%, 5% and 10% level. Importance (IV=1) denotes the importance index was computed using a constant intrinsic value, set equal to one, for all countries. Importance (IV=TS) and Importance (IV=GDP ratio) denote the importance indices computed using the world trade shares and the GDP per capita ratios as intrinsic values for the corresponding years. 29 0.267 0.227 174 DATA APPENDIX TABLE A1. RESULTS FOR LOCAL MEASURES OF ECONOMIC INTEGRATION World Trade Share Total Trade to GDP 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 Afghanistan Albania Algeria Andorra Angola Antigua and Barbuda Argentina Armenia Aruba Australia Austria Azerbaijan Bahamas Bahrain Bangladesh Barbados Belarus Belgium-Luxembourg Belize Benin Bermuda Bhutan Bolivia Bosnia Herzegovina Br. Virgin Isds Brazil Brunei Darussalam Bulgaria Burkina Faso Burundi Cambodia Cameroon Canada Cape Verde Cayman Isds Central African Rep. Chad Chile China China, Hong Kong SAR 1992 1998 1992* 1998* 0.005 0.008 0.296 0.015 0.086 0.003 0.398 0.001 0.012 1.061 1.399 0.004 0.041 0.049 0.082 0.010 0.019 3.522 0.006 0.009 0.016 0.002 0.024 0.008 0.002 0.854 0.057 0.091 0.005 0.003 0.009 0.037 3.537 0.002 0.011 0.003 0.003 0.278 2.901 2.516 0.005 0.011 0.207 0.011 0.053 0.004 0.560 0.006 0.015 1.047 1.181 0.015 0.024 0.046 0.112 0.012 0.146 2.986 0.006 0.011 0.013 0.001 0.031 0.036 0.016 1.096 0.035 0.092 0.010 0.002 0.019 0.033 4.013 0.003 0.012 0.003 0.002 0.309 3.921 2.167 51.196 40.152 71.097 197.145 15.718 94.418 35.095 76.213 198.540 19.025 96.621 93.043 130.292 125.980 58.920 78.199 45.907 16.081 87.000 37.685 36.379 22.024 36.295 52.884 65.765 40.339 45.615 62.454 27.761 259.622 55.991 52.758 135.817 204.961 17.500 94.028 40.202 78.645 185.350 28.012 97.431 126.471 129.938 104.711 58.838 74.871 47.528 17.705 101.073 39.892 30.542 62.636 44.478 71.659 87.228 43.626 51.138 57.074 38.915 275.828 0% (A) 21.8085 20.2128 64.3617 14.8936 25.0000 26.5957 62.7660 9.5745 14.8936 77.6596 92.5532 8.5106 31.9149 30.3191 58.5106 51.5957 11.7021 93.6170 44.1489 49.4681 52.6596 19.1489 53.1915 9.5745 17.5532 62.7660 52.1277 69.1489 20.2128 22.3404 19.1489 30.8511 81.9149 16.4894 21.2766 19.1489 17.0213 64.8936 86.7021 78.1915 Import - Degree Node Centrality (Index) 1992 1998 1%(A) 1% (B) 2% (A) 0% (A) 1%(A) 1% (B) 0.0000 0.0000 0.0000 33.5079 0.0000 0.0000 0.0000 0.0000 0.0000 49.2147 1.0471 0.0000 4.2553 4.7870 1.5957 70.6806 5.2356 2.0940 0.0000 0.0000 0.0000 56.5445 0.0000 0.0000 0.5319 0.0000 0.5319 39.2670 0.5236 0.0000 1.0638 0.0000 0.5319 36.6492 1.5707 0.0000 5.8511 10.1060 4.2553 71.2042 4.1885 14.1360 0.0000 0.0000 0.0000 31.9372 0.0000 0.0000 0.5319 0.5320 0.0000 24.0838 0.5236 0.5240 16.4894 22.8720 8.5106 90.5759 15.1832 23.5600 26.5957 20.7450 11.7021 97.3822 19.8953 16.7540 0.0000 0.0000 0.0000 50.2618 1.5707 1.0470 1.0638 0.0000 0.5319 50.7853 0.5236 0.0000 1.0638 3.1910 0.5319 40.3141 0.5236 3.6650 3.1915 0.0000 1.0638 79.0576 7.8534 1.0470 4.7872 3.7230 1.5957 63.3508 3.1414 3.6650 1.0638 0.5320 0.5319 67.5393 6.8063 5.2360 56.3830 57.9790 35.6383 96.8586 58.6387 56.0210 0.5319 0.0000 0.0000 42.4084 0.0000 0.0000 1.5957 0.0000 1.0638 64.3979 2.0942 0.5240 1.0638 0.0000 0.5319 37.1728 0.0000 0.0000 0.0000 0.0000 0.0000 20.4188 0.0000 0.0000 1.5957 0.5320 0.5319 62.3037 0.5236 0.5240 0.5319 0.5320 0.5319 32.9843 2.0942 1.0470 0.0000 0.0000 0.0000 32.4607 0.5236 0.5240 14.8936 32.4470 10.1064 81.6754 16.2304 28.7960 0.0000 0.0000 0.0000 47.6440 0.0000 0.0000 10.6383 1.0640 6.3830 73.8220 4.1885 4.7120 0.5319 0.0000 0.0000 59.6859 1.0471 0.0000 0.0000 0.0000 0.0000 41.3613 1.0471 0.0000 0.0000 0.0000 0.0000 30.3665 0.0000 0.0000 0.0000 0.0000 0.0000 60.7330 2.6178 0.5240 32.9787 32.4470 12.7660 94.7644 36.1257 31.9370 0.0000 0.0000 0.0000 25.1309 0.0000 0.0000 0.0000 0.0000 0.0000 27.2251 0.5236 0.0000 0.0000 0.0000 0.0000 32.4607 0.0000 0.0000 0.0000 0.0000 0.0000 29.3194 0.0000 0.0000 6.3830 5.3190 4.2553 63.3508 6.8063 5.2360 31.3830 55.3190 23.9362 87.4346 32.4607 71.2040 21.2766 39.8940 13.8298 87.4346 21.9895 29.8430 30 2% (A) 0.0000 0.5236 3.1414 0.0000 0.0000 1.0471 2.6178 0.0000 0.0000 6.2827 12.0419 1.5707 0.0000 0.0000 5.7592 2.6178 2.6178 41.8848 0.0000 1.5707 0.0000 