Specialization Patterns In International Trade

Transcript

Specialization Patterns in International Trade⇤ Walter Steingress February 11, 2015 Abstract I document new facts on the pattern of international specialization by looking at export and import concentration. As a result of international trade, countries normally specialize in a few sectors, which tend to get exported, and diversify the importing sectors. To measure specialization, I compute concentration indexes for the value of exports and imports and decompose the overall concentration into the extensive product margin (number of products traded) and intensive product margin (volume of products traded). Using detailed product-level trade data for 160 countries, I find that exports are more concentrated than imports, specialization occurs mainly on the intensive product margin, and larger economies have more diversified exports and imports because they trade more products. Based on these novel facts, I assess the ability of the Eaton-Kortum model, the workhorse model of modern Ricardian trade theory, to account for the observed patterns. The results show that specialization through comparative advantage induced by technology differences can explain the qualitative and quantitative facts. Also, I evaluate the role of the key determinants of specialization: the degree of comparative advantage, the elasticity of substitution and geography. Keywords: Ricardian Trade Theory, Comparative Advantage, Specialization, Import Concentration, Export Concentration ⇤I thank Kristian Behrens, Andriana Bellou, Rui Castro, Jonathan Eaton, Stefania Garetto, Ulrich Hounyo, Joseph Kaboski, Raja Kali, Baris Kaymak, Michael Siemer, Ari Van Assche, Silvana Tenreyro and Michael Waugh for their useful comments and suggestions. This paper also benefited greatly from comments by seminar participants at Boston University, Carleton University, Georgetown, the University of Laval, HEC Montreal and the University of Montreal. All errors are my own. Contact: Banque de France, 31 Rue Croix des Petites Champs, Paris 75001, France (e-mail: [email protected]). 1 1 Introduction The pattern of specialization is at the core of international trade theory. A consequence of international trade is that countries do not need to produce all their goods, instead they can specialize in the production of certain goods in exchange for others. Trade theory offers different explanations of how countries specialize in the number and sales volume of goods. Assessing the empirical relevance of the underlying theory is of vital interest since it not only allows evaluating the gains from trade through specialization but also informs how the trade pattern affects the structure of an economy. For example, a high degree of specialization increases the likelihood that product specific shocks have aggregate effects in terms of output volatility and/or an impact on the terms of trade. The contribution of the paper is twofold. Firstly, it uncovers new facts on the pattern of specialization by looking at export and import concentration. It decomposes the overall level of concentration into a measure for the extensive and intensive product margin and documents concentration levels for exports and imports on all margins. The extensive product margin indicates the degree of specialization in the number of goods traded. The concentration index on the intensive margin measures specialization in the volume of goods traded. Secondly, the paper evaluates the Eaton and Kortum (2002) model’s ability to account for the observed specialization patterns. Specifically, it assesses the model based on three basic questions about specialization: What explains the level of specialization in exports and imports? What determines the gap between specialization in exports and imports? Does specialization occur on the intensive or extensive product margin? Based on detailed product-level trade data for 160 countries, the results show that, on average, countries specialize more in exports relative to imports, with Gini coefficients of 0.98 and 0.91 respectively. The decomposition reveals that specialization of exports occurs predominately on the extensive margin. Countries receive their export revenues from few products. At the same time, countries import a wide range of products but concentrate their expenditure towards a small number of products. Hence specialization of imports is driven by the intensive margin. The difference between the concentration levels of exports and imports is due to the extensive margin. Countries specialize in exporting few goods and diversify on imports by acquiring various products from abroad. Focusing on cross-country differences, I find that larger economies have more diversified imports and exports. This is mostly along the extensive margin, i.e. large economies export and import a wider product range. Having documented the observed specialization pattern, I employ a standard Ricardian trade model developed by Eaton and Kortum (2002) to evaluate its ability to reproduce the stylized facts. A key implication of this model is that it uncovers how comparative advantage due to technology differences determines specialization endogenously on both the extensive and the intensive prod- 2 uct margins. Furthermore, it identifies geography together with the elasticity of substitution and the degree of absolute and comparative advantage as the main determinants of specialization. A higher level of technology increases a country’s absolute advantage and diversifies the extensive margin of exports by broadening the product range it exports. The degree of comparative advantage heightens the sensitivity of concentration to changes in unit costs, thereby dictating specialization on both margins. Trade costs decrease comparative advantage and increase specialization on the extensive and intensive margin. A higher elasticity of substitution provides for better substitution between intermediate goods and allows countries to concentrate their expenditure in low price sectors. As a consequence, concentration on the intensive margin increases. To calibrate the model, I follow Waugh (2010) and use data and the structure of the model to infer trade costs, technology, the elasticity of substitution and the degree of comparative advantage. Not surprisingly, the simulated results show that the model produces the observed specialization pattern qualitatively with countries being specialized in exports and diversified in imports on all margins. More importantly, the simulated model also reproduces the degree of concentration on the extensive versus the intensive margin for both, exports and imports. However, the obtained levels for exports are too high in comparison to the data. Focusing on the variation across countries, the simulated model replicates the fact that larger economies are more diversified in exports but fails to account for the observed cross-country pattern of imports. At this point, it is important to note that the Ricardian model shares with other models of international trade, most notably monopolistic competition models based on Krugman (1980) and Armington models like Anderson and Van Wincoop (2003), the ability to develop quantitative predictions about specialization patterns on the intensive and extensive product margin. However, in these models tradable goods are differentiated by location of production since each country is the sole producer of a good. Countries specialize completely and demand all product country combinations. When applying this definition of the product space to the data, the analysis shows that countries are more concentrated in imports than in exports because they import only a small subset of available goods. This result implies that the empirical implications depend on the definition of the product space, i.e. differentiated versus homogenous goods. Consistent with the Ricardian model, the main empirical analysis is based on assumption that foreign varieties are perfect substitutes to domestic ones and local producers compete directly with imports for the lowest price. The robustness section discusses the alternative results based on the Armington assumption. This paper contributes to the international trade literature that analyses the relationship between the pattern of trade and specialization in commodities. Starting with MacDougall (1951), Balassa (1963), Golub and Hsieh (2000) and Costinot et al. (2012) test the Ricardian prediction that countries export relatively more of the commodities they are relatively more productive in. Unlike these papers, my analysis does not intend to explain why countries specialize in a certain commodity or group of commodities. Instead, it uses the level of concentration in trade data to 3 shed light on the factors that drive specialization in the number and the volume of goods traded. The levels of concentration in each trade direction contain information on the pattern of trade and as such they provide a new quantitative test of the extent of specialization observed in the data. The analysis presented in this paper also relates to a growing literature in quantifying the importance of Ricardian comparative advantage in explaining trade patterns using the EatonKortum framework, see, for example, Chor (2010), Shikher (2011), Levchenko and Zhang (2011) and Costinot et al. (2012). These papers specify a multi-sector Ricardian model with both interand intra-industry trade in order to derive implications on the sectorial level. In contrast, I abstract from intra-industry trade and attach a sectoral interpretation to the continuum of traded goods within the standard Eaton-Kortum framework. Given this notion, the number of traded sectors arises endogenously and is not fix as in the previous papers. While the standard model has been primarily used to explain bilateral trade flows and trade volume, (see, for example, Eaton and Kortum (2002), Alvarez and Lucas (2007) and Waugh (2010)), I focus on the implications on the pattern of trade and analyze how geography, tastes and absolute and comparative advantage induce countries to specialize in narrow sectors. Finally, my investigation adds to the empirical growth literature that analyzes the relationship between income and trade patterns on the intensive and extensive product margins, see Hummels and Klenow (2005) and Cadot et al. (2011). Contrary to the previous papers, I apply the decomposition also to imports and use the resulting empirical evidence to assess the ability of the Eaton-Kortum model to explain the relationship of income differences and the concentration of exports and imports along both margins. While Hummels and Klenow (2005) stress that models with Krugman firm-level product differentiation can explain why larger economies export more goods, my analysis shows that the Ricardian model of Eaton-Kortum offers an alternative framework to describe the observed patterns. The novel approach of linking cross-country variation of export and import concentration to test the Eaton-Kortum model sheds light on how the interaction between preferences, technology and geography establishes trade patterns on the intensive and extensive product margin. As such, the Eaton-Kortum framework can provide theoretical guidance for future work. The rest of the paper is organized as follows. Section 2 describes the data and presents the empirical evidence of import and export concentration. Section 3 lays out the theoretical framework. Section 4 describes the calibration that allows the model to replicate the empirical facts. Section 5 estimates trade costs and presents the simulation results based on the estimated trade costs. Section 6 discusses the robustness of the results while section 7 concludes. 4 2 Empirical evidence and data The starting point of my analysis is an empirical assessment of the observed specialization patterns in world trade using detailed product level trade data. Before describing the data and the empirical evidence, we examine the properties of the concentration measurements used, which form the basis of the qualitative and quantitative tests of the model. 2.1 Concentration measurements I compute two measures of specialization for product level sales, the Gini coefficient and the Theil index. The Theil index has the advantage of being decomposable into an extensive and intensive product margin measure. For concreteness, I focus on exports - concentration measures for imports are entirely analogous. The two measurements are defined as follows. Let k index a product among the N products in operation in the world economy, let Rk be the corresponding export sales revenue, say, in a given country. The export Gini in this country is defined as : G= 2 (ÂkN=1 kRk ) N ÂkN=1 Rk N+1 N (1) where export revenues of product k, Rk , are indexed in increasing order, i.e. Rk < Rk+1 , and N denotes the total number of products in the world. A Gini coefficient of zero expresses complete diversification across trade revenues, i.e. (1) a country exports all products and (2) the revenues are the same across them. An index of one expresses complete specialization in which case export revenues stem from one product only. Alternatively, the Theil index is a weighted average of the log difference from the mean export revenue ( R¯ ) and defined by the following formula 1 T= N  k2 N Rk ln R¯ ✓ Rk R¯ ◆ (2) The index takes the value of zero in the case of complete diversification and ln( N ) in the case of complete specialization. Cadot et al. (2011) decompose the Theil index into a measure for the intensive and extensive product margin, T = T ext + T int . The extensive Theil index ( T ext ) captures the concentration in the number of products (extensive product margin) whereas the intensive Theil ( T int ) measures the concentration in the sales volume of products (intensive product margin). The intensive Theil index is given by: T int = 1 Nx  k2 Nx Rk ln R¯ x ✓ Rk R¯ x ◆ (3) and the extensive Theil index is T ext = ln ✓ N Nx ◆ (4) Nx denotes the number of exported products and R¯ x represents the mean value of exported products. 5 2.2 Data To build my empirical evidence, I use the BACI data set provided by CEPII (Gaulier and Zignago (2009)) and choose the 6 digit HS 1992 product classification scheme is the preferred level of disaggregation. I follow Hummels and Klenow (2005) and refer to import flows of the same 6-digit product from different trading partners to different varieties of the same product. I assume that the tradable goods sector corresponds to manufactures defined to be the aggregate across all 34 BEA manufacturing industries, see Feenstra et al. (1997).1 Using a correspondence table provided by Feenstra et al. (1997), I identify 4529 tradable manufacturing products. The baseline sample covers 160 countries representing all regions and all levels of development between 1992 and 2009 (18 years). In total, the sample consists of 2880 observations (country-years). Note the data contains import and export flows within 6 digit product categories. The model I am assessing is Ricardian and does not feature trade between varieties of the same product. To establish a mapping between the model and the data, I net out the within product component by considering net trade flows instead of gross trade flows.2 To measure the importance of trade between products and trade between varieties, I follow Grubel and Lloyd (1975) and calculate the percentage share of trade between products with respect to total trade. I find an average value of 81 percent across countries. The overall share of total net trades flows with respect to total gross trade flows is 65 percent. Both findings suggest that the majority of trade flows between countries in this sample is across products.3 Based on net trade flows at the product level, I calculate concentration indexes for each country on all margins for each year and then take the average over the whole sample period. Because the concentration indices employed are independent of scale, the calculation on a year-to-year basis avoids the need to deflate the data. Figure 1 plots the mean export against the mean import concentration for each country together with the 45 degree line. In terms of overall concentration, Figures 1(a) and 1(b), the vast majority of observed levels lie above the 45 degree line highlighting the fact that exports are more concentrated than imports for almost all countries. On the intensive product margin, Figure 1(c), the specialization level of exports is similar to imports with slightly higher levels of concentration for exports. Figure 1(d) plots the results for the extensive product margin with countries exporting fewer products than they import. Table 1 summarizes the sample statistics with the average year-by-year indices over the 2880 1 This is a simplification, but it is reasonable as a first-order approximation because, for all countries in the sample, this represents on average 76 percent of all merchandise imports; the median is 91 percent. 2 I compute total net exports at the 6 digit product level and consider a country as an exporter of that product if net exports are positive and an importer otherwise. 3 In the appendix I present an alternative approach to account for observed intra-industry trade in the data. The basic idea is to develop a measurement device that enables the model to characterize trade within and across products. The suggested procedure converts the product units in the model to product units in the data and allows examining specialization patterns based on gross trade flows. In the rest of the paper, I follow the net trade flow approach. I present the estimation and results of the alternative procedure in the appendix. 6 Table 1: Mean concentration indexes over 2880 country-year pairs Gini Level of concentration % share of overall concentration Theil Exports (X) Theil Imports (M) Exports Imports Extensive Margin Intensive Margin Total Extensive Margin Intensive Margin Total 0.98 0.91 2.60 2.13 4.73 1.10 1.61 2.71 55% 45% 40% 60% country-year pairs. As implied by Figure 1, exports are more concentrated than imports on all margins. With respect to overall concentration, the summary statistics reveal high levels of export and import concentration with a Gini coefficient of 0.98 for exports and 0.91 for imports. In the case of exports, the high level of concentration is due to the fact that countries export few products and hence specialization is primarily driven by the extensive margin. For imports, the decomposition favors an alternative explanation. Countries import a fairly wide range of products but concentrate their trade in the value of few products. Focusing on the gap between export and import concentration, Table 1 shows that differences between exports and imports are mainly explained by the extensive margin. The Theil of 1.10 on the extensive margin of imports implies that, on average, a country net imports a 33.3 percent of all products. On the other hand, the extensive Theil of exports indicates that a country net exports 7.4 percent of the product space. In terms of the intensive margin, a country receives roughly 50% of its export revenues from 1% of the products it exports and spends 50% of its import expenditure on 2% of the products it imports. Overall, these results are consistent with the idea that openness to trade spurs countries to specialize in few exporting sectors and diversify the importing sectors. Turning to cross country differences, the empirical evidence shows that larger economies diversify more than smaller economies. Figure 2 plots the log of the mean levels of concentration as a function of market size including the best linear fit for all margins. Market size is measured by the log of the average GDP relative to the United States (USA = 0). As Figures 2(a) and 2(b) show, the overall Theil index decreases with respect to relative GDP, i.e. smaller economies specialize more. This relationship is more pronounced for exports than for imports with an R square of 0.58 compared to 0.41. The decomposition reveals that specialization on the intensive margin does not vary with market size for both, exports (Figures 2(e)) and imports (2(f)). The main driver of specialization differences across countries is the extensive margin. Particularly robust is the linear relationship on the extensive margin for exports with an R square of 0.75. Bigger economies are more diversified because they export more products, which is consistent with Koren and Tenreyro (2007)’s observation that larger economies are more diversified because they produce and export more products. The relationship between market size and specialization on the extensive margin of imports follows a L shape pattern. As the size of an economy increases, countries diversify 7 on imports until reaching a certain market size after which concentration is roughly equal across countries. At this point, the key qualitative and quantitative facts have been established. First, exports are more specialized than imports. Second, the extensive margin drives concentration of exports and the intensive margin for imports. Third, the target levels of concentration are displayed in Table 1. Fourth, the cross-country patterns imply a negative relationship between market size and specialization caused by the extensive margin, i.e. larger economies export and import more products. The rest of the paper evaluates the Ricardian model’s ability to account for these stylized facts. Next, I present the relevant parts of the Alvarez and Lucas (2007) extension of the EatonKortum framework. 3 Model The Eaton–Kortum model is Ricardian, with a continuum of goods produced under a constantreturns technology. In this paper, we focus on the Alvarez and Lucas (2007) model and include capital as in Waugh (2010). Next, I derive the relevant theoretical predictions on the pattern of trade and evaluate the importance of the key model parameters for specialization of imports and exports. Consider a world economy with I countries, where each country produces tradable intermediate goods as well as non-tradable composite and final goods. Following Alvarez and Lucas (2007), define x = ( x1 , ..., x I ) as a vector of technology draws for any given tradable good and refer to it I . The production of an intermediate good in country i is defined by: as “good x” with x 2 R+ qi ( xi ) = xi q [kai s1i ] qmi 1 a b b . Technology xi differs between goods and is drawn independently from a common exponential distribution with density f and a country specific technology parameter li , i.e. xi ⇠ exp(1/li ). Denote the interest rate by ri , the wage by wi and the price of the intermediate aggregate good by pm,i . The intermediate good sector is perfectly competitive. Intermediate good producers minimize input costs and sell the tradable intermediate good at price pi ( xi ) = Bxiq [ria wi1 where B = b b (1 b) (1 b ) . a b 1 b ] pmi . The continuum of intermediate input goods x enters the production of the composite good qi symmetrically with a constant elasticity of substitution (h > 0) qi = ˆ 0 • q ( x )1 1/h 8 h/(1 h ) f( x )dx . The produced aggregate intermediate good qi can then be allocated costless towards the production of final goods or being used as an input in the production of intermediate goods. Similarly, capital and labor can be used either to produce intermediate or final goods. Finally, consumers draw their utility linearly from the final good. All markets are perfectly competitive. Since these features are not central to the implications I derived in this paper, I omit them. The interested reader is refereed to Alvarez and Lucas (2007) for the full description of the model. 3.1 General equilibrium Once a country opens to international goods markets, the intermediate goods are the only goods traded. Final goods are not traded and capital and labor are immobile between countries. Trading intermediate goods between countries is costly. We define “Iceberg” transportation costs for good x from country i to country j by kij where kij < 1 8 i 6= j and kii = 1 8i. As in Alvarez and Lucas (2007), we also consider tariffs. wij is the tariff charged by country i on goods imported from country j. Tariffs distort international trade but do not entail a physical loss of resources. Incorporating the trade costs, composite good producers in country i will buy the intermediate good x from country j that offers the lowest price 2 pi ( x ) = B min 4 j 3 a b 1 b ] pmj q xj 5 . kij wij [r aj w1j (5) Equation 5 shows that whether country i specializes in the production of good x depends on the productivity realizations, factor prices and trade costs. If country i does not offer a good at lowest costs in the local market, the good is imported. Following Alvarez and Lucas, the resulting price index of tradable goods in country i is 0 where A = G(1 + q (h I 0 B pmi = ( AB) @  @ j =1 b 1 b w j pmj kij wij 1 A 1/q 1 C lj A q 1)) is the Gamma function evaluated at point (1 + q (h (6) 1)). Next, we calculate the expenditure shares for each country i. Let Dij be the fraction of country i’s per capita spending pmi qi on tradables that is spent on goods from country j. Then, we can write total spending of i on goods from j as pmi qi Dij = ˆ Bij pi ( x )qi ( x )f( x )dx where Bij defines the set of goods country j attains a minimum in equation 5. Note that Dij is simply the probability that country j is selling good x in country i at the lowest price and calculated to be 9 1/q Dij = ( AB) 0 @ a b 1 b ] pmj [r aj w1j pmi kij wij 1 1/q A lj. (7) Equation 7 shows that in this model the sensitivity of trade between countries i and j depends on the level of technology l, trade costs w, geographic barriers k and the technological parameter q (reflecting the heterogeneity of goods in production) and is independent of the elasticity of substitution h. This result is due to the assumption that h is common across countries and does not distort relative good prices across countries. Note also that by the law of large numbers, the probability that country i imports from country j is identical to the share of goods country i imports from j. In this sense, trade shares respond to costs and geographic barriers at the extensive margin: As a source becomes more expensive or remote it exports/imports a narrower range of goods. It is important to keep in mind that the number of intermediate input industries that enter the production of the composite good is fixed. Each country uses the whole continuum of intermediate goods to produce composite goods. There are no gains of trade due to an increased number of varieties. Welfare gains are realized through incomplete specialization. Domestic production competes with imports and countries specialize through the reallocation of resources made available by the exit of inefficient domestic producers. To close the model, we impose that total payments to foreigners (imports) are equal to total receipts from foreigners (exports) for all countries i Li pmi qi I  Dij wij = j =1 I  L j pmj q j Dji w ji (8) j =1 The previous equation implies an excess demand system which depends only on wages. Solving this system, describes the equilibrium wage for each country together with the corresponding equilibrium prices and quantities. Next, I describe the predictions on export and import concentration on both margins. 3.2 Concentration of exports and imports In the model, the pattern of trade is established by domestic producers competing with importers for selling intermediate goods in the local market. Given the equilibrium price, p( x ), and quantity, q( x ), the total expenditure that country i spends (c.i.f.) on imported good x, RiM ( x ), is: RiM ( x ) = Li pi ( x )qi ( x ) x2 / Bii I is the set of goods where country i obtains the minimum price at home. Equivawhere Bii ⇢ R+ lently, domestic producers export their good to all foreign markets where they attain the minimum price. The set of exporting goods is simply a collection of the set of goods country i exports to any destination j, x 2 [ jI6=i B ji . As a result, (f.o.b.) export revenue sales of good x, Ri,X ( x ), are given by: 10 RiX ( x ) = I  Lk pk (x)qk (x)kki wki x 2 [ jI6=i B ji k 6 =i Given the described pattern of trade, the concentration index for imports is identified. To show this, I decompose the overall concentration into a concentration measure for the intensive and extensive product margin. Using equation 3, the Theil index for the concentration of imports on the intensive production margin can be written as: int TiM = ˆ x2 / Bii RiM ( x ) ln R¯ iM ✓ RiM ( x ) R¯ iM ◆ f( x )dx In the appendix I show that the distribution of import expenditures follows a Fréchet distribution with shape parameter 1/q (h 1) and scale parameter si . Solving the integral, the intensive Theil index of imports for country i becomes: int TiM = ln (G(1 + q (1 ˆ h ))) 0 1 ⇣ ln u( q (1 h )) ⌘ e u du (9) where G(.) stands for the Gamma function. Import specialization on the intensive margin is independent of equilibrium prices, trade costs, geography and the level of technology l. It is solely determined by preferences (i.e the elasticities of substitution) and heterogeneity in production (i.e. the degree of comparative advantage). A higher elasticity of substitution (h ) increases specialization by allowing producers in the composite intermediate good sector to better substitute cheap for expensive products and concentrate expenditure towards these sectors. Similar, an increase in the degree of comparative advantage (q ), which corresponds to a higher variance of productivity realizations and therefore an increase in unit price differences across goods, heightens the degree of concentration. To compute the concentration of imports on the extensive margin, note that the set of goods produced is disjoint form the set of goods imported. Consequently, we can express the share of goods imported as 1 minus the share of goods produced, (1 Dii ). The Theil index for the extensive margin of imports is equal to : ext TiM = ln ✓ N NiM ◆ = ln(1 Dii ) (10) where Dii = ( AB) 1/q [ria wi1 a ] pmi ! b/q li and depends on the level of technology and equilibrium prices. To assess the level of specialization in exports, I simulate the model within a discrete product space in the following section. I then 11 calculate the export concentration index on the intensive margin according to equation 3. Having outlined the pattern of trade and the corresponding implications on the specialization pattern of exports and imports, the next section discusses the simulation of the model. It contains special cases of equilibria designed to spell out step-by-step the main implications of the model on export and import concentration and in further instance on specialization. 4 Calibration and simulation To simulate the theoretical model, which assumes an infinite amount of goods, I "discretize" the Fréchet distribution of total factor productivity and calculate the respective trade value for each product x. Concerning the parameters of the model, we need values for a, b, g, h and q. For a, b and g, I use the same values as Alvarez and Lucas (2007). We set the capital share to a = 0.3, the efficient labor share in the tradable goods sector to b = 0.5 and the labor share in the production of non-tradable final goods to a = 0.75. To calibrate the elasticity of substitution (h) and the variance of the productivity draws (q), I use the model’s implication on the revenue and the quantity distribution of imports. As shown in the previous section, the distribution of import expenditure follows a Fréchet distribution with shape parameter, 1/q (h 1). Similarly, it can be shown that the distribution of quantities imported also follows a Fréchet distribution with shape parameter, 1/(qh ). Using the fact that the Theil index on the intensive margin solely depends on the shape parameter, we first calculate the average Theil int = 1.61, and for imported quantities, T int = 3.58.4 Then, using index for import expenditure, TiM iQ equation 9, we get the corresponding shape parameters and obtain 2 equations with 2 unknowns. The solution of the system consists of an elasticity of substitution (h = 8) and a degree of comparative advantage (q = 0.10). The elasticity of substitution is high but still in the parameter range found in the literature, see Broda and Weinstein (2006). The degree of comparative advantage also lies in the parameter range 0.08 to 0.15 estimated in the literature, see Eaton and Kortum (2002). However, compared to more recent estimates by Simonovska and Waugh (2011), q = 0.10 is rather low. In the following subsections, I analyze import and export concentration in special cases of the equilibrium by assuming different trading schemes. Doing so builds intuition of how taste, technology and geography determine specialization. To illustrate the impact of each factor separately, it is instructive to start the analysis by assuming symmetric countries and introduce heterogeneity across countries later on. Finally, I show that for a particular configuration of trade costs the 4 In addition to the dollar value of imports, BACI also reports the volume of goods imported, which are measured in tons. To calculate the Theil index of the volume of imports, we follow the procedure applied to the import revenues. We int = 3.58 represents the first calculate the net volume of imports for each good and then apply equation 3. The value TiQ cross-country average over the sample period. 