0.0000 0.0000 1.5707 0.0000 9.9476 0.0000 1.5707 1.0471 0.5236 0.0000 1.5707 14.1361 0.0000 0.0000 0.0000 0.0000 5.2356 21.4660 12.0419 Importance (Indices) 1992 1998 IV=1 (A) IV= Trade Share (A) IV= GDP ratio (A) IV= GDP ratio (B) IV=1 (A) IV= Trade Share (A) IV= GDP ratio (A) IV= GDP ratio (B) 1.000009 0.000051 0.000229 0.000056 1.000023 0.000050 0.000307 0.000101 1.000008 0.000076 6.391498 6.391366 1.000061 0.000110 9.519243 9.517703 1.000356 0.002961 18.339408 18.338620 1.000391 0.002069 14.094820 14.093288 1.000011 0.000147 0.000605 0.000013 1.000012 0.000108 0.000434 0.000035 1.000080 0.000859 6.616348 6.614312 1.000044 0.000533 0.001587 0.000773 1.000051 0.000027 51.981323 51.979640 1.000100 0.000041 46.270562 46.267650 1.000947 0.003980 33.301830 33.307900 1.001010 0.005604 37.007575 37.012529 1.000004 0.000009 10.686923 10.686900 1.000021 0.000063 8.119077 8.118839 1.000030 0.000121 0.000566 0.000331 1.000031 0.000147 0.000577 0.000429 1.002347 0.010623 76.285661 76.300752 1.002176 0.010478 77.610933 77.629868 1.001577 0.014006 77.363925 77.363254 1.001603 0.011822 71.452152 71.450111 1.000018 0.000039 0.000226 0.000059 1.000308 0.000153 7.699721 7.699017 1.000069 0.000406 0.001961 0.000761 1.000048 0.000238 0.000971 0.000453 1.000352 0.000488 0.010866 0.006434 1.000055 0.000464 0.001203 0.001498 1.000916 0.000824 4.951931 4.950241 1.000745 0.001126 5.060631 5.058723 1.000231 0.000100 51.105952 51.107429 1.000231 0.000123 49.828717 49.830107 1.000130 0.000186 28.507949 28.507852 1.000490 0.001464 22.290609 22.288741 1.007874 0.035258 77.467568 77.458359 1.007374 0.029891 71.332601 71.316856 1.000031 0.000065 22.577584 22.577016 1.000042 0.000056 20.464405 20.464032 1.000213 0.000086 3.887956 3.887689 1.000327 0.000108 3.657450 3.656828 1.000067 0.000163 0.000980 0.000739 1.000020 0.000126 0.000652 0.000194 1.000004 0.000019 0.000071 0.000005 1.000006 0.000015 0.000069 0.000007 1.000117 0.000239 9.231486 9.230795 1.000075 0.000312 8.565582 8.564637 1.000063 0.000082 0.000244 0.000212 1.000475 0.000360 0.006992 0.000963 1.000008 0.000017 0.000316 0.000035 1.000034 0.000158 0.000850 0.000180 1.001677 0.008551 23.087249 23.110156 1.002553 0.010975 21.834634 21.827948 1.000026 0.000566 0.000998 0.001503 1.000024 0.000353 0.000897 0.000724 1.000979 0.000910 24.484019 24.473479 1.000283 0.000923 17.360599 17.360930 1.000015 0.000048 3.432496 3.432433 1.000111 0.000100 3.040029 3.039718 1.000007 0.000029 2.948036 2.947979 1.000053 0.000023 1.961360 1.961179 1.000011 0.000094 0.000577 0.000088 1.000016 0.000188 3.987437 3.986910 1.000014 0.000373 6.854501 6.854724 1.000150 0.000332 6.263795 6.263479 1.002507 0.035403 80.221615 80.219546 1.002350 0.040172 77.460183 77.448693 1.000015 0.000020 10.835566 10.835339 1.000011 0.000026 10.778145 10.777845 1.000018 0.000111 0.000764 0.000201 1.000030 0.000122 0.000705 0.000235 1.000001 0.000027 4.907603 4.907634 1.000003 0.000032 3.038041 3.038108 1.000001 0.000026 4.273180 4.273191 1.000014 0.000025 2.911908 2.911907 1.000529 0.002786 26.539030 26.536249 1.000628 0.003094 30.850239 30.847741 1.005212 0.029028 8.376523 8.405595 1.003421 0.039231 10.361804 10.443086 1.002065 0.025185 90.094537 90.115207 1.002043 0.021698 79.453587 79.449250 TABLE A1. RESULTS FOR LOCAL MEASURES OF ECONOMIC INTEGRATION (…continues) 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 China, Macao SAR Colombia Comoros Congo Costa Rica Cote d'lvoire Croatia Cuba Cyprus Czech Rep. Czechoslovakia Dem. People's Rep. of Korea Dem. Rep. of the Congo Denmark Djibouti Dominica Dominican Rep. Ecuador Egypt El Salvador Equatorial Guinea Eritrea Estonia Ethiopia Faeroe Isds Fiji Finland Fmr Ethiopia France French Polynesia Gabon Gambia Georgia Germany Ghana Greece Greenland Grenada Guatemala Guinea Guinea-Bissau Guyana Haiti Honduras Hungary Iceland India Indonesia Iran Iraq 0.053 0.199 0.002 0.029 0.078 0.058 0.111 0.031 0.061 n.a. 0.323 0.022 0.027 1.005 0.005 0.004 0.086 0.086 0.206 0.029 0.001 n.a. 0.006 n.a. 0.012 0.015 0.651 0.012 6.538 0.009 0.043 0.006 0.003 11.460 0.047 0.378 0.011 0.002 0.056 0.016 0.001 0.010 0.007 0.036 0.300 0.047 0.592 0.847 0.401 0.014 0.039 0.255 0.001 0.021 0.116 0.078 0.122 0.039 0.045 