12 Eaton-Kortum model is able to replicate the specialization patterns observed in the data. 4.1 Symmetric countries All countries are identical. Trade costs are symmetric and set to kij = k 8 i 6= j with kii = 1 and wij = 1 8i, j. Due to symmetry, factor prices equalize across countries. The corresponding trade share matrix D is symmetric and the (i, j) element is given by: Di,j = (k )1/q 1 8i 6= j and Di,i = 1/q 1 + ( I 1)(k ) 1 + ( I 1)(k )1/q In free trade, k = 1, each country’s intermediate good producers specialize in a distinct set of goods equal to the relative size of the economy and export all products produced, Dii = Dij = 1/I. The corresponding share of imported products is 1 1)/I. In this case, Ricardian Dii = ( I specialization forces are strongest and the gap between export and import concentration reaches a maximum. Concentration on the Extensive Margin Including trade costs, the concentration index of imports equals the share of goods country i imports from all countries in the world and is given by: ext TiM = ln((1 Dii )) = ln(1 + ( I 1)(k )1/q ) ln(( I 1)(k )1/q ) Concentration at the extensive margin of imports increases with trade barriers k and decreases with the number of trading partners I 1 and the degree of comparative advantage q. Regarding exports, the extensive Theil index is given by the number of products exported to any destination divided by the total number of products in the world. To count the number of products exported, define the set of products exported as the union of the set of products exported to each destination, Uex = [ jI6=i B ji . Because the set of products exported to destination j overlaps with the set of products exported to destination k, B ji \ Bki 6= ∆, I apply the Inclusion Exclusion principle to avoid double counting. As I show in the appendix, under the assumption of symmetry, the extensive Theil index of exports is given by: ext TiX = ln I 1 Â( k =1 1) k 1 I k ! ak ! (11) where the share of products exported to k destinations, ak , is given by: ak = (k )1/q k + ( I k )(k )1/q The concentration of exports increases with geographical barriers, the degree of comparative advantage and the number of trading partners. In general, a larger number of trading partners increases competition between production and imports in the domestic market resulting in the 13 Table 2: Simulated export and import concentration indexes for benchmark parameters. Gini Parameters Exports Imports (k = 1) ( k = 0.7) ( k = 0.7, NT=10) Data 0.99 0.99 0.98 0.98 0.72 0.77 0.86 0.91 Theil Exports (X) Extensive Intensive Margin Margin 5.01 1.91 5.04 1.18 2.47 2.45 2.60 2.13 Total 6.92 6.22 4.92 4.73 Theil Imports (M) Extensive Intensive Margin Margin 0.01 1.61 0.10 1.61 1.09 1.61 1.10 1.61 Total 1.62 1.71 2.70 2.71 production of fewer goods at home and an increase in the number of goods imported. Also, more trading partners increase competition among exporters in foreign markets forcing countries to specialize more on the extensive margin of exports. Impediments to trade, i.e. a reduction in k, and a higher degree of comparative advantage, q, reduce import competition and, as a result, fewer goods are exported and imported. Notice that in the special case of free trade all goods produced are exported and concentration of production equals concentration of exports. With trade costs, countries export a subset of produced goods leading to more concentration of exports relative to production. Concentration on the Intensive Margin As noted previously the distribution of import expenditure follows a Fréchet distribution and is pined down by the elasticity of substitution (h) and the degree of comparative advantage (q). Concerning the distribution of export revenues, the simulation shows that it depends positively on the elasticity of substitution (h), the degree of comparative advantage (q) and geographical barriers (k ). The number of trading partners has non-monotone effects on the concentration of exports at the intensive margin. Few trading partners leads to high level of concentration because high and low productive goods enter export markets. However, as the number of trading partners increases, the degree of competition in the export markets also increases and only products with high productivity are exported. In the case of free trade, countries export all their goods to all destinations and, given that preferences are identical, export and import concentration on the intensive margin equalize. The results presented in Table 2 show that the free trade calibration of Alvarez and Lucas (2007) is able to replicate the qualitative fact that, overall, exports are more concentrated than imports. While the simulated overall level of export concentration attains the degree of specialization observed in the data, in the benchmark free trade parametrization countries diversify excessively in imports because they import too many goods. Next, I introduce 42 percent symmetric trade costs to all trading partners, k = 0.7. Row 3 of Table 2 shows the results. Impediments to trade reduce the number of products exported and imported and concentration on the extensive margin increases for both. Note that higher trade 14 costs lower the level of concentration on the intensive margin of exports. Due to the increase in trade costs, only very efficient producers export and their export revenues are more evenly distributed across products and trade partners. Still, the gap between export and import concentration remains substantial. The reason is that the degree of competition countries face in export and domestic markets is too high. In the symmetric setting the only way to reduce competition is to limit the amount of trading partners. Using equation 11, the number of trading partners (NT) corresponding to the empirical Theil index is 10, see fourth row of Table 2. Limiting the number of trading partners (NT) by introducing infinite trade costs to countries outside of the block reduces competition in all markets. Less competition in the domestic market increases the survival rate of local producers and reduces the amount of goods imported. Note that revenues of exporting industries are now geographically more concentrated and hence specialization on the intensive margin of exports intensifies. In sum, with the introduction of symmetric trade costs, the model can replicate the mean levels of concentration observed in the data. The key parameters are trade costs. In particular, by creating trade blocks, which amounts to introduce zeros in the bilateral trade matrix, we can calibrate the model to explain the mean pattern of specialization. 4.2 Asymmetric countries In this section I analyze the effects of cross-country heterogeneity on specialization. The empirical facts imply a negative relationship between specialization and market size. For this reason, I introduce heterogeneity in technology li to reflect the observed GDP differences in the data. To start with, consider equation 8 in a free trade equilibrium: ( wi L i + r i Ki ) = N  j =1 w j L j + r j K j D ji which can be simplified to ⇣ li = C (wi Li + ri Ki ) wia ri1 a ⌘ bq (12) where C is a constant. Using equation 12, I back out the level of technology, as a function of GDP (wi Li + ri Ki ) and endowments Li and Ki assuming that they are chosen optimal. To calibrate l, I use GDP, capital and population data from the Penn World table. I follow Waugh (2010) and normalize the obtained parameters for li , Li and Ki relative to the United States. Concentration on the Extensive Margin Plugging in the equilibrium wage into equation 7, I get the corresponding trade share matrix D with the (i, j) element given by: D ji = ( wi L i + r i Ki ) 8j ( wk Lk + r k Kk ) ÂkI =1 15 (13) Equation 13 shows that under the assumption of free trade country i’s share of the number of products exported is equal to its relative level of GDP with respect to world GDP. Hence, larger economies export more and import less products compared to smaller economies. This result is at odds with the empirical evidence. In the data, larger economies export and import more products. 4.3 Asymmetric trade costs To reconcile the cross-country concentration differences for imports, I consider trade costs as a function of either a fixed export cost (ex j ) or a fixed import cost (imi ). While both types of costs can reconcile the fact that larger countries import more goods, the exposition focuses on the fixed export cost.5 In this case, each country pays a country specific cost to export, which is independent of the importing country j, k ji = exi 8i 6= j and ki,i = 1 8 j = i. Due to asymmetric trade costs, wages and composite good prices do not equalize. The trade share matrix is asymmetric and given by: D ji = ( AB) 1/q 1 b ! 1/q [wia ri1 a ] b pm,i pmj exi and li 8i 6 = j Dii = ( AB) 1/q wia ri1 pm,i a ! b/q li Focusing on the expression for the share of goods that country i exports to country j, D ji , shows that a higher export cost reduces the fraction of the good that arrives in destination j (exi #) and decreases the number of goods country i exports to any destination j. Solving for the equilibrium and assuming that composite good prices across countries are approximately equal, one can show that the share of goods imported is approximately: (1 Dii ) t ✓ 1 1 q C1 exi (wi Li + ri Ki ) ◆ (14) where C1 is a constant. Equation 14 shows that the share of goods imported is decreasing in the country specific exporting costs, (∂(1 Dii )/∂exi > 0). Lower exporting costs allow producers to pay higher factor prices at home and still be competitive in export markets. At the same time, higher unit costs of production reduce competitiveness at home and result in a larger share of imported goods. Hence, an exporter fixed effect can reconcile the fact that larger economies import and export more goods. The main difference between the import cost and the export cost in terms of import concentration lies in the implication on the price level of tradable goods. One can show that the export cost implies a nearly constant price level of tradable goods across countries, see Waugh (2010). As a result, unit cost differences between countries are predominantly driven by factor price differences. On the contrary, the import cost leads to large cross-country price level differences with smaller economies facing a higher tradable price level. In this case, unit cost differences are driven by factor as well as tradable goods price level differences. Based on Waugh (2010)’s results that 5 One can apply the same reasoning to the fixed import cost. 16 Table 3: Simulated export and import concentration indexes for asymmetric countries. Gini Parameters Exports Imports ( k = 1) ( k = ex, NT=10) Data 0.99 0.98 0.98 0.73 0.85 0.91 Theil Exports (X) Extensive Intensive Total Margin Margin 5.75 2.59 2.60 1.91 2.67 2.13 7.66 5.26 4.73 Theil Imports (M) Extensive Intensive Total Margin Margin 0.007 1.10 1.10 1.61 1.61 1.61 1.62 2.71 2.71 countries have similar price levels of tradable goods, I focus only on the case of the exporter fixed effect for the rest of my analysis. In sum, the introduction of asymmetric trade costs in form of a country specific cost to export or import allows the model to replicate the import specialization pattern across countries, in particular when larger economies face relative low costs to either export or import. Waugh (2010) argues that trade costs have to be asymmetric, with poor countries facing higher costs to export relative to rich countries, in order to reconcile bilateral trade volumes and price data. While both our approaches highlight the importance of asymmetric trade costs in explaining trade data, our analysis differs. Waugh uses the Eaton Kortum model to explain bilateral trade volumes and price data whereas I look on the models implications on the specialization pattern of exports and imports. In this respect, the results presented in this paper provide further evidence on the importance of asymmetry in trade costs when studying trade volumes and trade patterns across countries. Row 1 of table 3 presents simulations results in the case of asymmetric countries and free trade. Note that in relation to the symmetric country case introducing technology differences increases the mean level of concentration for exports and decreases the level of concentration for imports. The underlying reason is that the technology distribution is skewed towards less productive countries and these countries export fewer and import more goods. Beside these changes, the results are similar to the symmetric case. To reconcile the empirical evidence that larger countries import more goods, I introduce country specific costs to export with larger economies facing relatively lower export costs. In particular, I calculate the implied export cost from equation 14 by replacing the share of goods produced at home by the extensive Theil index of imports observed in the data , Dii = 1 Ext ). Row exp( TM 2 of table 3 shows the results of the corresponding mean concentration levels. In terms of the cross country pattern, Figure 3 plots the simulated (in red) and the empirical (in blue) concentration levels against GDP for both margins. The figures show that the country specific export cost in combination with technology and endowment differences can replicate the across country evidence on all margins. 17 In the previous section I analyzed special cases of the equilibrium to study the different factors that determine specialization in the Eaton Kortum model. The key determinants are the degree of comparative advantage, the elasticity of substitution and asymmetric trade costs. However, I treated trade costs as free parameters and showed that for a particular configuration of trade costs, the model is able to reproduce concentration levels at the mean as well as the cross-country specialization pattern for both exports and imports. In the next section, I estimate trade costs and technology parameters based on bilateral trade shares using the model’s structure and check whether for given trade shares the model is able to generate the observed specialization pattern in the data. 5 Estimating trade costs from bilateral trade shares The starting point of the estimation of technology and trade costs is a structural log-linear “gravity” equation that relates bilateral trade shares with trade costs and structural parameters of the model. To derive the relationship, simply divide each country i’s trade share from country j, see equation 7, by country i’s home trade share. Taking logs yields I log ✓ Dij Dii ◆ = Sj Si + 1 equations for each country i : 1 1 log(kij ) + log(wij ) q q in which Si presents the structural parameters and is defined as Si = log([ria wi1 (15) (1 b)/q ] b/q pmi l i ). In order to estimate trade costs k and technology l implied by equation 15 I use data on bilateral trade shares across 160 countries. I follow Bernard et al. (2003) and calculate the corresponding bilateral trade share matrix by the ratio of total gross imports of country i form country j, Mij , divided by absorption Absi Dij = a Mij . Absi Absorption is defined as total gross manufacturing output plus total imports, Mi , minus total exports, Xi . To compute absorption, we use gross manufacturing output data from UNIDO.6 Combined with trade data from BACI, we get the expenditure share, Dij , which equals the value of inputs consumed by country i imported from country j divided through the total value of inputs in country i. Note that instead of focusing on a particular year, I compute the expenditure share for each year of the period 1992 - 2009 and take the average expenditure share over the sample period.7 6 The details are in the appendix. resulting sample consists of 160 times 159 potential observations if each country trades with all other countries. In our sample the total number of observations is 9649 implying a large number of zeros in the bilateral matrix. For this reason, I conduct a robustness test where I estimate the model with the Poisson estimator proposed by Silva and Tenreyro (2006). The appendix presents the results. 7 The 18 Table 4: Estimation Results Summary Statistics Observations 9649 Geographical barriers Barrier [0,375) [375,740) [750,1500) [1500,3000) [3000,6000) [6000,max) Tariff Shared border In total there are only I 2 TSS 2,60E+05 SSR 4,67E+04 R2 0.82 Paremeter estimate -4,89 -5,76 -6,78 -7,98 -9,05 -9,81 -0,23 1,37 Standard error 0,10 0,06 0,04 0,03 0,02 0,03 0,10 0,09 % effect on cost 79,93% 99,60% 125,62% 160,66% 196,42% 224,64% 5,47% -15,19% I informative moments and I 2 parameters of interest. Thus, restric- tions on the parameter space are necessary. To create them, I follow Eaton and Kortum (2002) and assume the following functional form of trade costs. log kij = bij + dk + wij + ex j + eij Trade costs are a logarithmic function of distance (dk ) a shared border effect between country i and j (bij ), a tariff charged by country i to country j and an exporter fixed effect (ex j ). Tariff (wij ) represents the weighted average ad valorem tariff rate applied by country i to country j. The distance function is represented by a step function divided into 6 intervals. Intervals are in miles: [0, 375); [375, 750); [750, 1,500); [1,500, 3,000); [3,000, 6,000); and [6,000, maximum]. eij reflects barriers to trade arising from all other factors and is orthogonal to the regressors. The distance and common border variables are obtained from the comprehensive geography database compiled by CEPII. ˆ and strucTo recover technology, I follow Waugh (2010) and use the estimated trade costs, k, ˆ tural parameters, S, to compute the implied tradable good prices, pˆ m , by rewriting equation 6 in ˆ terms of S: pˆ mi = ( AB) I Âe Sˆj j =1 kˆ ij wij 1/q ! q From the obtained prices and the estimates Sˆi , I get the convolution of wages and technology, log(wi b/q li ). Then, given the bilateral trade shares, Dij and the balanced trade condition in equa- tion 8, I follow Alvarez and Lucas (2007) and and use the relationship between factor payments 19 Table 5: Simulated concentration level with exporter fixed effect Gini Model Simulation Data Exports Imports 0.99 0.89 0.98 0.91 Theil Exports (X) Extensive Intensive Total Margin Margin 4.83 59% 2.60 55% 3.32 41% 2.13 45% Theil Imports (M) Extensive Intensive Total Margin Margin 8.15 0.84 34% 1.10 40% 4.73 1.61 66% 1.61 60% 2.45 2.71 and total revenue to calculate equilibrium wages.8 wi = ✓ (1 1 s f i ) Li ◆ I  Lj wj (1 j =1 s f j) Fj D ji w ji ! where s f i is the labor share in the production of final goods sfi = (1 g(1 (1 b) Fi ) g) bFi + g(1 (1 b) Fi ) and Fi is the fraction of country i spending on tradable goods net of tariff expenses. Fi = I  Dji w ji j =1 The obtained equilibrium wages together with tradable good prices, determine the implied technology levels lˆ for each country given the structural estimates of the gravity equation. Table 4 summarizes the regression outcome of the gravity equation. In terms of fitting bilateral trade flows, I obtain an R2 of 0.82 slightly lower than the R2 of 0.83 reported by Waugh. The obtained coefficients on trade costs are consistent with the gravity literature, where distance and tariffs are an impediment to trade. The magnitudes of the coefficients reported in Table 4 are similar to those in Eaton and Kortum (2002) and in Waugh (2010), which consider a similar sample of countries without tariffs. The overall size of the trade costs in terms of percentage are similar to those reported in Anderson and Van Wincoop (2004). Next, I feed the model with the estimated trade costs and technology.9 Table 5 presents the mean concentration levels for the simulated countries. The results show that the calibrated model replicates the fact that countries are more specialized in exports than in imports on all margins. Focusing on the obtained levels reveals that countries concentrate excessively on exports with re8 Given 9 factor endowments and optimal factor choice, the interest rates equals: ri = a/(1 See Table 7 at the end of the paper. 20 a)wi ( Li /Ki ) spect to the data. The simulated concentration levels are almost twice as high as the ones observed in the data. Mean export (import) concentration on the extensive margin is 4.83 (0.84) compared to 2.60 (1.10) in the data. This implies that in the simulated model countries export (import) 0.8 (43.2) percent of the product space compared to 7.4 (33.3) precent in the data. Figure 4 plots the corresponding cross country pattern for simulated and empirically observed concentration levels against the log of GDP. The model replicates the empirical pattern with export concentration decreasing in market size. However, the simulated concentration levels on the extensive margin are too high, particularly for smaller economies. Countries specialize excessively on the number of products exported. On the importing side, the calibrated model is unable to replicate the L shape relationship between market size and concentration. The relationship does not reveal any particular pattern. However, simulated countries tend to import more goods than in the data. Turning the attention to the intensive margin, Figures 4(e) and 4(f), the results show that, consistent with the data, the model predicts no relationship between concentration and market size. In sum, the calibrated model is able to replicate the qualitative pattern for exports but produces relatively high levels of concentration compared to the data, particularly on the extensive margin. A potential explanation for the excessive concentration in exports lies in the underlying productivity distribution. While the model reproduces the bilateral trade volumes, it fails to capture the underlying distribution of trade volume across products. To shed light on why countries trade too few products, I follow Haveman and Hummels (2004) and plot the empirical and the simulated density of the number of exporters and importers per product.10 Figure 5 shows the results. In the case of exports, simulated countries export their goods to too many destinations. The assumed productivity distribution generates such efficient producers that even firms facing high trade costs can sell their products to many destinations in the world. As a consequence, the number of exporting countries per product is small. In the data (in blue) more than a third of the products are exported by 25 or more countries. In the simulation (in red) no product is exported by more than 25 countries. Turning the attention to imports, Figure 5(b) shows that, contrary to exports, the simulated distribution of the number of countries importing a product is closely related to the empirical one. 5.1 Discussion of results There are several potential reasons why the model is not able to reproduce the cross country pattern of import concentration on the extensive margin. Note that the model implies that expenditure shares equal to product shares in the tradable sector. The product share, pi , is defined as 10 To get the empirical distribution of the number of exporters and importers per product, I count for each HS code the number of countries that net export or net import the product. Similarly, the model implied distribution represents the number of exporters and importers for each simulated product. 21 the number of products imported, Ni,M , divided by the total number of potential goods, N, i.e. number of HS codes : pi = Ni,M N (16) and the expenditure share, mi , equals the total value of imports, Mi , divided by domestic absorption, Absi . mi = Mi Absi (17) Figure 6 plots the empirical relationship between the two shares. The red line marks the 45 degree line where the two are equal. Notice that countries below the 45 degree import a lot of goods and spend relative little on those goods, whereas countries above the 45 degree line import few goods and spend a lot on them. One potential reason of why countries are not aligned along the 45 degree line can be that countries differ in the number of intermediate goods used in production. When calculating the share of goods imported, I divide the total number of net products imported by the total number of HS codes, which is common to all countries. If countries do not make use of all tradable goods (for example they do not have the underlying technology to use a particular intermediate good), then the calculated import product share for these countries is downward biased. Ethier (1982) argues that larger economies use a higher number of intermediate goods because of increasing returns to scale in the production process of the final good. To shed light on the potential role of market size on the number of intermediate products in the economy, we impose equality between product shares and expenditures shares, (pi = mi ). Given this assumption, we can rewrite this equation as: Mi Absi = Ni,M N (18) implying that the average per product import expenditure equals the average per product tradable expenditure. Since the number of tradable goods is the same for all countries, we expect that the elasticity of the average per product import expenditure with respect to absorption is 1. Figure 7 plots the relationship. Note that the figure reveals a strong positive correlation with a R2 = 0.84 and an elasticity of 0.6, significantly different from 1. Ethier’s argument that larger economies have a higher degree of specialization and use a larger number of intermediate inputs in the production of tradable goods can explain why the elasticity is below 1. In this case, the number of tradable goods would be country specific and increases with the size of the tradable sector. Non-homothetic preferences may represent an alternative explanation for the fact that some 22 countries spend, on average, relatively more on few imported goods. Note that according to equation 18, the ratio of the per product import expenditure with respect to per product tradable expenditure should be one. This result relies on the assumption of homothetic preferences. Figure 8 plots the log of the ratio against the log of GDP per capita. The figure shows a negative correlation of -0.67 with a R2 = 0.23. This evidence is consistent with non-homothetic preferences, where poorer countries spend relatively more per imported good compared to rich ones. 6 6.1 Robustness Alternative classification schemes This part addresses concerns on the robustness of the empirical observed concentration indexes. In particular, the level of disaggregation as well as the classification scheme chosen may affect the empirical concentration measures and the decomposition of the intensive and extensive margin. For this reason, I re-calculated the concentration indexes on all margins by defining a product as the equivalent of (1) a 4 digit SITC code and (2) a 6-digit NAICS code instead of a 6 digit HS code. The advantage of the NAICS and SITC classification system is that products are grouped according to economic functions such as their material and physical properties rather than for tariff purposes as in the HS system. Table 6 reports the calculated concentration indexes based on SITC and NAICS together with the correlation with respect to HS based based concentration indexes. The qualitative estimates all classification are very similar. Exports are more concentrated than imports. Concentration is driven by the extensive margin for exports and on the intensive margin for imports. In terms of cross country evidence, larger countries import and export more goods. Strikingly, the L pattern of the extensive margin also appears when using the SITC and the NAICS classification. Differences between the various classification schemes appear in the levels of import concentration. The reason is that the total number of 4 digit SITC codes is 642 and of NAICS codes 460, significantly lower than the 4529 HS codes. However, overall the high correlation across the different classification highlight the level of generality the results apply. 6.2 Intra-industry trade In this paragraph I want to address the discrepancy of the product space between the data and the model caused by intra-industry trade. In the main part of the paper I establish correspondence between the model and the data by netting out within product trade. This approach leaves valuable information unused and may bias the results. In an alternative approach, I deal with intra-industry trade by developing a “measurement device” that enables the model to characterize intra and inter industry trade. The basic idea is that in reality the true state of the world is indeed Ricardian, i.e. varieties are in fact products, but the data are not sufficiently disaggregated to capture the true product level. Instead, these “Ricardian products” are aggregated into sectors 23 Table 6: Mean concentration indexes for gross trade flows based on the Armington assumption: 160 countries Gini Mean index (HS 6 digit) % share of overall concentration Exports Imports 0.98 0.9 Theil Exports (X) Extensive Intensive Total Margin Margin 1.81 2.59 41% 59% 4.40 Theil Imports (M) Extensive Intensive Total Margin Margin 3.53 2.78 56% 44% 6.31 according to a classification scheme, i.e. HS codes. The suggested procedure converts the measurement of product units in the model to product units in the data and allows to examine gross trade flows. Because the classification scheme is unobserved, I assume that varieties are randomly assigned to an HS code following a Poisson process. Using the structure of the model, I can then estimate the Poisson parameter and characterize the “measurement device”. I obtain a value of 0.94 for the Poisson parameter implying that, on average, one “Ricardian product” comprises an HS product category. Based on this result, I group simulated Ricardian products randomly into artificial HS codes and calculate the implied concentration indexes. The results, presented in detail in the appendix, show that this approach leads to similar results as the net trade flow approach. 6.3 Implications of alternative trade models Finally, I want to relate my analysis to alternative trade models, in particular, to monopolistic competition models based on Krugman (1980) and Armington models based on Anderson and Van Wincoop (2003). The key difference with respect to the Ricardian model is that in both types of models tradable goods are differentiated by location of production. Applying this definition of the product space to the data implies that each country is the sole producer/exporter of an HS codes and demands all country product combinations. Hence, the number of potential goods exported is 4529 and the number of potential goods imported is 4529 times 159 trading partners. Table 6 presents the corresponding concentration indexes. The results shows that, conversely to the above results, countries are more specialized in imports than in exports and the extensive margin drives the import concentration. The reason is that a country imports only 3 percent of the products its demands, which equals 4529 products times 159 countries. This implies that the empirical implications used to evaluate a model depend on the definition of the product space, i.e. differentiated versus homogenous goods. While it is certainly possible to generate the results of Table 6 within a model based on the Armington assumption, the underlying mechanism to generate speculation will be very different.11 In this paper, the analysis is based on the assumption 11 For example by introducing fixed costs of trade (see Romer (1994)) or declining marginal utility of varieties (see 24 that foreign varieties are perfect substitutes to domestic ones consistent with the Ricardian model. One motivating observation is that the Grubel and Lloyd (1975) index of 0.19 indicates that the majority of the trade flows is inter-industry (81 percent) rather than intra-industry. However, I cannot reject these alternative hypotheses for the observed concentration patterns and would like to pursue them in future research. 