0.531 n.a. 0.019 0.018 0.836 0.004 0.003 0.113 0.105 0.194 0.055 0.007 0.003 0.076 0.020 0.008 0.011 0.693 n.a. 5.547 0.013 0.032 0.004 0.013 9.254 0.043 0.382 0.008 0.002 0.087 0.014 0.001 0.008 0.012 0.061 0.475 0.045 0.751 0.798 0.261 0.062 141.237 32.657 56.864 86.538 74.714 59.724 147.228 105.810 n.a. 75.980 49.319 65.788 128.603 122.847 70.156 59.056 52.547 44.360 110.265 n.a. 114.612 n.a. 111.199 49.426 21.234 43.927 81.789 124.449 84.161 49.322 43.398 44.585 105.304 41.119 54.979 53.128 159.240 37.022 63.078 67.541 65.629 16.592 48.947 30.245 - 117.359 35.152 62.036 136.199 87.684 76.911 95.244 97.655 95.900 n.a. 42.244 65.956 106.906 111.196 94.461 54.652 49.029 56.907 177.558 157.232 32.770 116.299 66.184 n.a. 43.828 95.477 114.676 67.538 46.788 58.353 40.370 107.831 42.204 43.800 48.721 192.789 31.549 84.614 74.276 68.393 24.792 52.654 47.506 - 32.9787 71.2766 19.1489 26.5957 47.8723 31.3830 65.4255 27.1277 64.8936 n.a. 34.0426 23.4043 22.3404 90.4255 26.0638 24.4681 29.2553 47.8723 37.2340 37.2340 15.9574 n.a. 13.2979 n.a. 51.0638 52.6596 80.3191 53.1915 97.8723 18.6170 23.4043 25.0000 11.7021 98.4043 73.4043 38.2979 46.2766 24.4681 31.9149 27.6596 18.0851 50.5319 26.5957 34.0426 73.9362 54.7872 69.1489 66.4894 30.8511 17.5532 0.0000 6.9149 0.0000 0.0000 1.5957 1.5957 3.7234 0.0000 1.5957 n.a. 2.6596 0.0000 0.0000 15.9574 0.5319 1.5957 1.0638 2.1277 2.6596 2.1277 0.0000 n.a. 0.5319 n.a. 0.0000 1.5957 10.1064 2.1277 81.3830 0.0000 0.0000 0.0000 0.0000 87.7660 2.6596 4.2553 0.0000 1.0638 0.5319 0.0000 0.0000 2.6596 0.0000 1.5957 6.3830 0.5319 18.0851 13.2979 7.4468 1.0638 0.0000 0.0000 42.9319 0.0000 0.0000 0.0000 1.000022 6.9150 3.7234 83.7696 6.2827 7.3300 3.1414 1.000987 0.0000 0.0000 30.3665 0.0000 0.5240 0.0000 1.000010 0.0000 0.0000 41.8848 1.0471 0.0000 0.0000 1.000013 2.6600 1.5957 65.4450 2.6178 3.6650 2.0942 1.000393 1.0640 0.5319 72.7749 4.7120 9.9480 3.1414 1.000115 2.1280 2.1277 76.4398 4.7120 3.1410 3.6649 1.000759 0.0000 0.0000 39.2670 0.5236 1.0470 0.5236 1.000047 0.5320 1.5957 74.8691 1.0471 0.5240 0.5236 1.000159 n.a. n.a. 86.9110 7.3298 10.4710 4.1885 n.a. 4.2550 1.5957 n.a. n.a. n.a. n.a. 1.000201 0.0000 0.0000 38.2199 0.0000 0.0000 0.0000 1.000021 0.0000 0.0000 31.9372 1.5707 0.0000 0.5236 1.000020 25.0000 8.5106 91.6230 12.5654 14.1360 6.2827 1.002766 0.5320 0.5319 34.0314 0.5236 0.0000 0.5236 1.000078 0.0000 1.0638 35.6021 2.0942 0.0000 1.0471 1.000070 0.5320 0.5319 39.2670 1.0471 0.0000 0.5236 1.000132 2.6600 0.5319 71.2042 4.7120 2.6180 1.5707 1.000126 1.0640 1.0638 77.4869 8.3770 4.1880 4.1885 1.000261 1.5960 2.1277 61.2565 4.7120 2.6180 3.1414 1.000137 0.0000 0.0000 23.5602 0.0000 0.0000 0.0000 1.000001 n.a. n.a. 21.9895 0.0000 0.0000 0.0000 n.a. 0.0000 0.0000 73.2984 4.1885 2.0940 2.6178 1.000013 n.a. n.a. 65.9686 1.5707 1.0470 1.5707 n.a. 0.5320 0.0000 53.4031 0.0000 0.0000 0.0000 1.000012 2.6600 1.0638 28.7958 0.0000 0.5240 0.0000 1.000093 13.2980 5.3191 87.9581 8.3770 10.9950 4.7120 1.000983 0.5320 1.5957 n.a. n.a. n.a. n.a. 1.000889 85.6380 70.2128 98.9529 80.1047 84.8170 63.8743 1.013698 0.0000 0.0000 64.3979 0.0000 0.5240 0.0000 1.000014 0.5320 0.0000 35.6021 1.0471 0.0000 0.0000 1.000007 0.0000 0.0000 48.1675 0.5236 0.5240 0.0000 1.000006 0.5320 0.0000 35.0785 1.0471 0.5240 0.5236 1.000003 92.0210 78.7234 98.9529 82.1990 90.5760 70.1571 1.019359 1.0640 1.5957 46.5969 2.6178 1.5710 2.0942 1.000201 1.5960 0.5319 84.2932 14.6597 6.2830 7.8534 1.000357 0.0000 0.0000 50.2618 0.0000 0.0000 0.0000 1.000013 0.0000 0.5319 51.3089 1.5707 0.0000 0.5236 1.000053 1.0640 0.5319 54.4503 3.1414 4.1880 2.6178 1.000075 0.0000 0.0000 57.5916 0.0000 1.0470 0.0000 1.000010 0.0000 0.0000 25.1309 0.5236 0.0000 0.0000 1.000003 0.5320 0.5319 35.6021 1.5707 1.0470 1.0471 1.000136 0.0000 0.0000 36.6492 0.0000 0.0000 0.0000 1.000004 0.5320 0.5319 55.4974 2.6178 2.0940 1.5707 1.000095 6.3830 1.5957 82.7225 10.4712 8.9010 4.7120 1.000419 1.0640 0.5319 64.9215 1.0471 0.5240 0.5236 1.000068 22.3400 14.3617 83.7696 20.4188 34.0310 14.1361 1.002892 12.2340 6.3830 89.0052 13.0890 25.1310 5.7592 1.000902 6.9150 3.1915 56.0209 7.3298 8.3770 3.6649 1.000401 0.5320 0.5319 32.9843 0.5236 2.0940 0.5236 1.000132 31 0.000532 0.001990 