7 Conclusion I have argued that export and import concentration in combination with a decomposition into an extensive and intensive product margin concentration measure provide new quantitative and qualitative evidence on specialization patterns in world trade. Based on detailed trade data, the calculations show that exports are more concentrated than imports on all margins and specialization is dominated by the extensive product margin for exports and by the intensive product margin for imports. The extensive product margin explains the gap between export and import concentration and drives specialization differences across countries. Larger economies diversify more because they export and import more products. Furthermore, I show that the Eaton Kortum model is consistent with the observed patterns and partly replicates the stylized facts as well as the cross-country differences qualitatively and quantitatively. Overall, my results stress the importance of the role that geography and absolute as well as comparative advantage play in determining the pattern of specialization. By looking through the lenses of export and import concentration, this paper analyses how openness to trade changes the production structure of an economy and how these changes relate to income. My results show that the relationship between income and concentration is primarily driven at the extensive margin. This relationship has important macroeconomic policy implications. Specialization increases a country’s exposure to shocks specific to the sectors in which the economy concentrates. As a result, the likelihood that product specific shocks have aggregate effects in terms of output volatility and/or a negative impact on the terms of trade increases with openness. Diversifying along the extensive margin reduces such risks whereas specialization along the intensive margin by exporting industries that have already a comparative advantage to more destinations intensifies the risk. Analyzing this question is an avenue for future research. Ottaviano et al. (2002)) 25 Figures Gini index − HS 6 digit Total Theil index − HS 6 digit Mean of the index from 1992−2009 Mean of the index from 1992−2009 6 4 2 Export Concentration .95 .9 .85 ITA GER NGA TCD YEM OMN GAB COG IRN KWT GNB SAU QAT MLI COM ARE BFA WSM BWA AZEBDI RWA KIR BMU BEN SDN MWI GIN STP CAF VEN DZA ZMB MRT NER VUT GMB CUB JAM ECUSYR MOZ PNG ETH UGA SLEBHS ATG CMR GUY CPV ERI GHA PRY MNG NOR TTO TGO ARM BHR KAZ DJI BOL ISL BLZ MLT BRB KGZ TZA MUS CRI COL ZWE FJI GEO SEN NIC KHM RUS PER PAN KEN SWZ CHL MDGNPL JOR HND EGY SLV LBN LVA GTM BGD CYPPHL ALB VNM AUS DOM IRL ARG LTU MDA BIH ISR MEX URY AFG MYSBLR CAN NZL MKD TUN MAR ZAF GRC LKA EST PAK BRA FIN UKR SRB HRV SVK KOR PRT HUN IDN ROM BGR DNK SWE THA ESP SVN IND TUR POL CHE FRA GBR CZE BEL JPN AUT NLD USA CHN GER ITA 0 .75 .8 Export Concentration 8 TCD GAB KWT COM YEM COG GNB QAT BDI KIR RWA MLI MRT WSM DZA OMN GIN BMU CAF BEN BWANGA VUT SDN SAU ARE STP MWI BFA PNG NER CUB GMB ZMB AZE CPV MOZ BHS CMR GUY IRN ETH JAM ERI GHA VEN ECU UGA TTO PRY MNG TGO BHR BOL ISL DJI SLE ATG BRB ARMATG ARM SEN KAZ TZA BLZ NIC MLT MUS KGZ FJI KHM SYR GEO MDG ZWE HND JOR PER NORCOL CRI SWZ CHL BGD KEN SLV LBN DOM PAN ALB GTM CYP MDA NPL URY RUS LVA EGY IRL BIH AUS ARG NZL LTU MARVNM MKD PHL GRC LKA BLR AFG PAK MYS MEX TUN EST SRB CANHRV ISR ZAF FIN PRTROM BRA UKR SVK HUN BGR KOR DNK IDN POLSVNSWE TUR THA ESPCHE AUT CZE BEL NLD GBR IND FRA JPN USA CHN 1 8 .75 .8 .85 .9 .95 1 0 2 4 Import Concentration (a) Gini coefficient 8 (b) Theil index Mean of the index from 1992−2009 6 Extensive Theil index − HS 6 digit Mean of the index from 1992−2009 6 Intensive Theil index − HS 6 digit 4 0 PAN KIR TCDSTP GNB WSM ERI CPV BDI VUT RWA BMU BENGMB CAF DJI GAB COG MRT ATG BFA GIN MWI SDN BWA BHS PNG MLI MOZNER ETH DZA GUY BRB CUB UGA QAT YEM MNGARM BLZ TGO ZMB KWT CMR JAM SEN AZE NGA GHA PRY NIC SLE BOL KHM TZA BHR FJI ISL TTO OMN ALB GEOKGZ MLT MDG LBN NPL KAZ MUS CYP AFG JOR SAU SLV VEN ECU HND DOM SWZ BIH MDA BGD URY GTM ZWE CRI MKD KEN PAN GRC LVA IRN PER HRV LKA SRB CHL MAR SYR LTU TUN ARE BLR EST NOR EGY COL NZL VNM ARG PRT ISR IRL PHL ROM AUS PAK UKR FIN MEX SVK SVN CAN BGR HUN DNK POL RUS MYS TUR AUT CHE SWE ESP CZE BRA BEL ZAF GBR IDN THA KOR NLD FRA USA IND JPN ITA GER CHN 2 4 2 Export Concentration COM IRN ARE NGA OMN SAU YEM SYR KWT VEN NOR QAT ECU MLI AZE RUS GAB COG BWA COL ZMB BFA DZA JAM TCD SLE KAZ SDN MWI CRI TTO GIN ZWE CHL PER CMR NER CUB EGY GHA PRY KEN AUS MRT MOZ IRLMYS BEN VNM BHR MLT PHL ETH UGA MEX PNG MUS ARG ISL BRA ISRZAF BDI GUY CAF SWZ CAN BOL MNG BHS LVA BMU KGZ RWA GNB WSM TGO KORIND JOR HND NPL GTM ARM TZA LTU IDN GEO FJISLV BGD BLR GMB ATG MDG NZL PAK FIN THAJPN HUN FRA UKR BLZ SVK KHM NIC SEN TUN URY MDA LBN CYP SWE DOM ESP MAR STPBIH GBR PRT VUT NLD BRB EST CHE DNK TUR CHN MKD LKA BGR GRC USA CZE BEL ALB POL GER ROM COM SVN SRB KIR HRV AFG DJI AUT ERI CPV ITA 0 Export Concentration 6 Import Concentration 0 2 4 6 0 Import Concentration 2 4 6 Import Concentration (c) Intensive margin (d) Extensive margin Figure 1: Average export versus import concentration for the period 1992 to 2009 for 160 countries 26 6 4 Import Concentration USA PLW STPKIR −10 −5 0 −10 Log of GDP relative to the US USA −5 0 Log of GDP relative to the US R2=0.58 R2=0.41 (a) Overall concentration of exports (b) Overall concentration of imports Extensive Theil index − HS 6 digit Mean of the log index from 1992−2009 Mean of the log index from 1992−2009 6 Extensive Theil index − HS 6 digit 6 COM Import Concentration USA 0 4 KIR STP VUT AFG GNB ERIGNQTCD COM WSM SLE DJI CAF ATG HTI BMU LAO IRQ ZAR ARM RWA GMB BDI MRT KGZ TKM CPV NER COG BEN MNG NPLUZB MWI SYC SUR BFAGEO TGO MOZ MLI KHM AGO FJI AZE ZWE BRB BLZ YEM BIH SWZ MDA ALB LBY ZMB UGA QAT CMR SEN SDN BHR SYR KAZ CIV MDG ETH GERCHN CUB TZA BLR GHA NIC EST ITA PAK IND JPN LVA KEN MKD KWT BGD DOM MLT OMN LTU HND MUS BGR JAM VNM TTO UKR PRY PAN LKA URY BOL SLV JOR LBN IRN KOR NGA ISL CYP FRA SVN NLD THA IDN CRI CZE SVK GBR LUX BRA SWE CHE BEL GTM ZAF AUT POL RUS TUR PHL MYS HUN TUN ROM PER ISR ECU ARE ESP DNK MAR EGY FIN SRB ARG COL HRV IRL DZA NZL PRT MEX NOR SGP VEN CHL GRC SAU AUSCAN USA 0 2 PLW 4 GNQTCD BDI ERI RWA AGO IRQ BMU BEN GMB SYC DJI CAF COG BFA SDN MRT MWI ZAR ATG LBY DZA SURNERMOZ TKM MLI UGAETH YEM CUB ZMB BRB HTI JAMQAT KWT NGA MNG BLZ TGO LAO ARM AZE CMR GHA SENPRY TZA NIC BOL KGZ ISL SLE TTO OMN KHM FJI BHR MDG UZB GEO ALB CIV LBN ECUKAZ MLT CYP NPL JOR VEN SAU HND MUS SLV DOM AFG SWZ MDA BIH URY ZWE BGDGRC IRN CRI GTM PER KEN CHL MKDPAN LVALTU LKA HRV MARARE SRB SYR EGY TUN NOR EST LUX BLR NZLVNM COL ARGAUS SGP PRT ISR IRL PHL ROM PAK UKR MEX CAN FIN SVK HUN BGR SVN POL DNK RUS TUR ESP MYS AUT BRA ZAF SWE CZE CHE BEL IDN GBR THA NLD KOR FRA IND JPN ITA GERCHN GNB VUT WSM CPV STP 2 PLW KIR Export Concentration AFG VUT BMU PAN ATG GNQ GNB COM ERI TCD WSM IRQ ARM SLE CAF HTI DJI MRTNER KHM BEN GMB SYC BDI TKM MLT GEO LAO PHL RWA IND KGZ COG BFA TGO CPV ZAR UZB MWI AGO MLI MOZ NPL YEM AZE QAT MYS SURMNG OMN BHR CIV SEN ZWE CYP SDN ZMB CHN CMR LBY SWZBRB BLZ ALB UGA FJI ITA GERJPN KWT PRY BIH ISR MDG TZA SGP CHE KAZ MDA GHA LUX ETH PAK KEN THA LBN TTO SYR IRL NIC BGD NLDKOR HND ARE DOM JOR DZA BEL JAM GBR EST VNM SLV HUN BOL CUB LKA NGA ISLMKD MUS LTU UKR BRA LVA SAUIRN BLR ZAF SVK EGY ECU RUS NZL CRI IDNCAN GRC BGR AUT SWE DNK URY NOR MEX CZE FIN TUN AUS GTM COL TUR VEN SVN PER ESP FRA PRT MAR ARG CHL SRB POL HRV ROM 0 2 4 STP AGO GNQ SUR MOZ MLIZAR BMU CAF BDI MRTNER ZMB JAM VUT COG GMB TCD GNB SLE ERI RWA CMR SDN CPV ATG BEN IRQ GHA KWT BFA TTO CUB QATLBY ARM KGZ UGA ETH MNG ISLSENTKM TGO BOL DZA TZA LAO DJI YEM PRY MLT UZBECU PER BHR MWI NGA BLZ CHL NIC HTI ZWE BRB MDG NPL AZE KHM KAZ JOR MUS CIV GEO CRI LBN HND SWZ FJI SGP SAUIRN ALB OMNDOM ARE BGD CYP PHL ARG VEN MDA URY IRL SLV PAN ISR KENGTM BIH LVA LUX NZL ZAF AUSCAN MYS MAR MKD AFG GRC MEX TUN BLR VNM NOR SYR PAK LKA RUS COL EST LTU FIN UKR SRB EGY HRV PRT SVK HUN BRA KOR ROM THA SWE DNK BGR TUR ESPFRA SVN POL IND CHE IDN CZE GBR BEL AUT JPN NLD GERCHN ITA COMWSM SYC 2 6 KIR PLW 0 Export Concentration 8 Total Theil index − HS 6 digit Mean of the log index from 1992−2009 8 Total Theil index − HS 6 digit Mean of the log index from 1992−2009 −10 −5 0 −10 Log of GDP relative to the US R2=0.75 (d) Extensive margin of imports Mean of the log index from 1992−2009 4 STP KIR PLW BMU PHL MLT IND MYS CYP SGP ATG ISR OMN IRL LUX AFG DZA SAU KHM ARE CHE PRY GRC QAT THA CIV LBN KWT BHR BEL ITA NLD NZLHUN VUT GBR BGD AUS SYC SEN BEN NOR EGY TTO ARM MEX IRQ BRA ECU CHN VEN KEN ZAF PAK YEM GEO JOR RUS AGO SDN CAN DNK IRN KOR ISL NGA COL FIN SVK NER SLV AUT CMR PRT HND BOL GHA TZA LKA TUN CHL SWE AZE JAM CRI MDGUGA DOM GTM ZMB IDNESP GERJPN VNM BFA ARGTUR TGO MUS KAZ PER CZE ETH NIC MRT UKR MAR MLI LBY LTU SRB HRV UZB SWZ ZWE SYR ALB MOZ HTI URY GMBBLZ MKD POL SVN EST MWI BGR COG LVA TKM GNQ FRA BDI KGZ CPV NPL CUBROM MNG CAF BIH RWA BLR MDA SURBRB COMWSM LAOZAR DJI FJI SLE ERI TCD GNB USA 0 0 1 USA PAN 3 ZMB JAM SUR MOZ SLE NERMLI PER SGP CHL TTOAGO ZAR SYC PHL ZAF ZWE KGZ CMR ECUIRL MRT MLT ARE ISRMYS ARG CAF ISL GHA CAN CRI IRN UZB KWT SWZ ARMBHR NPL AUS WSM BMU COGMUS KOR JOR QAT RUS MEX BOL IND KAZ TZA HND SAU PRY SENPAN MNG CIV GMB GNQ VNM BGD NZL FIN BRA PAK LUX CUB HUN FRA MDG THA ESP LBNSDN UKR LVA LAO JPN URY KHM TGO MDA NICGEO NGA UGA ATG KEN SWE SVK BLR MAR DOM TUN NOR BDI LBY PRT GTM VEN CYP COL BLZ CHE FJI SLV GBR BIH AZE TUR IDN MKD DNK EST ETH BFA HTI SYR YEM CZE BEL ALB LTU GRC GERCHN DZA OMN BGR POL VUT SVN EGY LKA ROM NLD BRB TKM SRB AUT HRV IRQ BENAFG KIR COM MWI ERI TCD ITA CPV PLW DJI RWA GNB STP 2 Import Concentration 5 Intensive Theil index − HS 6 digit Mean of the log index from 1992−2009 5 Intensive Theil index − HS 6 digit 4 3 2 Export Concentration 0 R2=0.54 (c) Extensive margin of exports 1 −5 Log of GDP relative to the US −10 −5 0 −10 Log of GDP relative to the US −5 0 Log of GDP relative to the US R2=0.01 R2=0.12 (e) Intensive margin of exports (f) Intensive margin of imports Figure 2: Average export and import concentration versus the log of average relative GDP with respect to the United States for the period 1992 to 2009 for 160 countries. 27 Total Theil Index Total Theil Index Data versus simulation Data versus simulation 12 8 Red − Simulated Data Blue − Empirical Data 11 Import Concentration Export Concentration 9 GNB 8 AGO VUTBLZ STP GNQ KIR COM WSM SYC ERI SUR PLW MOZ ZAR MLI KIR BMU NER CAF ZMB BDI PLW MLI COM MRT MDA MKD MOZ JAM VUT BEN GMB COG RWA TCD AZE ATG UGA WSM CPV GNB GMB CIV SLE TKM SDN ARM SYC DJI ERI RWA GEO CMR CPV ATG TTO BEN YEM GHA IRQ KWT BFA BMU CUB ZWE MRT JAM QAT LBY ARM KGZ STP UGA TZA QAT FJI CAF BDI SLE ETH GHA MNG ISL TUR RUS LKA SWZ TUN LBY TGO BOL DZA TZA LAO SEN SVKPER DJI SUR TKM DNK YEM NER PRY ECU MLT UZB GNQ COG BHR NOR MWI NGA MNG SGP MLT THA BLZ CHL TCD NIC HTI KAZ BRB ROM NGA BFA PAK LAO KGZ ZWE BRB MUS ALB HTI NIC ZAR MDG NPL AZE KHM KAZ ISL GRC ZMB JOR SEN CZE MUS CIV AFG GEO EST BHR NPL CRI PAN LBN HND SWZ FJI URY IRN BIH LVA SGP SAU CMR JOR CYP AGO LUX UZB ARE TTO PRY ALB SLV SVN DOM OMN KEN BOL KWT LTU ETH BGD CYP SDN LBN GER HRV DOM VEN PHLARG SYR CRI BGR MDA BLR IRQ URY IRL GTM SLV PAN IDN BRA MAR ISR AUS CUB ECU IRN KEN GTM BIH SRB NZL BEL SAU ARG LVA ARE VEN NLD POL MEX LUX FIN CAN PHL UKR VNM PRT BGD CHE NZL ZAF AUS SWE ITA PER AUT MYS MAR ESP CHL COL ISR KOR FRA IRL MKD EGY CAN VNM AFG TUN GBR GRC HUN BLR NOR SYR PAK MEX LKA RUS COL EST LTU FIN IND UKR SRB EGY HRV PRT SVKHUN BRA JPN KOR THA SWE DNK BGR ROM TURESP SVN FRA POL IND CHN CHE IDN CZE GBRJPN BEL AUT NLD USA GERCHNUSA ITA 7 6 5 4 3 2 1 −6 10 −4 10 −2 10 Red − Simulated Data Blue − Empirical Data 7 10 6 AFG VUT BMU PAN PLW ATG KIR GNQ STP COM GNB WSM VUT GNB KIR COM STP ATG ERI TCD CPV GMB BLZ SYC WSM IRQ DJI CAF BMU ARM BDI ZWE SLE HTI GNQ ERI MWI BRB DJICAF FJI MUS NER MRT KHM BEN GMB SUR SYCBDI COG MDA RWA TKM TGO MNG SEN SWZ MLT GEO LAO PHL RWA IND HTI ALB TCD ZMB GEO KGZ COG BFA MLT BEN TGO CPV ZAR LAO TTO MOZ UZB MWI BFA AGO MLI ARM AFG ISL MOZ NPL CYP MNG NIC YEM MDG AZE QAT KHM MYS BIH JOR SUR MKD EST OMN BHR CIV SEN UZB ZWE PRY NPL CYP USA BLR HND UGA SDN ZMB GHA AZE URY CIV CMR BOL PAN AGO TZA LVA JAM LBY SVN KWT SVK SWZ YEM ETH BLZ ALB UGA KEN SDN FJI ITAGERCHN KWT LUX CRI LBN PRY SLV LTU BIH QAT ISR MDG BRB TZA SGP OMN KAZ SYR MDA GHA LUX ETH DOM PAK SRB LKA KEN JPN BGR THA LBN SYR LBY IRL TUN NIC TTO ECU GTM BGD CHE HRV NLD HND ARE KAZ HUN DOM IRQ JOR DZA BEL MAR JAM PER PHL GBR EST VNM SLV HUN IRN BOL CUB KOR COL LKA NGA IRL ISL DZA MUS LTU UKR BRA VNM ISR CHL MKD LVA SAU BLR NGA ZAF ARE UKR SVK EGY DNK ECU RUS FIN ROM NZL CZE CRI IDN GRC BGR SGP MYS AUT DNK SWE VEN SAU USA CHE URY NOR PAK PRT CZE MEX FIN TUN AUS GRC GTM SWE COL TUR VEN BRA BEL CAN SVN JPN ZAF PER ESP RUS FRA PRT MAR KOR IRN NLD AUS CHL GBR THA SRB TUR POL ARG GERCHN HRV CAN ROMARG IND IDN ESP FRA ITA MEX 5 PLW 4 3 2 1 −6 10 0 10 Log of GDP relative to the US −4 (a) Overall concentration of exports 0 10 Extensive Theil Index Data versus simulation Data versus simulation 6 6 Red − Simulated Data Blue − Empirical Data STP COM PLW PLW KIR STP 4 3 2 1 0 −6 10 Red − Simulated Data Blue − Empirical Data 5 GNQ GNB TCD COM VUT BDI RWA WSM AGO CPV ERI GNB BMU IRQ BEN GMB SYC CAF DJI WSM COG ZAR SDN BFA VUT MWI ATG CPV MRT ETHLBY DJI SUR GMB MOZTKM SYC NER DZA MLI UGA CUB YEM ZWE BMU CAF BLZ BDI SUR MRT ERI SLE SWZ FJI ZMB QAT BRB HTI MNG BLZTGO LAO JAM ARM COG AZE KWTNGA GNQ NER CMR GHA SEN MNG TZA MWI RWA NIC BRB BEN TCD BOL PRY ARM LAO MLT BFA MOZ MDA MLI KGZ NIC HTI MUS ALB KGZ ZAR ISL AFG SEN SLE TTO MDG KHM FJI BHR MKD MDG GEO ZMB OMN EST UZB ALB GEO BHR BIH TKM UGA NPL KAZ CIV GHA TTO LBN CMR LVA PRY PAN JOR JAM MLT CYP HND TZA UZB YEM CYP AGO ECU VENSAU URY BOL LUX AZE NPL JOR LTU ETH HND QAT MUS OMN SVN SLV KEN LBN AFG DOM CRI SYR LBY KWT TUN SDN BGR DOM IRQ LKA HRV BLR GTM SVK CUB SWZMDABIH ZWE MAR BGD IRN SRB URY ECU CRI KEN KAZ NZL VNM CHL GRC MKD BGD CZE ROM PER PHL NGA ZAF PRT ARE BEL PAK ARG DNK FIN UKR SGP SWE NOR IRL PAN THA LVAHRV LKA CHE VEN AUT PER CHL MYS MAR SRB ISR DZA EGY COL POL SAU SYR HUN IRN TUR NLD GRC IDN MEX LTU EGY BRA TUN ARE AUS NOR RUS GBR ESTLUX ITA BLR NZL ESP COL FRA KOR GER CAN IND ARG VNM AUS SGP PRT ISR IRL PHL ROM PAK CAN UKR MEX FIN SVKHUN BGR SVN POL DNK RUS JPN CHN TURESP MYS AUT BRA ZAF SWE CZE CHE BEL IDN GBR THA NLD KOR FRA USA IND ITA JPN GER CHNUSA −4 10 −2 10 Import Concentration 5 Export Concentration 10 (b) Overall concentration of imports Extensive Theil Index PLW 4 KIR STP 3 2 1 0 −6 10 0 10 Log of GDP relative to the US VUT GNB −4 8 Red − Simulated Data Blue − Empirical Data 7 Import Concentration 5 0 −6 10 GNB BLZ MKD AZE MLI MDA TUR RUS VUT ERI MOZUGA DZA CIV ZMB BEN TKM SURRWA DNK MOZ JAM YEM MLI NORTHA SLE NER GEO LKA SGP QAT PER TUN AGO JAM SVK LBY SGP CHL TZA PAK NGA TTO ROM GRC KAZ ARM GHA ZAR SYC PHL ZAF ZWE CZE KGZ IRL CMR ECU MYS MRTMLT ARE ISR ARG CAN GER CAF ISL GHA KWT CRI IRN UZB BMU SWZ ARM BHR GMB NPL AUS WSM COG MUS KOR JOR IDN QAT KAZ SAU RUS MEX BRA BOL SYC ATG IRN CPV IND PAN HND TZA MRT PRY SEN DJI BMU MNG CIV GMB GNQ KOR VNM FJI NZL BGD DOM ZWE CAN NLD FIN BRA PAK SAU LUX MEX CUB ITA MWI HUN MLT URY SLE FRA ESP PAN MDG NER THA KWT BDI TGO NPL POL LBN WSM MNG UKR MUS LVA ALB LAO JPN VEN URY SUR BEL GNQ HND ZMB BFA HRV HTI ARG KHM SVN KGZ GEO COG KEN BHR SLV LUX TCD SDN ZAR SWZ ARE TGO EST ESP LVA ISL AGO NIC GTM MDA NGA LAO BGR GBR UGA UZB BRB IRQ NZL MDG MYS ATG CAF FIN KEN SWE JOR SVK SEN CMR BLR CHE ECU LTU SYR UKR LBN CUB MAR OMN PER DOM COL AFG AUT TUN BOL ETH CRI ISR PHL NOR PRT BDI SRB CHL EGY SWE BIH LBY ZAF TUR TTO BGD INDJPN PRY GTM VNM VEN IRL CYP HUN COL BLZ CHE FJI SLV BIH IDN GBR KIR COM MKD DNK EST ETH BFA STP HTI SYR YEM CZE BEL ALB LTU GRC CHNUSA GERCHN AZE DZA PLW OMN BGR VUT SVN POL EGY LKA NLD BRB TKM SRB ROM AUT HRV BENAFG KIR COM IRQ MWI ERI TCD ITA CPV PLW RWA GNBDJI STP −4 10 −2 10 Red − Simulated Data Blue − Empirical Data 7 6 1 0 10 Data versus simulation 8 2 10 Intensive Theil Index Data versus simulation 3 −2 10 (d) Extensive margin of imports Intensive Theil Index 4 AFG ERI TCD GNQ COM GNB PLW KIR COM WSM STP VUT SLE ATG DJI CPV GMB ATG CAF SYC DJI BLZ CAF