0.000019 0.000295 0.000778 0.000580 0.001112 0.000309 0.000613 n.a. 0.003237 0.000221 0.000266 0.010055 0.000054 0.000038 0.000862 0.000856 0.002063 0.000290 0.000014 n.a. 0.000063 n.a. 0.000118 0.000147 0.006512 0.000118 0.065446 0.000086 0.000430 0.000063 0.000027 0.114713 0.000467 0.003787 0.000114 0.000015 0.000562 0.000161 0.000014 0.000103 0.000068 0.000365 0.003006 0.000468 0.005924 0.008482 0.004019 0.000135 84.266554 18.725478 8.100254 7.371080 19.138250 7.527817 0.012796 20.162670 51.771525 n.a. 43.834048 0.000499 2.029670 82.970932 0.000237 24.279924 12.419053 14.440028 13.175909 13.381317 4.507980 n.a. 26.525010 n.a. 0.001143 18.196769 66.954891 1.879307 75.369818 0.000323 31.191388 4.515771 0.000123 78.175352 4.681447 45.358450 0.000858 18.564411 13.472651 9.862295 2.301102 8.383903 3.355387 7.970074 31.153068 77.136290 6.524222 12.014279 18.234856 0.002043 84.266090 18.705536 8.100119 7.371379 19.137486 7.527781 0.007945 20.162237 51.768941 n.a. 43.834930 0.000450 2.029734 82.976124 0.000161 24.279494 12.418306 14.440323 13.173511 13.381113 4.507975 n.a. 26.525017 n.a 0.001468 18.196334 66.959255 1.879147 75.370332 0.000010 31.191907 4.515733 0.000201 78.168246 4.680999 45.354995 0.000780 18.563843 13.472739 9.862378 2.301062 8.383381 3.355330 7.969794 31.152232 77.136019 6.524502 12.017141 18.233998 0.001906 1.000021 1.000663 1.000005 1.000065 1.000270 1.000425 1.000669 1.000068 1.000121 1.000728 n.a. 1.000019 1.000125 1.001847 1.000094 1.000089 1.000117 1.000313 1.000566 1.000384 1.000003 1.000009 1.000289 1.000562 1.000013 1.000011 1.000903 n.a. 1.011009 1.000024 1.000041 1.000030 1.000205 1.014137 1.000254 1.001479 1.000009 1.000080 1.000426 1.000048 1.000021 1.000090 1.000020 1.000201 1.000644 1.000094 1.003470 1.000873 1.000759 1.000151 0.000389 0.002551 0.000006 0.000207 0.001163 0.000783 0.001224 0.000387 0.000448 0.005315 n.a. 0.000194 0.000184 0.008367 0.000044 0.000028 0.001136 0.001049 0.001939 0.000553 0.000065 0.000034 0.000762 0.000205 0.000078 0.000108 0.006934 n.a. 0.055524 0.000126 0.000318 0.000041 0.000135 0.092627 0.000432 0.003826 0.000076 0.000024 0.000866 0.000137 0.000014 0.000082 0.000119 0.000612 0.004756 0.000448 0.007520 0.007986 0.002618 0.000623 67.703848 17.516604 5.320368 4.688901 17.545786 6.384617 27.011545 0.001800 0.003470 44.120983 n.a. 0.000651 0.000597 81.305629 0.000164 0.001028 14.889685 11.039027 12.396435 13.915658 9.634090 0.000082 28.620795 1.909343 0.000800 16.454331 68.959225 n.a. 68.127625 0.000431 23.583104 3.472129 15.043293 69.996274 4.298029 43.749310 0.000721 16.886398 12.786251 8.991573 2.011349 11.083694 7.339439 6.981768 30.621892 76.684430 7.311329 12.153833 16.910491 0.002591 67.703425 17.516803 5.320331 4.688767 17.546037 6.387173 27.005988 0.001925 0.001063 44.120556 n.a. 0.000688 0.000548 81.307929 0.000018 0.000634 14.888716 11.038490 12.393346 13.915102 9.634139 0.000012 28.620568 1.909060 0.000778 16.454199 68.964426 n.a. 68.132996 0.000075 23.583185 3.472138 15.042653 70.000616 4.297655 43.744003 0.000728 16.885904 12.786220 8.991812 2.011409 11.083283 7.339247 6.981299 30.621389 76.684017 7.312936 12.160376 16.909027 0.003176 TABLE A1. RESULTS FOR LOCAL MEASURES OF ECONOMIC INTEGRATION (…continues) 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 Ireland Israel Italy Jamaica Japan Jordan Kazakhstan Kenya Kiribati Kuwait Kyrgyzstan Lao People's Dem. Rep. Latvia Lebanon Liberia Libya Lithuania Madagascar Malawi Malaysia Maldives Mali Malta Marshall Isds Mauritania Mauritius Mexico Micronesia Mongolia Morocco Mozambique Myanmar Nepal Neth. Antilles Netherlands New Caledonia New Zealand Nicaragua Niger Nigeria Norway Oman Pakistan Palau Panama Papua New Guinea Paraguay Peru Philippines Poland 0.708 0.408 5.099 0.045 6.952 0.059 0.012 0.038 0.001 0.119 0.001 0.004 0.013 0.047 0.065 0.193 0.049 0.011 0.012 1.135 0.003 0.008 0.054 0.001 0.010 0.043 1.534 0.001 0.004 0.168 0.017 0.020 0.011 0.052 3.873 0.014 0.267 0.013 0.006 0.282 0.851 0.122 0.224 0.001 0.133 0.029 0.030 0.105 0.359 0.428 0.982 0.468 4.255 0.046 6.423 0.049 0.094 0.052 0.001 0.175 0.012 0.007 0.059 0.071 0.065 0.114 0.088 0.012 0.010 1.325 0.004 0.011 0.044 0.002 0.011 0.037 2.325 0.001 0.008 0.176 0.010 0.033 0.015 0.039 3.169 0.011 0.226 0.023 0.007 0.169 0.747 0.109 0.170 0.001 0.052 0.027 0.040 0.129 0.592 0.690 112.447 82.065 38.980 110.347 18.733 128.946 149.337 52.981 102.375 78.126 38.048 103.494 103.128 93.540 40.939 57.611 140.611 50.323 177.891 105.392 131.269 36.481 85.339 54.334 47.447 34.394 101.836 54.824 67.688 38.687 63.302 70.882 83.145 34.792 69.915 93.312 62.612 24.541 58.850 43.689 130.501 79.234 47.448 122.601 16.692 127.796 77.828 68.533 93.563 79.814 62.011 107.387 71.152 126.110 49.203 64.933 189.409 56.834 194.689 99.741 125.468 52.920 119.317 58.292 52.689 56.627 99.291 58.597 85.110 39.769 78.713 71.779 90.117 36.653 76.166 101.263 100.573 27.236 91.258 50.939 82.9787 53.1915 98.4043 59.0426 45.7447 52.6596 11.7021 54.7872 21.2766 39.3617 9.0426 19.1489 13.2979 34.5745 26.5957 30.3191 29.2553 50.0000 24.4681 79.7872 18.6170 25.5319 61.7021 9.0426 22.3404 64.8936 72.8723 6.9149 16.4894 36.7021 23.9362 53.7234 23.9362 46.8085 95.7447 19.1489 68.0851 26.0638 23.4043 38.8298 78.1915 47.3404 72.3404 6.3830 36.7021 22.8723 38.2979 61.1702 55.8511 78.7234 5.8511 4.2553 79.2553 4.2553 30.8511 4.2553 0.5319 1.5957 0.0000 2.1277 0.0000 0.0000 0.5319 1.0638 1.5957 2.6596 7.4468 1.0638 1.0638 14.3617 0.0000 0.0000 0.0000 0.0000 0.0000 3.7234 12.2340 0.0000 0.0000 2.6596 1.0638 0.5319 0.0000 3.1915 62.2340 0.0000 4.2553 1.0638 0.0000 1.0638 10.1064 1.0638 10.6383 0.0000 2.6596 0.5319 0.5319 4.7872 11.1702 17.0213 4.7870 1.5960 82.4470 5.8510 45.2130 2.1280 0.5320 6.3830 0.0000 1.0640 0.0000 0.0000 0.5320 0.5320 0.0000 2.1280 7.4470 0.5320 0.0000 18.0850 0.0000 0.0000 0.5320 0.0000 0.0000 1.5960 13.2980 0.0000 0.0000 1.5960 0.0000 0.5320 0.0000 6.9150 70.2130 0.0000 9.0430 1.0640 0.0000 3.7230 14.3620 2.6600 10.6380 0.0000 3.7230 0.5320 1.0640 1.5960 1.5960 13.8300 2.6596 1.0638 63.8298 2.6596 21.2766 1.5957 0.0000 1.5957 0.0000 1.5957 0.0000 0.0000 0.5319 0.5319 0.0000 1.0638 7.4468 0.0000 0.5319 7.9787 0.0000 0.0000 0.0000 0.0000 0.0000 1.0638 7.9787 0.0000 0.0000 0.5319 0.5319 0.5319 0.0000 2.1277 44.6809 0.0000 2.1277 1.0638 0.0000 1.0638 6.3830 1.0638 7.4468 0.0000 1.0638 0.0000 0.0000 2.1277 6.3830 10.6383 90.5759 72.2513 97.3822 64.3979 96.8586 58.6387 35.6021 60.7330 22.5131 71.7277 46.5969 27.2251 56.5445 87.9581 38.2199 35.6021 49.2147 58.1152 45.5497 80.6283 31.9372 38.7435 62.8272 20.4188 35.0785 69.1099 89.0052 11.5183 34.5550 73.2984 31.9372 32.9843 40.8377 36.6492 98.9529 24.0838 49.7382 55.4974 52.8796 72.2513 84.2932 61.7801 79.5812 12.0419 53.4031 60.7330 43.4555 67.5393 75.9162 82.7225 7.8534 6.8063 67.0157 4.7120 60.2094 2.0942 2.6178 2.0942 0.5236 4.1885 2.6178 0.0000 1.0471 3.6649 1.5707 3.6649 2.6178 0.5236 1.5707 13.0890 0.0000 2.6178 0.5236 0.0000 0.5236 3.1414 13.6126 0.0000 0.0000 3.1414 0.5236 0.0000 0.5236 1.0471 59.1623 0.5236 0.5236 2.6178 2.0942 1.5707 12.5654 1.0471 8.9005 0.0000 3.1414 1.5707 1.5707 4.7120 10.9948 18.8482 10.9950 1.5710 80.6280 1.5710 88.4820 1.0470 2.0940 6.2830 0.0000 3.1410 1.5710 0.0000 1.5710 1.0470 0.0000 2.0940 2.6180 1.0470 0.0000 25.6540 0.0000 0.0000 0.5240 0.0000 1.0470 1.5710 16.7540 0.0000 0.0000 3.1410 0.5240 0.0000 0.0000 1.5710 69.6340 0.0000 3.6650 0.5240 0.5240 4.7120 8.9010 2.6180 8.3770 0.0000 3.6650 1.0470 1.0470 1.5710 3.6650 11.5180 32 3.6649 1.5707 53.4031 2.6178 46.5969 1.5707 1.5707 0.5236 0.0000 2.6178 1.5707 0.0000 1.0471 1.5707 0.5236 1.5707 1.5707 0.0000 1.0471 6.8063 0.0000 2.6178 0.0000 0.0000 0.5236 1.5707 2.0942 0.0000 0.0000 2.0942 0.0000 0.0000 0.5236 0.5236 37.1728 0.5236 0.5236 2.0942 0.5236 1.5707 8.9005 1.0471 5.2356 0.0000 1.5707 1.0471 0.5236 2.0942 5.7592 7.8534 1.000686 1.000291 1.011094 1.000318 1.004673 1.001289 1.000021 1.000159 1.000013 1.000286 1.000003 1.000005 1.000039 1.000177 1.000081 1.000173 1.001448 1.000054 1.000045 1.001132 1.000010 1.000007 1.000053 1.000006 1.000006 1.000216 1.001456 1.000000 1.000003 1.000208 1.000057 1.000055 1.000022 1.000209 1.005041 1.000015 1.000513 1.000115 1.000008 1.000184 1.001142 1.000191 1.001318 1.000001 1.000193 1.000032 1.000056 1.000313 1.001001 1.001839 0.007088 0.004082 0.051045 0.000446 0.069567 0.000588 0.000122 0.000384 0.000006 0.001195 0.000009 0.000037 