ZWE BMU HTI BDI SLE LAO IRQ ZAR ARM RWA GMB ERI FJI GNQ MRT COG SUR NER TGO KGZ TKM CPV BDI SWZ NER COG MNG RWA BRB TCD MWI KGZ BEN MNG MDA NPL MWI LAO SYC SUR BFA TGO GEO MLT MLI HTI MOZ MLI BFA ARM KHM UZB ALB FJI MUS NIC MDG ISL ZMB ZAR AFG MKD SEN GEO AGO BIH AZE KHM ZWE MDA BLZ BHR TTO EST YEM PRY BIH TKM JOR HND NPL GHA CYP PAN BRBALB UGA SWZ CIV AGO CMR AZE LBY ZMB LVA TZA ETH JAM BOL UGA URY KEN QAT CMR SEN SLV SDN LUX YEM UZB CRI LTU LBN BHR SYR KAZ CIV MDG QAT ETH SDN OMN GERCHN CUB TZA BLR LKA BGR GHA SVN NIC EST DOM LBY GTM TUN HRV ITA BLR PAK LVA INDJPN ECU KEN MKD IRQ KWT DOM MLT SRB LTU OMN KAZ HND MUS BGR JAM VNM CUB UKR TTO PRY HUN SVK ISR KWT USA PAN PER LKA URY IRL BOL SLV JOR MAR BGD DZA LBN IRN KOR NGA ISLCYP FRA SVN NZL NLD THA VNM CHL IDN FIN CRI NGA ARE UKR CZE USA SVK GBR LUX ROM MYS BRA DNK SWE CHE BEL ZAF GTM COL SGP AUT POL RUS PHL EGY TUR PRT VEN MYS NOR HUN TUN ROM ISR PER ECU ARE ESP GRC DNK AUT PAK MAR EGY SWE FIN SRB ZAF ARG COL HRV JPN IRL BEL DZA IRN CHN THA NZL NLD PRT MEX POL CAN NOR TUR ARG IDN KOR SGP VEN CHL SAU AUS GER IND GRC SAU CAN ESP FRA GBR ITA MEX RUS BRA Log of GDP relative to the US (c) Extensive margin of exports Export Concentration −2 10 Log of GDP relative to the US 6 5 3 2 1 0 −6 10 0 10 Log of GDP relative to the US PAN 4 BMU PHL MLT IND MYS CYP SGP ATG ISR OMN USA TKM IRL LUX KWT BLR SVK SEN AFG DZA KHM SAU MUS ARE CHE PRY GRC QAT BRA THA CIVLBN BHR MWI GEO KWT ZMB RUS BEL UZB TTO HUN CYP ITA BRB NLD ALB SAU NZL PHL COL GBR PLW VUT SVN BGD AUS SYC SEN BEN NOR MDA EGY TTO ARM SRB HTI IRQ MEX ZAR BRA ECU VEN CHN GNQ ZAF KEN PAK YEM GEO JOR RUS MAR AGO SDN CAN DNK GER IRN HUN KOR ISL NGA AUT COL EGY FIN NER SVK UKR JPN SGP AUS BEL PER ARE VEN SLV ROM NOR ECU CMR PRT CHE HND BOL URY CZE GRC GHA DZA VNM AFG TZA YEM KAZ TUN LKA QAT IRL CAN CHL AZE SWE ZAF CHNUSA MYS MEX ITA FRA JAM LBN GER BLZ CUB RWA LBY NLD CPV OMN CRI LTU MDG IND BDI VUT MLT NZL BMU IRN GTM DOM ZMB ISL LUX TUR IDN HRV IRQ BFA VNM POL KHM SYR ARG TGO GMB ESP NER ATG SYC TZA KEN BGR THA KAZ PER MOZ CZE CMR ETH DJI NPL MNG NIC MRT UGA FJI SUR EST LVA CIV UKR ERI BIH ARM MAR SWZ SLV TCD MKD ISR AGO MUS WSM SLE CAF MLI HND LBY MDG BHR BEN GNB PRY COG GHA LTU TGO SRB KGZ HRV LAO PAN UZB ZWE SWZ SYR ALB MOZ HTI KIR COM URY STP GMB MKD POL SVN EST ROM MWI BGR COG TKM LVA GNQ FRA BDI KGZ CPV NPL BRB CUB MNG BLZ CAF BIH RWA BLR MDA SUR COM LAO WSM FJI DJI ZAR STP SLE ERI TCD KIR GNB PLW −4 10 −2 10 0 10 Log of GDP relative to the US (e) Intensive margin of exports (f) Intensive margin of imports Figure 3: Simulated (in red) and empirical observed (in blue) export and import concentration versus GDP across 160 countries. The simulation uses parameterized trade costs to match the data using a country specific export cost. 28 Total Theil Index Total Theil Index Data versus simulation Data versus simulation 20 ATG 18 8 Red − Simulated Data Blue − Empirical Data Red − Simulated Data Blue − Empirical Data 7 PLW KIR 14 Import Concentration Export Concentration 16 PLW STP BMU GNB CPV ERI BDI SYC TCD SEN UZB WSM COM SLE GEO AZE MNG IRQ CAF SDNNGA CUB VUT ARM BIH MLI NER TKM ALB NPL MRT MUSMLT BRB FJI AFG MKD ZARSUR DOM DJI MWI LBY DZA ECU ETH JAM COG KGZ GNQ CYP CMR COL VEN TZA RWA TGO MDG OMN ISL SLV TTO HRV BHR GHA HTI BLZ HND QATCIV BFA BOL NICMDA PRY SYR NZL IRN SAU GMB AGOLAO SRB BEN KHM LKALTU YEM LBN PAN MOZ KAZ KEN ZWE LVA ARE EST AGO MAR ZMB TUN PER SVK GNQ SWZ COM JOR PHL ZAF UGA WSM SYC SUR MOZ CRI ZAR MLI KIR BMU NER ZMB CAF LUX BDI KWT BLRROM AUS PLW GTM JAMBGD VNM VUT COG GMB MRT ISR TCD GNB EGY SVN UKR BGR SDN ERI RWA CMR CPV SLE ATG TTO MEX BRA GRC HUN BEN GHA IRQ KWT BFA CUB QAT LBY ARM KGZ STP UGA TUR URY CHL ETH MNG ISL IND TGO NOR BOL ITA DZA TZA LAO SEN DJI TKM YEM PRY ECU MLT UZB BHR MWI NGA PAK POL PER ARG BLZ CHL NIC HTI CZE ZWE BRB MDG NPL SWE AZE KHM KAZ JOR CIVCRI GEO RUS AUT SAU LBN HND THA SWZ MUS FJI IRN SGP DNK PRT ARE ALB OMN BEL BGD CYP VEN PHL DOM ARG MDA IDN FINCHE URY IRL SLV PAN ISR KEN GTM IRL BIH NLD KOR LVA AUS LUX CAN NZL ZAF MYS MAR MKD VNM AFG CHN GRC ESP MEX TUN BLR NOR SYR LKA PAK RUS COL EST LTU FIN UKR SRB EGY KOR HRV PRT SVKHUN JPN BRAGBR ESP THA SWE DNK BGR ROM TUR CAN SVN FRA POL IND CHE IDN GBRJPN CZE BEL SGP AUT FRA GER MYS NLD GERCHNUSA ITA USA 12 10 8 6 4 2 −6 −4 10 −2 10 10 6 2 1 10 −6 −4 10 Extensive Theil Index Data versus simulation BMU PLW VUTCPV WSM AFG TCD ERI GNB DJI 8 6 4 2 0 −6 MNG SYC SUR BRB SLE TKM HTI ISL NPL BIH RWA GMB BDI BLZ ALB BFA MLI MWI KGZ LAOBOL DZA GNQ FJI SDN BEN NER YEM CUB PRY MDA MLT ARM LBY GEO CAFAGO MDG LKA AZE MRT ZAR MKD LUX BHR TGO COM CYP KHM LBN MUS UZBSYR JAM PLW KIRCOG UGA KWT TTO GNQ GNB MOZ ECU HRV TCD BGD VUT TZA LVA SRB NICBDI STP ETHWSM SWZ ERI RWA AGO JOR CPV LTU EST MARIRQ PHL ZMB BMU CMR SVN NGA BEN GMB SYC NZL CAF VEN SLV DJI BLR DOM KEN OMN COG KAZ URY ZAR SDN BFA SEN TUN MRT MWI ATG PAN LBY IRN SUR ETH GTMTKM HND MOZ PER BGR SVK EGY NER DZA QAT MLI UGA CUB YEM ZWE ZMB TGO HTI BRB QAT MNG BLZ LAO JAM NGA DNK ARM KWT GHA AZE CMR GHA COL SEN TZA CIV VNM NIC BOL PRY GRC CHL ISL CZE HUN SLE TTO KHM FJI KGZ BHR ROM MDG NOR OMN UZB KAZ GEO ALB AUSCHE CRI CIV PRT LBN MLT CYP SAU ECU VEN NPL JOR HND MUS SLV DOM AFG UKR IRL BEL BIH ZWE SWZMDA BGD URY CRIGTM KEN AUT NLD POL CHL GRCIRN MKD PER FIN KOR PAK PAN LVA LKA HRV MAR SRB SYR SAU LTU LUX EGY TUN ARE NOR EST NZL BLR COL ISR TUR ARG VNM ARE AUS SGP PRT ISR RUS CAN IRL PHL ROM ZAF IDN PAK UKR MEX CAN FIN SVK BGR HUN SWE SVN POL DNK RUS TUR MYS ARG AUT BRA ZAF ESP IND SWE CZE CHE BEL IDN GBR FRA JPN THA SGP MEX BRA NLD KOR FRA ESP USA INDITA THA JPN CHN GBR MYS ITAGER GER CHN USA −4 10 Red − Simulated Data Blue − Empirical Data PLW 5 Import Concentration Export Concentration 6 Red − Simulated Data Blue − Empirical Data PLW STP COM IRQ 0 10 Extensive Theil Index Data versus simulation 10 10 (b) Overall concentration of imports 14 KIR −2 10 Log of GDP relative to the US (a) Overall concentration of exports 12 PAN ATG GNQ GNB COM ERI TCD WSM IRQ ARM GNQCAF SLE CMR HTI DJI MRT BRA SYR IRN NER KHM BEN GMB SYC TKM PER PHL AUS MLT GEO LAO RWA IND MWI TKM BDI COG ZAF KGZ JPN BFA TGO DOMVEN CPV UZB ZAR TCD UZB MWI AGO MLI BFA FJI BHR ERI MOZ NPL MNG YEM AZE KWT SLV QATNZL MYS YEM AGO SVN ISR COL TTO SUR CIV KAZ OMN BHR CIV SEN RUS AZE KGZ ROM CHL ZWE ARG SRB CYP USA TUN MDG BIH SAU SDN ZMB SDN ZWEGTM CHNCHN BDI CAF CMR BLR USA ARM LBY PHL TUR SWZ BOL KEN BLZ ALB UGA ITA FJI UGA BGD ETH GER KWT RWA DZA PRY BIH ISR LKA MDG ECU IND TZA BRB SWZ SGP NPL MDA TTO EGY HTI GHA LUX LBY ETH PAK KEN JPN HUN HRV THA LBN MOZ GNBWSM SYRKAZ AUT IRL NIC BEN SLE ALB TZA BGD MUSMLT CHE NLD SEN HND ARE SUR CRI MAR JOR DOM JOR DZA BEL CPV MLI GRC FIN JAM UKR GBR EST PAK SLV VNM HUN IRNKOR BOL QAT CUB LKA NGA ISL OMN BGR KHMMUS SYC LTU UKR BRA MKD PRY ARE MKD SAU LVA LAO NGA BLR ZAF SVK NER EGY ECU RUS KOR FRA NZL CHE POL CRI IDN GRC BGR NICLBN AUT SWE DNK GHA SWE CYP URY NOR MEX FIN CZE TUN AUS GTM COL TUR ZARMNG ESP TGO COM CUBISL URY VEN IDN KIR VNM SVN CAN PER ESP STP FRA ITA PRT CZE MRT MAR SVK ARG DJIIRQ COG MDA BRB CHL LVALTU VUTGMB SRB POL NOR PRT ZMBJAM ROM DNK BELGBR GER AFG THA GEO LUX MEX HND EST HRV NLD IRL PAN SGP MYS ATG BMUBLZ CAN KIR STP 4 Log of GDP relative to the US ATG BMU PLW 3 0 AFG VUT 5 −2 10 10 4 KIR DJICAF ATG GNQBMU CMR HTI LAO BRA IRQ ZAR IRN PER ARM RWA GMB SYR TKM DOM AUS MRT KGZ VEN CPV BDI COG NER ZAF SLV NPL TKM BEN MNG MWI SYC SUR BFA GEO UZB TGO BFA JPN UZB TTO MOZ MLI COLNZL KHM FJI FJI AGO ERI AZE ZWE BHR AGO CIV AZE CHL ZWE KWT BLZ MDG RUS ISR ARG BDI GTM BIH YEMBLR MDA YEM ARM ECU BRB SDN HTIUGA KEN SWZ RWA ALB BOL KAZ SWZ LBY ROM TUR SAU ZMB SRB USA UGA MOZ QAT CRI CMR SEN PHL CAF SDNKAZ BGD KGZ CHN LKA SUR ETH TZA BHR SYR CIV MDG ETH EGY CUBUKR IND GERCHN BLR TZA MUS NPL BEN GHA SEN NIC EST HUN NIC JPN DZA BIH MAR WSMMLI PAKFIN ITA LVA IND CUB KEN MKD CPV KWT DOM MLT JOR LTU OMN HND MUS BGR JAM SLE ALB SYC VNM TTO UKR POL PRY PAN NGA GNB NER LBY LKA URY BOL BGD SLV JOR SWE ITA USA LBN PRY IRN KOR GRC NGA ISL URY FRA SVN IDN NLD THA MEX BGR PAK IDN CRI KHMISL BRBZAR LAO GHA LTU CZE SVK GBR LUX BRA ESP SWE CYP CHE BEL TUN GTM ZAF JAM HRV AUT POL RUS TUR PHL MYS HUN TUN COM ROM OMN PRT ISR PER ECU ARE ESP QAT LVA DNK MAR TGO COG FIN EGY HND ARE SRB NOR ARG COL HRV KIR IRL DZA MKD MDA SVN NZL PRT MEX KOR ZMBLBN CAN NOR SGP MRT GEO VEN CYPMLT CHL AUT AUS CZE SAU VUTGMB GBR GER SVK IRQ VNMGRC THA STP EST DNK MNGPAN DJI ATG BMUBLZ IRL MYS SGP LUX AFG NLD CHE BELCANFRA 2 MWI TCD 0 10 −6 10 Log of GDP relative to the US −4 −2 10 10 0 10 Log of GDP relative to the US (c) Extensive margin of exports (d) Extensive margin of imports Intensive Theil Index Intensive Theil Index Data versus simulation Data versus simulation 10 8 Red − Simulated Data Blue − Empirical Data 9 AFG ERI TCD GNQ COM WSM SLE 1 0 VUT GNB STP 3 Red − Simulated Data Blue − Empirical Data 8 SEN 7 NGA 6 6 ATG UZB Import Concentration Export Concentration 7 SAU ARECOL DOM GEO GHA CIV ZAF AZE VEN QATCMR HND SDN OMN MEX BRA ETH ARM MUS SLV ITA ECU IRN ISR MRT TZA CUB IND ZWE MKD PAN ZAR JAM CRI HRV NZL AUS MLT NERCOG THA ARG TUR TTO KAZ ROM GNB MLI NICKEN UKR CHN TGO CYP LBY SYC PER SWE VNM SLE FJI TUN SVK SRB LTU BIH MLI ZMB ERI ZMB SLE MAR SUR ALB SYR RUS MOZ JAM EST BHR NER PER TCD AGO PAK POLESP SGP MDGNPL CHL KIR TTO DZA ZAR HUN SYC CPV MOZ PHL GRC GBR IDN WSM MWI GNQ ZAF ZWE KGZ LVA CMR ECU IRL PHL MYS GTM MRTMLT KGZ ARE BLR ISR ARG CHL AUT CAF ISL GHA MNG NOR CAN KWT CRI LBN STP JOR IRN SWZ JPN UZB KHM BMU SWZ ARM EGY BHR NPL AUS WSM COG MUS JOR QAT KAZ SAU BGR MEX RUS BOL TKM FINKOR GER IND PAN HND TZA PRY SEN MNG CIV GMBMDA CZE GNQ BGD VNM NZL BMU FIN BRA LUX PAK CUB HUN FRA MDG THA BEL LBN UKR LVA LAO URY JPN KHM GEO MYS PLW SDN AGO TGOMDA ESP NIC BRBUGA UGA ATGBLZ KEN SWE SVK PRY BLR PRT MARNGA DOM TUN KOR NOR BDI LBY PRT BOL GTM LKA VEN CYP COL CHE FJI GBR BIHSLV TUR IDN NLD IRL RWA MKD DNK EST FRA ETH BFA HTI SVN DJI SYR YEM SGP CZE BEL USA ALB LTU GRC GERCHN AZE OMN BGR VUT POL SVN BGD EGY LKA YEM ROM URY NLD VUT BENBLZ BRB TKM BFACOM SRBDZA SUR AUT HRV CAN BENAFG COM IRQ KIR LAOERIKWT IRQ MWI DNK TCD ITA CPV CHE ISL PLW DJI RWA HTIGNB LUX AFG STP GMB CAF 5 BDI 4 3 2 1 5 PAN 4 3 BMU PHL MLT SVN MYS IND CYPTUN SGP ATG AUT ISR OMN IRL CHE USA MLTAFG FRA LUX HRV DZA KHM SAU BIH MNG ARE CHE PRY GRC QAT THA CIV BHR MKD LBY DJI GNB KWT QAT LBN BEL STP BEL HUN ITA AFG VNM ARE OMN KOR NLD NZL VUT GBR LUX BGD CYP AUS KGZ SYC DZA SEN BEN NOR EGY SVK TTO ARM DNK MEX IRQ LAO NLD BRA GRC ECU ALB PAK CZE VEN CHNCHN IRQ GMB TGO KEN ZAF KHM PAK KAZ YEM GEO JOR RUS LBN SLE ROM AGO SDN CAN DNK BGR GER IRN KOR ISL NGA YEM COL FIN GBR SVK NER JPN MRT BHR KIR SRB SLV AUT CMR PRT HND BOL GHA NPL THA TKM GER ETH KWT TZA IND LKA EST TUN CHL CAF PRY AZE SWE ISR HUN JAM PHL EGY JOR SGP IRL CRI SAU MDG GTM DOM ERI MDA ZMB COM VUT GHA IDN TUR BFA VNM COG UZB BGD ARG MLI TGO ESP CPV BLZ ZMB MYS KAZ PER ZAR SYR CZE WSM ETH USAJPN NIC MRT ATG UGA UKR MAR FIN GEO MAR SYC SDN LKA BEN ESP MUS NER URY NGA MLI CIV PAN LBY RUS MUS CHL FJI LTU TUR AUS SRB NOR HRV SEN BOL UZB SWZ SWE ZWE SYR TZA ALB MOZ KEN LVA GNQ CMR HTI ARG URY ZAF TCD PRT IDN AGO GTM GMB MDG ISL MKD NZL SVN POL JAM POLBRA BDI ARM EST ROM LTU RWA MWI BGR COG LVA BMU BFA TKM PER UKR GNQ BRB IRN FRACAN HND BDI SWZ BLR AZE KGZ MWI MOZ SUR CPV ZWE NPL BRB CUB VEN HTIUGA COL BLZ MNG ECU CAF CRI ITA BIH RWA TTO NIC BLR MDA CUB MEX DOM SUR SLV PLW COM WSM DJI FJI LAO ZAR STP SLE ERI TCD KIR GNB 2 1 PLW 0 −6 10 −4 10 −2 10 0 0 10 −6 10 Log of GDP relative to the US −4 10 −2 10 0 10 Log of GDP relative to the US (e) Intensive margin of exports (f) Intensive margin of imports Figure 4: Simulated (in red) and empirical observed (in blue) export and import concentration versus GDP across 160 countries. The simulation is based on estimated trade costs form bilateral trade shares including an exported fixed effect. 29 Number of exporters per product Number of importers per product Data versus simulation Data versus simulation 0.1 0.018 Red − Simulated Data Blue − Empirical Data 0.09 0.08 0.014 0.012 Frequency Frequency 0.07 0.06 0.05 0.04 0.01 0.008 0.006 0.03 0.02 0.004 0.01 0.002 0 0 Red − Simulated Data Blue − Empirical Data 0.016 10 20 30 40 0 0 50 Number of exporters per product 20 40 60 80 100 120 140 160 Number of importers per product (a) Share of products per exporting country (b) Share of products per importing country Figure 5: The simulated (in red) and empirical observed (in blue) share of the number of products traded against the number of trading countries. 30 Import expenditure versus product share 1 Import expenditure share (1−D) 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 AFG ATG LUX VUT DJI HND SGP PAN BMU ZMB ZAR EST COG JAM MKD MLT MRT SVK GEO IRQ TGO QAT GHA SYC BIHALB STP CHE MDA COM GMB NER KIR BLZ MYS IRL OMN NLDCYP LVA BRB BEL AGO WSM ARE NIC LAO SUR CAF NGA TUN HTI CPV FJI MDG KHM MNG MEX MLI MUS ISL LKA TZA LTU LBN ARMBEN THA PHL LBY ETH HUNHRV KGZ MOZ CRI KWT JOR PRY SVN AUT ISR UGA GNB VNM NPL DNKDZA KAZ UZB ZWE SEN CZE SWZ BHR SAU GTM YEM BOL GNQ BGR SDN SLE BDI MAR CAN ERI SLV KEN NOR BLR MWI CHL GRC SWE AZE CIV RWA DOMBGD SRBPRT ROM ECU TKM TTO URYIDN GBR BFA FIN NZL UKR EGY POL GER PAK COL ESP FRA TUR PER VEN TCD CMRCUB ZAF RUSARG AUS ITA IRN KOR SYR IND USA CHN BRA JPN 0.1 0 0 PLW 0.2 0.4 0.6 0.8 1 Import product share (1−π) Figure 6: The import expenditure share versus the import product share. 31 Log of average per product import expenditure Average per product Import Expenditure −4 10 USA GER CHN GBR JPN FRA MEX IND ITA NLDCAN RUS ESP IDN THA AFG KOR BEL BRA CHE TUR MYS POL AUT IRN AUS SAU PAK NGA SWE PHL SGP UKR VNM ARE CZE KAZ EGY DNK GRC LBY HUN ARG IRL ISR ZAF NOR BGD UZB COL PRT DZA ROM SVK QAT KWT ZAR AGO BLR CHL SDN FIN NPL HTI VEN YEM CUB LKA PER GEO TKM AZE GNQ OMN BGR LAO MAR DOM ETH GTM BIHTZA TUN TCD NZL ARM LTU KHM SRB HRV ECU LBN LUX JAM GHA SVN CRI HND ERI KEN LVA UGA KGZ PAN ZMB VUT EST ALBCIV MLI SLVSYR MOZ NER BEN BOL SLECOG RWA PRY MDA MKD BHR JOR CMR MDG SEN URY BMU BFA MRTMNG CYP MWI NIC TTO DJI CAFTGO GNB SURBRB MLT MUS WSM ISL FJI BDI KIR COMATG GMB SYC BLZSWZ CPV ZWE STP −5 IRQ 10 −6 10 −7 10 −8 10 PLW −9 10 −5 10 −4 10 −3 −2 10 10 −1 10 0 10 Log of Absorption relative to the US Figure 7: The log of average per product import expenditure against log of Absorption. 32 Ratio of import expenditure vs tradable expenditure 2 Log of import expenditure ratio 10 KIR VUT AFG STP GNB COM ERI WSMGNQ DJI ATG BMU SLE CAF ZAR IRQ TCDHTI LAO GMB MRT COG ARM NER CPV SYC TGO KGZ GEO SUR BEN MNG BDI RWA FJI KHM NPL TKM BLZ AGO MOZMLI BIH UZB MDA ALB MWI ZMB BRB QAT EST ZWE BFA LBY GHA YEMHND AZEMKD MLT PAN JAM UGASENNIC SWZ LVA ETHMDG TZA PLW OMN LUX KAZ SDN BHR NLD SVK LTU MUS KWT NGA CHE LKA SGP BEL MYS BLR LBN CYP IRL CIV ISL PRY ARE TUN THA JOR CRI KENVNM BGR SVN PHL BOL SLV HUN MEX DOM CZE AUT ISRGER DNK GTM HRVTTO BGDCMR MAR SWE UKR URY CUB PAK IDNDZA CAN SAU GBR ROM NOR ECU SRB CHL GRC ITA POLPRT FIN FRA EGY TUR NZL CHN SYR ESP COL PER IRNRUS KOR USA ZAF IND VEN ARG AUS 1 10 0 10 BRA JPN −1 10 −3 10 −2 10 −1 10 0 10 1 10 Log of GDP per capita relative to the US Figure 8: The log of the ratio of average per product import expenditure with respect to average per product tradable expenditure against log of GDP per capita. 