0.000130 0.000468 0.000649 0.001931 0.000487 0.000112 0.000125 0.011360 0.000035 0.000077 0.000536 0.000009 0.000104 0.000431 0.015363 0.000008 0.000041 0.001682 0.000166 0.000201 0.000106 0.000521 0.038768 0.000144 0.002673 0.000126 0.000059 0.002819 0.008521 0.001224 0.002241 0.000008 0.001331 0.000288 0.000303 0.001049 0.003590 0.004282 56.001632 54.545777 75.100863 14.053449 87.378619 13.952203 0.000197 4.444755 0.000104 0.002355 0.000006 0.000092 25.450657 13.273144 0.002226 0.004225 0.017305 3.150096 2.200073 26.395285 0.000124 3.116485 0.001104 0.000015 4.852306 38.850725 28.830682 0.000011 0.000042 13.296199 3.567050 0.000612 4.271000 0.007739 75.920383 0.000453 57.477683 6.911151 3.328483 3.367481 80.964717 0.002294 6.830991 0.000015 20.652176 12.259805 19.056968 13.773895 10.878976 23.522992 56.003128 54.543460 75.093075 14.053281 87.427675 13.950818 0.000212 4.445427 0.000006 0.001714 0.000006 0.000048 25.450634 13.270984 0.000766 0.004869 0.020783 3.150029 2.199817 26.396912 0.000063 3.116484 0.000818 0.000002 4.852259 38.850174 28.828730 0.000003 0.000020 13.295626 3.566615 0.000640 4.270869 0.009677 75.922577 0.000149 57.480036 6.910525 3.328459 3.368128 80.972330 0.003054 6.830405 0.000011 20.651017 12.259273 19.056722 13.772718 10.876700 23.521093 1.000759 1.000411 1.009812 1.000545 1.011265 1.000275 1.000360 1.000178 1.000014 1.000351 1.000171 1.000011 1.000207 1.000359 1.000143 1.000186 1.000263 1.000044 1.000116 1.000953 1.000016 1.000207 1.000048 1.000004 1.000047 1.000221 1.000955 1.000000 1.000013 1.000417 1.000026 1.000033 1.000048 1.000123 1.004893 1.000053 1.000155 1.000171 1.000091 1.000665 1.002060 1.000290 1.000930 1.000000 1.000190 1.000157 1.000083 1.000397 1.000875 1.001129 0.009827 0.004683 0.042595 0.000465 0.064285 0.000493 0.000937 0.000520 0.000006 0.001749 0.000120 0.000068 0.000586 0.000714 0.000653 0.001144 0.000882 0.000124 0.000104 0.013264 0.000044 0.000106 0.000440 0.000016 0.000114 0.000373 0.023278 0.000011 0.000081 0.001765 0.000103 0.000333 0.000151 0.000386 0.031718 0.000112 0.002264 0.000229 0.000069 0.001687 0.007475 0.001089 0.001698 0.000007 0.000521 0.000275 0.000396 0.001287 0.005924 0.006905 69.629486 54.373270 68.166548 10.683938 76.691880 12.608815 18.629835 4.122530 0.000117 0.003751 8.264857 0.000157 21.855285 16.467114 0.002491 0.004208 24.595319 2.596810 2.438134 31.657577 0.000169 2.820555 46.471870 0.000060 4.211827 41.704624 25.294002 0.000017 0.000194 13.374015 3.183892 0.000729 4.392844 0.002139 73.479269 0.000266 56.122345 5.170872 2.610686 2.537720 81.688101 0.001889 6.360905 0.000010 19.294675 10.275025 14.790340 14.367943 10.608268 27.102359 69.632859 54.371724 68.172402 10.681875 76.720227 12.608154 18.629789 4.122799 0.000018 0.003700 8.264768 0.000087 21.854388 16.465323 0.001352 0.004119 24.594139 2.596943 2.437878 31.663962 0.000056 2.820262 46.471784 0.000058 4.211849 41.704240 25.294615 0.000013 0.000136 13.373186 3.183693 0.000448 4.392629 0.001570 73.483132 0.000176 56.122902 5.169903 2.610602 2.538467 81.688260 0.002214 6.361473 0.000009 19.295511 10.274614 14.789701 14.366063 10.606973 27.098062 TABLE A1. RESULTS FOR LOCAL MEASURES OF ECONOMIC INTEGRATION (…continues) 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 Portugal Qatar Rep. of Korea Rep. of Moldova Romania Russian Federation Rwanda Saint Kitts and Nevis Saint Lucia Saint Vincent and the Grenadines Samoa Sao Tome and Principe Saudi Arabia Senegal Serbia and Montenegro Seychelles Sierra Leone Singapore Slovakia Slovenia So. African Customs Union Solomon Isds Somalia Spain Sri Lanka Sudan Suriname Sweden Switzerland Syria Tajikistan TFYR of Macedonia Thailand Timor-Leste Togo Tonga Trinidad and Tobago Tunisia Turkey Turkmenistan Uganda Ukraine United Arab Emirates United Kingdom United Rep. of Tanzania Uruguay USA Uzbekistan Vanuatu Venezuela Viet Nam Yemen Zambia Zimbabwe 0.692 0.054 2.137 0.002 0.100 0.933 0.003 0.002 0.006 0.003 0.003 0.000 1.045 0.021 0.167 0.004 0.008 1.766 n.a. 0.156 0.439 0.003 0.003 2.298 0.081 0.013 0.014 1.324 1.871 0.070 0.001 n.a. 0.997 0.000 0.013 0.001 0.045 0.150 0.538 0.004 0.007 0.054 0.212 5.415 0.020 0.053 13.744 0.008 0.001 0.392 0.060 0.041 0.027 0.050 0.591 0.083 1.970 0.020 0.202 1.148 0.003 0.002 0.004 0.003 0.001 0.001 0.684 0.022 0.067 0.004 0.003 1.779 0.226 0.183 0.453 0.003 0.002 2.326 0.087 0.027 0.010 1.456 1.570 0.072 0.006 0.028 0.935 0.000 0.010 0.000 0.054 0.139 0.686 0.010 0.016 0.271 0.484 5.300 0.023 0.066 15.589 0.043 0.002 0.323 0.179 0.035 0.019 0.034 71.323 63.644 116.370 45.109 52.823 23.855 140.507 163.290 140.178 97.932 95.195 80.617 55.373 121.023 36.635 375.556 n.a. 138.556 24.190 38.297 66.074 49.377 59.544 70.834 52.695 n.a. 71.543 82.292 73.030 85.362 32.766 27.789 54.155 106.450 50.376 44.069 43.543 20.696 72.690 53.222 47.742 44.834 70.610 49.147 67.480 66.375 122.062 58.545 50.701 36.704 127.539 137.100 112.562 115.388 115.775 74.158 69.226 128.891 46.370 352.436 117.998 113.324 35.560 45.453 79.420 10.947 71.957 68.547 71.553 85.776 69.722 94.107 89.068 44.592 31.179 80.383 127.932 56.488 55.821 43.079 23.750 85.778 49.683 84.208 73.319 72.583 76.482 85.6383 53.1915 79.7872 10.6383 34.0426 37.7660 20.7447 18.6170 47.8723 22.8723 15.9574 16.4894 75.5319 30.3191 60.6383 35.6383 23.9362 63.8298 n.a. 80.8511 80.8511 14.8936 22.8723 96.2766 55.8511 27.1277 23.9362 43.0851 93.0851 32.9787 9.0426 n.a. 79.7872 4.2553 24.4681 14.8936 59.0426 62.2340 79.2553 10.6383 26.0638 17.5532 59.0426 91.4894 29.2553 29.7872 96.8085 10.6383 18.6170 54.2553 27.1277 29.7872 28.1915 59.0426 23.4043 2.1277 28.7234 0.5319 2.1277 10.1064 0.0000 0.5319 3.7234 0.5319 0.5319 0.0000 18.6170 0.5319 6.9149 0.5319 0.0000 26.5957 n.a. 6.9149 7.9787 0.0000 0.5319 58.5106 3.7234 0.5319 1.0638 11.7021 25.5319 1.5957 0.0000 n.a. 25.0000 0.0000 1.0638 0.5319 4.2553 3.7234 19.1489 0.0000 0.5319 1.0638 3.1915 74.4681 0.5319 1.5957 88.2979 0.5319 0.0000 8.5106 0.5319 1.5957 1.0638 2.1277 8.5110 0.5320 40.9570 0.0000 2.1280 12.2340 0.0000 0.0000 2.1280 0.5320 0.0000 0.0000 17.5530 0.5320 4.2550 0.5320 0.0000 30.3190 n.a. 2.6600 12.2340 0.0000 0.0000 38.8300 0.5320 0.0000 0.0000 19.1490 47.3400 1.5960 0.0000 n.a. 27.1280 0.0000 0.0000 0.0000 7.9790 2.1280 15.4260 0.0000 0.0000 1.5960 6.9150 85.1060 0.0000 1.0640 94.1490 0.5320 0.0000 15.4260 0.5320 0.5320 0.0000 3.7230 12.7660 1.5957 23.9362 0.0000 0.5319 5.8511 0.0000 0.0000 2.6596 0.5319 0.0000 0.0000 10.6383 0.0000 3.1915 0.5319 0.0000 18.0851 n.a. 2.6596 4.7872 0.0000 0.5319 40.4255 1.5957 0.0000 0.5319 4.7872 13.2979 0.5319 0.0000 n.a. 15.4255 0.0000 0.5319 0.5319 2.6596 0.0000 11.1702 0.0000 0.5319 0.5319 1.5957 63.2979 0.5319 0.5319 81.3830 0.0000 0.0000 3.7234 0.0000 0.5319 1.0638 0.5319 88.4817 57.5916 89.0052 51.8325 79.5812 82.7225 56.5445 22.5131 47.1204 43.4555 17.8010 21.9895 80.1047 61.7801 79.5812 30.8901 35.0785 76.9633 75.9162 85.8639 91.0995 18.8482 26.7016 97.3822 44.5026 72.2513 42.4084 89.5288 94.7644 45.5497 27.2251 62.8272 92.6702 3.1414 59.1623 17.8010 64.3979 74.8691 84.2932 30.3665 73.8220 81.1518 53.9267 98.4293 72.7749 63.8743 97.9058 33.5079 21.4660 65.9686 43.4555 38.7435 59.1623 40.8377 21.4660 1.0471 27.7487 1.0471 6.8063 24.0838 2.0942 1.0471 3.1414 2.0942 0.0000 0.0000 12.0419 1.0471 4.7120 0.0000 0.0000 22.5131 3.1414 4.7120 6.2827 0.0000 0.5236 63.8743 1.5707 4.7120 1.5707 16.7539 24.6073 2.6178 0.5236 3.1414 24.6073 0.0000 1.5707 0.0000 4.7120 2.0942 22.5131 0.5236 2.6178 11.5183 7.8534 73.2984 1.5707 2.0942 90.0524 1.0471 0.0000 8.9005 1.5707 1.5707 2.0942 1.0471 5.2360 1.0470 55.4970 0.5240 5.7590 33.5080 0.0000 0.0000 0.0000 1.5710 0.0000 0.0000 18.3250 4.7120 1.0470 0.0000 0.0000 30.8900 2.6180 2.6180 16.2300 0.0000 0.0000 51.3090 0.5240 1.0470 1.0470 35.0790 37.1730 1.5710 0.0000 0.5240 30.8900 0.0000 2.0940 0.0000 6.2830 1.0470 17.8010 0.5240 2.0940 14.6600 12.0420 85.8640 3.6650 1.5710 93.1940 1.0470 0.0000 13.0890 3.6650 0.5240 2.0940 1.0470 12.0419 0.0000 17.2775 0.0000 2.6178 17.2775 1.0471 0.5236 2.0942 0.5236 0.0000 0.0000 8.3770 0.5236 1.5707 0.0000 0.0000 15.7068 0.5236 2.6178 4.7120 0.0000 0.0000 43.9791 0.5236 1.5707 1.5707 5.7592 8.3770 1.0471 0.5236 0.5236 16.2304 0.0000 0.5236 0.0000 3.1414 0.5236 9.9476 0.5236 2.6178 6.8063 4.7120 59.6859 0.5236 1.0471 82.7225 0.5236 0.0000 3.6649 0.0000 1.0471 0.5236 1.0471 1.002935 1.000139 1.004080 1.000011 1.000124 1.001037 1.000014 1.000032 1.000344 