33 9 Tables Table 7: Country-Specific Technology and Trade Costs estimates Country USA AFG AGO ALB ARE ARG ARM ATG AUS AUT AZE BDI BEL BEN BFA BGD BGR BHR BIH BLR BLZ BMU BOL BRA BRB CAF CAN CHE CHL CHN CIV CMR COG COL COM CPV CRI CUB CYP CZE DJI DNK DOM DZA ECU EGY ERI ESP EST ETH FIN FJI FRA GBR GEO Exporter FE Standard error Precent cost Si Standard error (lUS /li )q 6,36 -0,46 -1,96 -3,31 2,98 2,19 -3,14 1,12 3,29 2,03 -3,41 -3,45 5,53 -3,11 -4,45 0,96 0,05 -0,83 -3,57 -1,40 -0,26 -1,26 -1,84 3,17 -1,49 -2,05 4,10 4,79 2,13 5,11 -0,12 -2,10 0,87 -0,04 -3,06 -3,16 0,32 -1,47 0,61 1,13 -1,23 2,57 -1,12 -2,29 -0,18 0,42 -4,87 3,76 1,75 -1,73 1,77 -1,88 4,56 4,86 -0,54 0,18 0,25 0,23 0,23 0,19 0,19 0,22 0,45 0,19 0,19 0,22 0,24 0,18 0,23 0,23 0,2 0,19 0,32 0,24 0,21 0,26 0,41 0,22 0,19 0,23 0,26 0,18 0,19 0,2 0,18 0,2 0,2 0,23 0,19 0,29 0,32 0,21 0,2 0,19 0,19 0,28 0,18 0,2 0,2 0,2 0,19 0,26 0,18 0,21 0,2 0,19 0,25 0,18 0,18 0,21 -53,47 5,53 26,79 48,53 -30,12 -23,01 45,55 -12,28 -32,75 -21,83 50,25 51,47 -48,63 45,41 70,73 -10,19 -0,59 10,52 53,34 18,12 3,53 16,48 24,88 -31,64 20,33 28,12 -38,98 -43,90 -22,51 -45,86 1,13 28,96 -9,83 0,63 44,10 46,15 -3,62 19,41 -6,89 -12,91 16,32 -26,76 14,60 31,68 2,25 -4,75 79,16 -36,45 -19,23 23,10 -19,42 25,42 -42,30 -44,32 6,46 0,84 -3,06 -0,97 -0,12 -0,71 1,54 0,2 -3,72 0,98 1,24 1,12 -0,45 -0,89 -0,38 0,6 0,27 1,01 0,26 1,1 2,1 -1,77 -1,91 0,39 1,71 -0,91 -1,11 0,43 -0,76 0,48 1,57 0,06 0,78 -2,63 1,13 -1,78 -0,66 0,06 0,86 -0,44 1,37 -2,99 0,97 0,65 0,61 0,57 0,83 0,12 0,81 -1,36 -0,6 1,73 -0,36 1,05 0,57 -1,25 0,13 0,19 0,16 0,16 0,14 0,14 0,16 0,3 0,13 0,13 0,16 0,16 0,13 0,15 0,15 0,14 0,14 0,23 0,17 0,15 0,18 0,28 0,15 0,13 0,16 0,19 0,13 0,13 0,14 0,13 0,14 0,14 0,17 0,13 0,19 0,2 0,15 0,14 0,14 0,13 0,2 0,13 0,14 0,13 0,14 0,13 0,19 0,13 0,14 0,13 0,13 0,17 0,13 0,13 0,15 1 193,42 23,5 9,63 2,6 2,24 9,61 12,93 1,2 0,77 12,2 49,84 0,85 45,53 36,85 10,69 2,79 1,87 6,66 2,17 8,17 5,66 10,01 2,22 5,95 21,31 0,99 0,9 2,34 2,22 8,73 8,12 16,64 5,89 42,22 16,29 3,43 9,57 3,51 1,04 50,23 0,8 3,72 17,61 6,51 9,14 43,83 1,19 2,27 70,73 0,59 5,57 0,8 1,06 15,46 34 Table 8: Country-Specific Technology and Trade Costs estimates - cont. Country GER GHA GMB GNB GNQ GRC GTM HND HRV HTI HUN IDN IND IRL IRN IRQ ISL ISR ITA JAM JOR JPN KAZ KEN KGZ KHM KIR KOR KWT LAO LBN LBY LKA LTU LUX LVA MAR MDA MDG MEX MKD MLI MLT MNG MOZ MRT MUS MWI MYS NER NGA NIC NLD NOR NPL NZL OMN Exporter FE Standard error Precent cost Si Standard error (lUS /li )q 4,74 1,14 -1,69 -3,13 -3,99 0,73 -1,41 1,26 -0,60 -3,14 0,43 4,30 3,76 3,90 -1,18 -3,12 0,08 1,26 3,96 0,76 -0,60 4,91 -0,28 -0,24 -3,04 -2,22 -2,77 4,42 -1,70 -3,15 -0,31 -1,81 0,98 -0,24 1,44 -0,64 0,73 -1,11 -0,95 3,42 -1,04 -2,42 0,30 -2,60 -1,13 -0,58 0,95 -3,87 5,40 -1,89 0,15 -1,13 5,66 1,83 -3,03 2,54 0,39 0,18 0,2 0,24 0,38 0,28 0,19 0,21 0,24 0,19 0,32 0,19 0,19 0,18 0,18 0,2 0,3 0,21 0,19 0,18 0,21 0,2 0,18 0,21 0,2 0,24 0,29 0,39 0,18 0,2 0,29 0,19 0,24 0,2 0,2 0,25 0,21 0,19 0,23 0,22 0,19 0,23 0,25 0,22 0,27 0,21 0,24 0,2 0,23 0,19 0,23 0,21 0,22 0,18 0,19 0,23 0,19 0,21 -43,54 -12,49 23,14 45,90 61,56 -8,59 18,61 -13,99 7,28 45,77 -5,15 -40,26 -36,12 -37,55 15,38 44,88 -1,09 -14,17 -38,00 -8,32 7,39 -44,65 3,08 3,22 43,58 30,65 38,98 -41,23 23,09 46,04 3,77 24,19 -11,03 2,69 -16,07 7,78 -8,11 14,13 12,10 -33,56 13,26 33,82 -3,45 36,44 14,84 7,23 -10,48 59,69 -47,75 25,67 -1,37 14,67 -49,44 -19,87 44,04 -26,41 -4,70 1,17 -1,78 -1,99 -0,89 0,39 0,93 0,41 -2,49 0,92 -0,5 1,49 0,21 1,03 -0,47 1,94 -1,13 -0,18 1,11 1,27 -1,7 0,24 1,95 1,08 -0,06 0,39 0,71 -1,68 1,4 0,84 0,54 -0,23 0,27 -0,37 0,6 -0,65 0,3 0,39 -0,33 -0,93 -0,1 -0,73 -0,45 -0,68 -0,51 -0,55 -2,13 -0,98 0,29 -0,74 -1,35 -1,19 -0,78 -0,88 1,02 0,37 0,58 -0,59 0,13 0,14 0,17 0,27 0,19 0,13 0,14 0,17 0,13 0,23 0,13 0,13 0,13 0,13 0,15 0,21 0,15 0,14 0,13 0,15 0,14 0,13 0,15 0,14 0,16 0,21 0,29 0,13 0,14 0,23 0,14 0,17 0,14 0,14 0,2 0,15 0,14 0,16 0,15 0,13 0,15 0,17 0,16 0,19 0,14 0,17 0,14 0,15 0,14 0,16 0,14 0,15 0,13 0,13 0,16 0,14 0,15 0,65 17,47 30,89 34,18 3,92 2,5 6,43 9,19 2,53 26,46 1,26 4,69 6,78 0,78 7,13 224,32 1,15 1,26 0,8 6,45 5,19 0,48 4,17 20,53 10,95 10,91 20,97 0,73 3,69 11,92 7,97 8,88 7,75 2,6 0,86 2,97 5,1 8,17 20,18 3,27 5,07 43,51 1,64 10,12 18,96 21,41 3,68 34,15 1,64 39,78 57,57 10,6 1,02 0,9 18,27 1,27 6,51 35 Table 9: Country-Specific Technology and Trade Costs estimates - cont. Country PAK PAN PER PHL PLW POL PRT PRY QAT ROM RUS RWA SAU SDN SEN SGP SLE SLV SRB STP SUR SVK SVN SWE SWZ SYC SYR TCD TGO THA TKM TTO TUN TUR TZA UGA UKR URY UZB VEN VNM VUT WSM YEM ZAF ZAR ZMB ZWE Exporter FE Standard error Precent cost Si Standard error (lUS /li )q 1,59 2,82 0,47 2,33 -9,10 0,87 1,76 -1,36 0,60 0,18 1,98 -3,73 1,34 -2,46 -0,69 6,66 -0,49 -1,74 -1,84 -2,21 -1,59 1,67 -0,38 2,74 -0,81 -1,17 -3,28 -5,68 -1,07 5,42 -4,02 -1,00 0,44 1,92 -0,25 -1,79 0,91 0,76 -2,14 -0,46 2,46 -0,93 -2,40 -2,67 3,49 1,02 1,85 -1,06 0,19 0,23 0,2 0,2 0,4 0,19 0,19 0,22 0,21 0,19 0,19 0,23 0,19 0,2 0,21 0,19 0,49 0,21 0,21 0,33 0,26 0,2 0,19 0,18 0,24 0,28 0,21 0,26 0,23 0,18 0,26 0,22 0,19 0,18 0,2 0,21 0,19 0,21 0,24 0,2 0,19 0,34 0,3 0,22 0,19 0,27 0,26 0,22 -17,08 -28,76 -5,45 -24,40 197,33 -9,93 -19,23 17,79 -7,14 -2,06 -21,33 57,16 -14,98 34,35 8,70 -55,17 5,65 23,31 24,59 29,84 21,12 -18,39 4,59 -28,26 10,26 15,16 48,21 98,20 14,05 -47,82 61,53 12,96 -4,70 -20,57 3,46 24,15 -10,55 -8,63 29,20 5,82 -25,57 11,75 32,82 37,58 -34,26 -11,18 -19,62 13,97 0,9 -2,16 1,17 0,07 4,52 1,78 0,7 0,6 -0,62 1,73 1,89 -0,15 0,36 -0,12 -0,57 -2,19 -0,94 0,42 1,36 -1,89 -0,75 -0,18 1,09 1,41 0,15 -1,46 2,26 0,93 -1,56 -0,68 1,08 0,46 0,01 1,38 -0,88 -0,32 1,75 0,44 0,68 1,35 0,24 -2,46 -1,26 0,39 0,48 -2,97 -2,55 0,16 0,14 0,17 0,14 0,14 0,32 0,13 0,13 0,17 0,15 0,13 0,13 0,15 0,13 0,13 0,14 0,13 0,33 0,15 0,15 0,24 0,18 0,14 0,13 0,13 0,19 0,2 0,15 0,18 0,15 0,13 0,19 0,15 0,14 0,13 0,13 0,14 0,14 0,15 0,18 0,14 0,13 0,23 0,22 0,15 0,13 0,2 0,17 0,16 8,28 6,67 3,64 4,89 0,19 1,57 1,47 7,95 2,69 2,39 2,15 67,74 4,12 41,79 14,97 0,98 22,08 4,95 3,15 30,19 3,21 1,73 1,01 0,66 3,87 3,51 7 52,38 32,81 2,93 12,48 2,04 3,76 2,55 30,26 37,54 3,01 2,72 12,54 4,18 6,16 16,57 9,02 31,04 2,25 58,54 18,9 9,86 36 References A LVAREZ , F. AND R. J. L UCAS (2007): “General equilibrium analysis of the Eaton Kortum model of international trade,” Journal of Monetary Economics, 54(6), 1726–1768. A NDERSON , J. E. AND E. VAN W INCOOP (2003): “Gravity with gravitas: a solution to the border puzzle,” American Economic Review, 93, 170–192. ——— (2004): “Trade costs,” Journal of Economic Literature, 42, 691–751. B ALASSA , B. (1963): “An empirical demonstration of classical comparative cost theory,” The Review of Economics and Statistics, 45, 231–238. B ERNARD , A. B., J. E ATON , J. B. J ENSON , AND S. K ORTUM (2003): “Plants and productivity in international trade,” American Economic Review, 93, 1268–1290. B RODA , C. AND D. W EINSTEIN (2006): “Globalization and the Gains from Variety,” Quaterly Journal of Economics, 121, 541–585. C ADOT, O., C. C ARRÈÈRE , AND V. S TRAUSS -K AHN (2011): “Export Diversification: What’s behind the Hump?” The Review of Economics and Statistics, 93, 590–605. C HOR , D. (2010): “Unpacking sources of comparative advantage: A quantitative approach,” Journal of International Economics, 82, 152–167. C OSTINOT, A., D. D ONALDSON , AND I. K OMUNJER (2012): “What goods do countries trade? A quantitative exploration of Ricardo’s ideas,” Review of Economic Studies, 79, 581–608. E ATON , J. AND S. K ORTUM (2002): “Technology, Geography, and Trade,” Econometrica, 70(5), 1741–1779. E THIER , W. J. (1982): “National and international returns to scale in the modern theory of international trade,” The American Economic Review, 389–405. F EENSTRA , R. C., R. E. L IPSEY, AND H. P. B OWEN (1997): “World Trade Flows, 1970–1992, with Production and Tariff Data,” National Bureau of Economic Research Working Paper, 5910. G AULIER , G. AND S. Z IGNAGO (2009): “BACI: International trade database at the product-level,” . G OLUB , S. S. AND C. T. H SIEH (2000): “Classical Ricardian theory of comparative advantage revisited,” Review of International Economics, 8, 221–234. G RUBEL , H. G. AND P. J. L LOYD (1975): Intra-industry trade: The theory and measurement of international trade in differentiated products, vol. 12, Macmillan London. 37 H AVEMAN , J. AND D. H UMMELS (2004): “Alternative hypotheses and the volume of trade: the gravity equation and the extent of specialization,” Canadian Journal of Economics/Revue canadienne d’économique, 37, 199–218. H UMMELS , D. AND P. K LENOW (2005): “The Variety and Quality of a Nations Exports,” American Economic Review, 95, 704–23. K OREN , M. AND S. T ENREYRO (2007): “Volatility and development,” The Quarterly Journal of Economics, 122, 243–287. K RUGMAN , P. (1980): “Scale Economies, Product Differentiation, and the Pattern of Trade,” The American Economic Review, 16, 670–675. L EVCHENKO , A. A. AND J. Z HANG (2011): “The evolution of comparative advantage: Measurement and welfare implications,” Tech. rep., National Bureau of Economic Research. M AC D OUGALL , G. D. A. (1951): “British and American exports: A study suggested by the theory of comparative costs. Part I,” The Economic Journal, 61, 697–724. O TTAVIANO , G. I. P., T. TABUCHI , AND J. F. T HISSE (2002): “Agglomeration and Trade Revisited,” International Economic Review, 43(2), 409–436. R OMER , P. (1994): “New goods, old theory, and the welfare costs of trade restrictions,” Journal of Development Economics, 43, 5–38. S HIKHER , S. (2011): “Capital, technology, and specialization in the neoclassical model,” Journal of International Economics, 83, 229–242. S ILVA , J. AND S. T ENREYRO (2006): “The log of gravity,” The Review of Economics and Statistics, 88, 641–658. S IMONOVSKA , I. AND M. WAUGH (2011): “The Elasticity of Trade: Estimates and Evidence,” Tech. rep., National Bureau of Economic Research. WAUGH , M. E. (2010): “International trade and income differences,” American Economic Review, 100, 2093–2124. 38 Appendix Chapter 1 The import expenditure distribution Note, the set of imported goods is defined as the sum over all the fraction of goods imported from all other countries in the world economy. Dn,k describes the fraction of goods country n imports from country k. Given that a country imports a good from only 1 source country (so the sets of products imported from different countries are mutually disjoint), implies that we can sum the fraction of goods imported over all trading partners to obtain the probability to import. I  Dnk Pr(imp) = k6=n The corresponding distribution function for prices ( p) is given by: Mn ( p ) = Mn ( p ) = b 1 b w p s ms Define u = ÂsN=1 ls ( kns wns ) 1 q ÂkI 6=n ÂkI 6=n ´p 0 ’sN6=n (1 Gns (q))dGnk (q) ÂkI 6=n ´p Dnk 1 Dnk 0 jn 1q q q 1 e 1 jn q q ÂkI 6=n Dnk 1 1 q q and du = jn 1q q q p ˆ Mn ( p ) = ✓ 0 1 dq u (e = 1u q q dq, ◆ dq we get: )du Hence, the import price distribution is independent of the source countries: Mn ( p ) = 1 1 jn p q e = Fn ( p) Using the import price distribution, we can derive the import expenditure distribution by the following transformation. Note, import expenditures of country n on good x in the case of CES preferences are given by: qn ( x ) pn ( x ) = ✓ min pi ( x ) i6=n ◆1 h h pmn qn ! b 1 b = w p min B i mi xiq ) kni wni i6=n !1 h h pmn qn (19) and the probability of importing at price p is given by Mn ( p). Hence, we can write the distribution function of import expenditure at price p as En ( p) = 1 e 1 (1 h ) jn k n 39 1 ( p ) q (1 h) h where k n = pmn qn is a constant. The corresponding Fréchet pdf is 1 e( p) = q (1 h) jn ✓ 1 j q (1 h ) n with location parameter sn = k n p kn ◆ ✓ ◆ 1 e p 1 q (1 h ) 1 p j n ( k n ) q (1 h) 1 12 . q (1 h ) and shape parameter a = Given that the price distribution is independent of the source country and follows a Fréchet distribution, we can calculate the corresponding concentration indexes analytically. The Theil index on the intensive margin for country n can be approximately written in terms of the continuos probability density function: TnW = " 1 Na  k 2 Ga ✓ Rk R¯ a ◆ ln ✓ Rk R¯ a ◆# ⇡ ˆ • 0 ✓ R R¯ a ◆ ln ✓ R R¯ a ◆ f ( R)dR Plugging in the density of the Fréchet distribution with location parameter sn = k n and shape parameter a = TnW 1 , q (1 h ) = ˆ 0 • 1 j q (1 h ) n we get: ✓ R R¯ a ◆ ln ✓ R R¯ a ◆ a sn ✓ R sn ◆ a 1 e ( R/sn ) a dR where R¯ a is the mean import expenditure. Solving the integral, we get: TnW = 1 G (1 1 a) ✓ˆ 0 • ⇣ ln u ( 1/a) ⌘ u ( 1/a) e u du ◆ ✓ ln G(1 1 ) a ◆ (20) The Theil index on the intensive margin of imports does not depend on the scale parameter sn and is hence identical across countries. The index is completely determined by the shape parameter a = 1/(q (h 1)) and depends only on the elasticity of substitution h and the degree of comparative advantage q. The integral in equation 20 cannot be solved analytically. To compute the exact Theil index implied by the shape parameter a, I approximate the integral numerically via the Gauss Laguerre procedure. The share of products exported The concentration index of exports on the extensive margin is given by the number of products exported to any destination divided by the total number of products in the world. The Law of Large Numbers implies that a country’s probability to export a good is equal to the share of products exported. Define the set of products that country n exports as the union of the set of products exported to each destination j, Uex = [ jI6=n B jn . Note that the set of products exported to desti- nation j overlaps with the set of products exported to destination k. The total number of products exported to both countries is the sum of the two sets minus the number of products that are in 12 The generic form of the Fréchet probability density function is: f ( x ) = 40 a s x s 1 a e ( x/s) a . both, i.e. B jn [ Bkn = B jn + Bkn B jn \ Bkn . Generalizing this expression to all possible desti- nations implies that the share of products country n exports follows the Inclusion and Exclusion Principle and is given by Uex = I 1  ( Ai ) i =1 +   i,j:1i < j I 1 Ai \ A j + · · · + ( 1) I Ai \ A j \ A k i,j,k:1i < j I 1 2 ( A1 \ · · · \ A I 1) where Ai defines the event that a product is exported to destination i, i.e. Ai contains all ⇥ ⇤ products where country i obtains the minimum price in country n, Ai = pn,i ( x )  min j6=n pn,j ( x ) . If we denote the intersection of all Ai ’s with an index L A L := \ ( Ai ) i2 L we can rewrite the set of products exported in a more compact form Uex = I 1 Â( 1) k i =1  1 ( AL ) L⇢{1,...,I 1},| L|=k The last sum runs over all subsets L of the indices 1, ..., I 1 where k describes the number of destinations a product is exported to. In the special case of symmetric countries, the number of products exported to k destinations is the same for all destinations and the intersection A L only depends on the cardinality of L. As a result, we can write the event to export to k destinations as the L = k intersections of A, ak = ( A L ) and the set of products exported simplifies to Uex = I 1 Â( 1) k k =1 where I 1 I k ! ak 1 is the total number of destinations. The resulting share of products exported can readily be calculated. For example, the share of products country n exports to one particular destination j is given by a1 = D jn and is equal to a1 = (k )1/q (1 + ( I 1)(k )1/q ) The share of products country n exports to any pair of destinations j and k is given by the 41 probability of obtaining the minimum price in those two destinations. a2 = Pr ⇢✓ ◆ ✓ ◆ ⇥ ⇤ p j ( x ) = pij ( x )  min plj ( x ) ^ pk ( x ) = pik ( x )  min [ plk ( x )] l 6 =i l 6 =i ⇥ ⇤ where the price of good x offered by country n in destination j is pn,j ( x ) = B kxnq 8 j 6= n and at home pn,n = Bxnq . Notice that the only difference between prices offered is whether the country sells in the home market or in the foreign market. Since k < 1 implies that p jj < p jl and pkk < pkl 8 j, k 6= l, we can write the set of products exported to destinations j and k by the corresponding probability to obtain the minimum price in the respective destinations ⇢ h i a2 = B Pr kxiq  min x qj , xkq , kxlq l 6=i,j,k, By the properties of the exponential distribution, this probability is equal to a2 = (k )1/q 2 + ( I 2)(k )1/q Proceeding in similar manner, one obtains the probability to export (and hence the share of products exported) to k destinations ak = (k )1/q k + ( I k )(k )1/q As a result, we can write the extensive Theil for exports as the inverse of the log of the share of products exported to any destination as ext Ti,X = ln I Â( 1) i =1 k 1 I k ! ak ! where ak = (k )1/q k + ( I k )(k )1/q Trade Data To build my empirical evidence, I use the BACI data based on the Comtrade data set collected by the United Nations. I choose the 6 digit HS 1992 product classification scheme as the preferred level of disaggregation. I assume that the tradable goods sector corresponds to manufactures defined to be the aggregate across all 34 BEA manufacturingindustries. Using a correspondance table provided by Feenstra et al. (1997), I identify 4529 tradable manufacturing products. I construct trade shares D following Bernard et al. (2003) and Waugh (2010) in the following manner: Di,j = Importsi,j Gross Mfg. Productioni - Exportsi + Importsi 42 In the numerator is the aggregate value of manufactured goods that country i imports from country j. These data are directly obtained from BACI. In the denominator is gross manufacturing production minus total manufactured exports plus manufactured imports (against all countries in the sample), see Eaton and Kortum (2002). Basically, this is simply computing an expenditure share by dividing the value of inputs consumed by country i imported from country j divided by the total value of inputs in country i. Gross manufacturing data are from either UNIDO (2012) or imputed from value added data from the UN National accounts. SITC 4 digit industry classification In the paper I analyzed total net trade flows at the 6 digit HS industry classification. This sections shows that the results found in the main part of the paper are not driven by the industry classification scheme and do apply in a more general sense. As a robustness check, I use the 4 digit SITC and the 6 digit NAICS classification scheme. The total number of SITC products is 642 and of NAICS products is 460. Table 10: Mean concentration indexes for net trade flows based on the 4 digit SITC and the 6 digit NAICS industry classifications: 160 countries Gini Mean index (SITC 4 digit) % share of overall concentration Correlation to HS Mean index (NAICS 6 digit) % share of overall concentration Correlation to HS X M 0.98 0.86 Theil Exports (X) Ext. Int. Total Mar. Mar. Theil Imports (M) Ext. Int. Total Mar. Mar. 2.16 1.86 0.70 1.29 54% 46% 35% 65% 4.01 1.99 0.95 0.88 0.94 0.91 0.95 0.62 0.83 0.73 0.97 0.82 1.98 1.75 3.73 0.52 1.23 1.75 53% 47% 29% 71% 0.95 0.86 0.53 0.79 0.94 0.89 0.91 0.66 Table 10 present the descriptive statistics for the SITC and the NAICS sample. The qualitative estimates of the SITC as well as the NAICS classification are very similar to the HS one. Exports are more concentrated than imports. Concentration is driven by the extensive margin for exports and on the intensive margin for imports. Also, in terms of cross country evidence, larger countries import and export more goods. Strikingly, the L pattern of the extensive margin also appears when using the SITC and the NAICS classification. 