1.000055 1.000027 1.000001 1.002576 1.000030 1.000632 1.000034 1.000002 1.002712 n.a. 1.000720 1.001412 1.000006 1.000024 1.005749 1.000336 1.000041 1.000052 1.000871 1.002429 1.000108 1.000003 n.a. 1.003648 1.000000 1.000105 1.000035 1.000399 1.000232 1.002224 1.000009 1.000063 1.000167 1.000339 1.011954 1.000071 1.000077 1.030205 1.000026 1.000011 1.000704 1.000054 1.000130 1.000071 1.000130 0.006927 0.000545 0.021391 0.000016 0.001003 0.009339 0.000033 0.000015 0.000065 0.000032 0.000031 0.000005 0.010455 0.000211 0.001667 0.000040 0.000075 0.017678 n.a. 0.001563 0.004389 0.000028 0.000032 0.023012 0.000815 0.000132 0.000141 0.013249 0.018729 0.000701 0.000012 n.a. 0.009980 0.000000 0.000127 0.000009 0.000446 0.001501 0.005389 0.000035 0.000072 0.000536 0.002125 0.054205 0.000201 0.000534 0.137579 0.000085 0.000014 0.003923 0.000603 0.000411 0.000269 0.000505 Notes: (A) Denotes network indicators computed with the Export Dependency Ratio (i.e. exports of i to j out of total exports of i). (B) Denotes network indicators computed with the Import Dependency Ratio (i.e. imports of i from j out of total imports of i). (-) data is not available. (n.a.) country is not applicable. 33 48.958924 0.001283 43.296530 0.000088 15.709880 34.591843 4.135726 32.415887 23.042458 24.845484 0.000384 5.662109 0.023868 5.809950 0.010419 32.119525 3.708233 70.374220 n.a. 40.260357 27.567363 0.000131 0.000091 56.057349 10.060867 0.000373 0.001005 75.394605 91.597355 13.738387 0.000030 n.a. 20.740801 0.000002 4.685590 0.000315 36.229550 19.666077 22.635384 0.000111 2.505207 31.258953 0.005415 68.633975 1.609739 30.500886 100.589834 0.000197 0.000127 26.985461 4.401174 3.742668 3.612574 10.032842 48.953445 0.001163 43.297885 0.000111 15.709628 34.597808 4.135655 32.415513 23.039809 24.845195 0.000077 5.662093 0.026355 5.809829 0.008673 32.119442 3.708302 70.376369 n.a. 40.259710 27.574227 0.000055 0.000018 56.041207 10.060141 0.000149 0.000545 75.403369 91.602777 13.738080 0.000026 n.a. 20.736510 0.000002 4.685344 0.000020 36.234577 19.665117 22.631722 0.000146 2.505065 31.259233 0.006727 68.594516 1.609492 30.500887 100.570956 0.000170 0.000011 26.988201 4.400749 3.742230 3.612199 10.032801 1.002177 1.000114 1.002952 1.000063 1.000457 1.005329 1.000111 1.000052 1.000217 1.000101 1.000002 1.000005 1.001837 1.000234 1.000594 1.000009 1.000009 1.002547 1.000301 1.000459 1.001291 1.000004 1.000031 1.006225 1.000218 1.000342 1.000092 1.001422 1.001843 1.000210 1.000044 1.000162 1.003496 1.000004 1.000102 1.000000 1.000400 1.000261 1.001967 1.000061 1.000532 1.000936 1.000784 1.010674 1.000164 1.000500 1.030747 1.000177 1.000002 1.000887 1.000119 1.000113 1.000172 1.000086 0.005922 0.000835 0.019716 0.000198 0.002023 0.011488 0.000034 0.000016 0.000042 0.000034 0.000013 0.000006 0.006845 0.000225 0.000669 0.000037 0.000026 0.017814 0.002265 0.001834 0.004535 0.000030 0.000020 0.023281 0.000875 0.000266 0.000104 0.014578 0.015712 0.000716 0.000059 0.000277 0.009357 0.000000 0.000098 0.000005 0.000538 0.001391 0.006873 0.000099 0.000160 0.002712 0.004850 0.053059 0.000233 0.000661 0.156061 0.000433 0.000018 0.003235 0.001794 0.000350 0.000191 0.000342 47.079143 0.001645 42.316816 7.120538 15.052632 22.068008 2.836722 40.179348 19.851502 20.882870 0.000061 3.984000 0.017393 4.991881 0.002705 38.270087 2.422815 81.719478 36.731237 46.757611 23.644578 0.000106 0.000049 53.851243 10.642130 0.001851 0.001020 69.801958 81.614385 13.231600 3.830060 14.945334 19.915362 0.000005 3.052881 0.000013 26.166296 19.713282 22.397474 0.000595 2.897545 14.077592 0.011292 69.508126 1.396282 32.803976 100.579422 0.002554 0.000061 19.340935 0.004792 3.604136 2.485421 8.688628 47.073592 0.002145 42.325537 7.120406 15.051980 22.077258 2.836568 40.178903 19.850265 20.882634 0.000054 3.983981 0.020386 4.991998 0.001766 38.269988 2.422806 81.721306 36.730354 46.757087 23.651177 0.000094 0.000011 53.840964 10.641773 0.000496 0.000852 69.809061 81.617213 13.230974 3.829951 14.944494 19.919083 0.000001 3.052868 0.000005 26.171609 19.712447 22.392069 0.000430 2.897287 14.078641 0.010076 69.482232 1.395869 32.803445 100.535228 0.002526 0.000058 19.341968 0.004734 3.604019 2.484929 8.688653