43 Poisson parameter approach The data contains intra-industry trade whereas the model is a pure Ricardian model. In this section I outline an alternative approach that converts the measurement of product units in the model to product units in the data. Suppose that the true level of disaggregation of Ricardian products, as defined in the Eaton and Kortum (2002) model, is unobserved and the classification scheme measures only an aggregate of those Ricardian products. For example, when products, in the sense of the Eaton and Kortum (2002) model, arrive at the boarder, custom agents aggregate those products into an industry according to the HS classification standard. The number of EK model products that custom agents assign to an HS industry classification is unobserved to the researcher. Given this interpretation, I model the classification process as a randomization device following a Poisson process with parameter µ. The parameter µ informs on how many EK Ricardian products, on average, comprise one HS code (the observed product unit in the data). To estimate the Poisson parameter, I proceed as follows. By the law of large numbers, the probability of importing a particular EK product equals the share of the number of EK products imported with respect to the total number of EK products. In the model, the probability that an EK product is imported equals P(imp EKprod ) = 1 Dii , where Dii is the probability of not importing an EK product. By independence, the probability of not importing any EK product within an HS µ code is Dii , where µ is the average number of products that comprise an HS code. As a result, we get the probability of importing an HS code (product unit in data), which corresponds to one minus the probability of not importing any EK products in that industry, P(imp HScode ) = 1 µ Dii . Since the probability of importing a product just equals the share of products imported, NM /N, we can use the definition of Theil index on the extensive margin, Tibm = ln( NM /N ) = ln 1 µ Dii to back out µ: µi = exp( Tibm ) ln( Dii ) ln 1 ! We compute the Poisson parameter for each country and take the average value as our estimate of µˆ = 1/I ÂiI=1 µi . The results imply that on average µˆ = 0.94 EK products comprise an HS code. code. Empirical evidence and simulation results In my simulation the total number of intermediate goods (N) is the product of the 4529 industries in the data times 0.94, the average number of products in an industry, N = 4258. One advantage of the this approach is that we can make use of the full data sample and do not lose 35 percent of trade flows when converting the data into net trade flows. Next, I present the empirical results for 44 the full sample together with the corresponding simulation results that replicate the data. Table 11: Mean concentration indexes for gross trade flows over 2880 country-year pairs Gini Level of concentration % share of total concentration Correlation with net trade Theil Exports (X) Theil Imports (M) X M Ext. Mar. Int. Mar. Total Ext. Mar. Int. Mar. Total 0.96 0.89 1.81 2.59 4.40 0.94 1.76 2.70 41% 59% 34% 66% 0.96 0.45 0.98 0.82 0.98 0.87 0.87 0.90 Figure 11 shows that, in general, the pattern of export and import concentration in the full sample is similar to the pattern in the net trade flow sample. Exports are more concentrated than imports for almost all countries and on all margins. The pattern on the extensive margin is displayed in Figure 11(d). Figure 11(c) shows the patterns on the intensive margin. Turning our attention to the quantitative differences, we observe that the overall level of concentration decreases with respect to the net trade flow sample for both exports and imports. The decomposition reveals that the effects are different across the margins. In the case of the extensive margin, concentration decreases with respect to the net trade flow sample whereas on the intensive margin concentration increases thus reversing the relative importance of each margin in terms of overall export concentration. Intra-industry trade increases the number of products traded and the sales value of the respective product. As a result, we observe a lower (higher) concentration index on the extensive (intensive) margins. The overall concentration index is primarily driven by the intensive margin with a share of 59% for exports and 66% for imports, see Table 12. Table 12: Simulated concentration level with Poisson parameter µ = 0.94 and exporter fixed effect Gini Model Simulation Data (gross trade flows) Data (net trade flows) X M 0.99 0.89 0.96 0.89 0.98 0.91 Theil Exports (X) Ext. Int. Total Mar. Mar. Theil Imports (M) Ext. Int. Total Mar. Mar. 4.97 60% 1.81 41% 2.60 1.15 39% 0.94 34% 1.10 45 3.32 40% 2.59 59% 2.13 8.29 4.40 4.73 1.76 61% 1.76 66% 1.61 2.91 2.70 2.71 Using data on gross trade flows I re-estimate trade cost and technology parameters based. I then simulate the model, calculate the resulting concentration indexes using the Poisson measurement device and compare the simulated results with the data. Table 12 shows the results. Export concentration is higher than import concentration on all margins. With respect to the decomposition, similar to the net trade case, the extensive margin dominates overall concentration for exports and the intensive margin dominates for imports. The obtained concentation levels of imports are close to the one in the data. In the case of exports, simulated concentration levels are too high. In terms of the cross country concentration pattern, the calibrated model fits the data well, particularly for exports. However, similar to the net trade results, the model cannot capture the negative relationship between import concentration and GDP. 46 8 Mean of the index from 1992−2009 IND USA CHN .85 ITA GER 6 TLS NGAYEM OMN ATG KWT MLI TCD BWA SAU IRN GAB COM GNB NER BDI ARE SYC QAT KIR COG RWA WSM GIN SDN SURCAF STP VUT TKMAZE DZA BFA VEN BEN MRT DJI MWI BMU MNE MOZ SYR BLR JAM ZMB GMB UGA ETH CUB BHS ECU NOR CPV HKG PNG CMR TTO GHA KAZ GUY MNG SLE ERI CIV TGO PRY ISL MLT BLZ MUS BHR KHM JOR ARM MAC SWZ BRB BOL FJI CRI SEN NPL PHL COL SLV BGD LBN CYP PAN SGP RUS CHL MDG KEN NIC HND GEO ZWE KGZ TZA IRL ALB VNM PER GTM EGY LUX PAK DOM MDA BIH MEX TUN LVA AUS LKA GRC MYS URY MKD ZAF NZL ARG MAR PRT FIN ISR ESP SRB SVK CAN LTU CHE AFG EST HUN HRV FRA KORUKR GBR BRA ROM IDN DNK SVN TUR SWE BGR AUT THA BEL POL JPN CZE IND NLD CHN GER ITAUSA 4 .95 .9 NLD Export Concentration JPN CZE 2 1 Total Theil index − SITC 4 digit 0 .75 .8 Export Concentration Gini index − SITC 4 digit Mean of the index from 1992−2009 TLS TCD NGA YEM GAB OMN ATG BWA GNB BDI COM KWT QAT SYC COG NER RWA MLI CAF SAU DZA GIN STPKIR SUR MRT DJI VUT TKM WSMSTP WSM BMU BEN BFA SDN IRN MWI MOZ GMB ZMB MNE JAM ARE CUB BHS CPV PNG ETH CMR UGA VEN GUY HKG ECU AZE ERI GHA TTO ISL MNG BLR SYR CIVMAC PRY KHM MUS TGO BHR KAZ SLE BLZ MLT NORSWZ FJI SEN ARM BRB BOLJOR LBN SLV NICBGD MDG NPL CRI HND CYP SGP TZA COL KEN GEO CHL KGZ ALB PER ZWE PAK GTM DOM PAN RUSIRLLUX VNM BIH LVATUN PHL MDA GRC LKA EGY NZL MAR URY MEX MKD AUS PRT MYS ARGESP SRB ZAF EST LTUFIN AFG HRV CAN ISR SVK ROM CHE KOR UKR BRA TUR SWE SVN DNK FRA HUN GBR BGR THA POL AUT BEL IDN .75 .8 .85 .9 .95 1 0 2 4 Import Concentration (a) Gini coefficient 8 (b) Theil index Mean of the index from 1992−2009 KWT 2 VEN AZE BLR NOR MLI QAT ECU RUS SDN TKM GAB MOZ PHL KAZ DZA ATGCOL NER JAM ZMB CIV GIN MNE SUR MWI CHL PAN CRI IRL TTO BWA ZAF VNM SGP COG HKG PRY UGA CMR MYS SYC BFA GHA MLT KEN ZWE SLE BEN MEX EGY ETH MRT GNB CUB MNG MUS AUS TCD ISL PER JOR KHM CAF CHE PAK ARG FRA HND BOL SLV CAN VUTBMU BHS PNG MAC LUX BHR FIN BRA HUN SVK ESP CYP GUY KOR TGO GTM IDN BGD BDI NPL ISR MDG GBR JPN TLS RWA NZL LVA DJI THA GEO TUN CHN ARM SWZ WSM IND UKR BLZ GER DOM SWE URY NIC PRT LKA KGZ TZA FJIBIH SRB MDA DNK AFG SEN CZE LBN GMB BEL MAR EST MKD TUR NLD AUT SVN POL BGR ROM GRC ALB LTU USA STPBRB HRV ITA COM KIR CPV COM KIR STP GNB VUT AFG 0 ERI 0 TLS TCD ERI BDI CPV RWA WSM BWA GMBCAF ATG DJI SYC COG NER GAB BFA MRT BMU BEN SUR GIN MLI BHS MWIBLZ CUB MNE DZA ETH BRB PNG UGA SDN QAT GUY TKM YEM JAM ZMBMNG TGO MOZ FJI SEN ARM SWZ HKG LBN CMR OMN ISL GHA BHR NGA SLE AZE MUS ALB KHM TTO NIC MAC MLT JOR NPL PRY BGD BOL KGZ TZA MDG GEO KWT CYP SLV CIVMDA BLR ECU VEN BIH KAZ GRC DOM HND SAU GTM MKD LKA KEN HRV TUN CRI LTU MAR ZWE SGP NOR LUX URY LVA PER PRT SYR P PAN AN PAK ARE SRB CHL IRN EST COL NZL EGY ISR IRL VNM ROM AUS MEX SVN PHL ESP FIN BGR TUR ARG SVK DNK POL AUT CAN BEL RUS HUN SWE MYS UKR KOR GBR CHE ZAF CZE IDN FRA THA BRA ITA USA JPN NLD IND GER CHN 4 4 ARE NGASAU OMN YEM SYR 2 Export Concentration 6 Extensive Theil index − SITC 4 digit Mean of the index from 1992−2009 6 Intensive Theil index − SITC 4 digit IRN 0 Export Concentration 6 Import Concentration 2 4 6 0 Import Concentration 2 4 Import Concentration (c) Intensive margin (d) Extensive margin Figure 9: Export versus import concentration on the 4 digit SITC level 47 6 Gini index − NAICS 6 digit Total Theil index − NAICS 6 digit Mean of the index from 1992−2009 Mean of the index from 1992−2009 6 AGO IRQ NGA BRN YEM TCD GNQ OMN IRNCOG QAT SAU GAB GNB KWT MLI AZE TKM BDI BFA VEN RWA BEN STP MRT MDV COM NOR SUR SDN TJK LCA ISL MOZ MWI JAM KNA CAF ZMB GIN NER GMB ATG VCT SYR DJI CPV GRD HKG COD ECU ETH TTO BHR GHA PRY UGA KAZ CMR SGP SLE TGO MNG BTN UZB RUS DMA BOL MUS SEN MAC CIV MLT JOR BLZ LAO KGZ KHM ARM CRI COL BGD LBN IRL FJI BRB CHL PER CYP HND GEO UKR EGY SLV KEN TZA MDG ZWE NPL GTM BIH PAK ALB URY MDA AUS DOM PHL LTU BLR TUN LVA NZL MYS ZAF LKA ARG FIN MAR GRC SRB CAN MEX VNM ESP SVK HUN HRV KOR BRA ISR PRT GBR EST BEL FRA ROU CHE SWE DNK IDN BGR NLD SVN AUT TURCZE THA POL JPN IND USA Export Concentration 3 4 5 Export Concentration .8 .9 1 AGO IRQ BRN NGA TCD YEM GNQ GAB QAT OMN GNB SAU COG MLI TKM RWA BDI MRT AZE COM BEN IRN BFA MDV STP VEN SUR ISL KWT KNA LCA TJK CAF SDN MOZ GIN JAM GRD GMB VCT CPV COD NER MWI DJI PAN HKG ZMB NOR ECU ETH TTO BHR GHA CMR TGO MNG PRY UGA SLE BTN DMA SEN MAC MUS SGP KAZ BOL SYR UZB CIV JOR LAO KHM ATG ARMBGD LBN RUS BLZ FJI MLT BRB KGZ PER CYP GEO HND COL IRL CHL TZA CRI SLV MDG KEN PAK GTM BIH ALB EGY NPL ZWE URY DOM MDA TUN AUS UKR LKA LVA GRCNZL PHL ZAF ARGMAR SRB MYS BLR LTU VNM CAN ESPFINMEX SVK HRV PRT KOR BRA HUN ROU ISR EST GBR BEL SWE CHE FRA DNK BGRIDN TWN NLD SVN THA TUR AUTPOL IND JPN USA CZE CHN 2 DEU ITA PAN TWN 1 .7 DEU CHN ITA .7 .8 .9 1 1 2 3 4 Import Concentration Import Concentration (a) Gini coefficient Between Theil index − NAICS 6 digit 4 Mean of the index from 1992−2009 4 Mean of the index from 1992−2009 COM AGO MDV IRN CPV GNQ GNB BRN MRT TCD GRD KNA IRQ DJI QAT HKG GMB LCA GAB RWA BDISTP BEN SDN KWT COD COG YEM SUR VCT GIN TKM JAM BFA NGA MLI BHR ISL PAN CYP MOZ AZE ATG LBN ETH BRB NER MNG SEN CAFBTN MAC FJI DMA TJK JOR SAU MLT MUS BLZ TGO CMR VEN OMN KHM UGA MWI GHA ARM PRY ALB ZMB KAZ SLE LAO GEO BGD TTO BIH SLV BOL UZB TZA GRC ECU SGP NPL DOM KGZ HND MDG NOR URY MDA CIV LVA EGY GTM MAR SRB SYR RUS PER KEN HRV CHL PAK IRN CRI LKA TUN LTU IRL PRT AUS NZL EST ZWE ISR COL ROU ARG BLR VNM UKR BEL ESP GBR PHL SVK FIN CAN ZAF AUT BGR SVN DNK POL CHE HUN MEX NLD TUR FRA BRA SWE MYS KOR USA CZE JPN IDN TWN THA IND ITA DEU OMN YEM IRQ SYR AZEBRN MLI COG TKM TCD BFA ECUZMB MWI TJK KWT RUS COL AGO GABQAT GNQ CRIIRL BDI UKR SGP TTO CAF ISL MOZ STP BEN NER CHLUZB PER GNBCIV JAM RWA KAZMYS ZWE ATG SLE GHA BOL PRY KGZ SUR SDN UGA PHL BLR KEN ETH MEX ZAF EGYCMR LAO TGO GIN PAK FIN CAN MRT KOR VCT AUS GTM DMA HUN BGD BTN MNG TWN HND NZL MUS ARG BHR SVK LCA TUN IDN BRA BLZ ARM MDG KHM VNM JOR LTU COD ESP MAC SEN FRA MLT NPL TZA KNA URY LKA THA SWE GEO SLV GMB IND MDA CHE DNK LVA BIH FJI DJI GBR HKG BEL CHN SRB MAR JPN LBN NLDBGR DOM BRB CZE GRD ISR ALB MDV DEU ROU EST TUR USA PRT SVN HRV POL COM CYP GRC AUT CPV ITA NOR VEN 1 Export Concentration 1 2 3 NGA SAU 2 Import Concentration PAN CHN 0 Export Concentration 2 3 6 (b) Theil index Within Theil index − NAICS 6 digit 1 5 3 4 0 (c) Intensive margin 1 2 Import Concentration 3 (d) Extensive margin Figure 10: Export versus import concentration on the 6 digit NAICS level 48 4 Gini index − HS 6 digit Total Theil index − HS 6 digit Mean of the index from 1992−2009 Mean of the index from 1992−2009 8 6 4 2 Export Concentration .95 .9 .85 ITA GER NGA TCD YEM OMN GAB COG IRN KWT GNB SAU QAT MLI COM ARE BFA WSM BWA AZEBDI RWA KIR BMU BEN SDN MWI GIN STP CAF VEN DZA ZMB MRT NER VUT CUB JAM ECUSYR MOZGMB BHS PNG ETH UGA SLEERIATG CMR GUY CPV GHA PRY NOR TTO MNG TGO ARM KAZ BHR DJI BOL ISL MLT BLZ BRB KGZ TZA MUS CRI COL ZWE FJI GEO SEN NIC KHM RUS PER PAN KEN SWZ CHL MDGNPL JOR HND EGY SLV LBN LVA GTM BGD CYPPHL ALB VNM AUS DOM IRL ARG LTU MDA BIH ISR MEX URY AFG MYSBLR CAN NZL MKD TUN MAR ZAF GRC LKA EST PAK BRA FIN UKR SRB HRV SVK KOR PRT HUN IDN ROM BGR DNK SWE THA ESP SVN IND TUR POL CHE FRA GBR CZE BEL JPN NLD AUT USA CHN GER ITA 0 .75 .8 Export Concentration 1 TCD GAB KWT COM YEM COG GNB QAT BDI KIR RWA MLI MRT WSM DZA OMN GIN BMU CAF BEN BWANGA VUT SDN SAU ARE STP MWI BFA PNG NER CUB GMB ZMB AZE CPV MOZ BHS CMR GUY IRN ETH JAM ERI GHA VEN ECU UGA TTO PRY MNG TGO BHR BOL ISL DJI SLE ATG BRB ARMATG ARM SEN KAZ TZA BLZ NIC MLT MUS KGZ FJI KHM SYR GEO MDG ZWE HND JOR PER NORCOL CRI SWZ CHL BGD KEN SLV LBN DOM PAN ALB GTM CYP MDA NPL URY RUS LVA EGY IRL BIH AUS ARG NZL LTU MARVNM MKD PHL GRC LKA BLR AFG PAK MYS MEX TUN EST SRB CANHRV ISR ZAF FIN PRTROM BRA UKR SVK HUN BGR DNK KOR IDN THA POLSVNSWE TUR ESPCHE AUT CZE BEL GBR IND NLD FRA JPN USA CHN .75 .8 .85 .9 .95 1 0 2 4 Import Concentration (a) Gini coefficient 8 (b) Theil index Mean of the index from 1992−2009 6 Extensive Theil index − HS 6 digit Mean of the index from 1992−2009 6 Intensive Theil index − HS 6 digit 4 0 PAN KIR TCDSTP GNB WSM ERI CPV BDI VUT RWA BMU BENGMB CAF DJI GAB COG MRT ATG BFA GIN MWI SDN BWA BHS PNG MLI MOZNER ETH DZA GUY BRB CUB UGA QAT YEM MNGARM BLZ TGO ZMB KWT CMR JAM SEN AZE NGA GHA PRY NIC SLE BOL KHM TZA BHR FJI ISL TTO OMN ALB GEOKGZ MLTMDG LBN NPL KAZ MUS CYP AFG JOR SAU SLV VEN ECUHND DOM SWZ BIH MDA BGD URY GTM ZWE CRI MKD KEN PAN GRC LVA IRN PER HRV LKA SRB CHL MAR SYR LTU TUN ARE BLR NOR EGY EST COL NZL VNM ARG PRT ISR IRL PHL ROM AUS PAK UKR FIN MEX SVK SVN CAN BGR HUN DNK POL RUS MYS TUR AUT CHE SWE ESP CZE BRA BEL ZAF GBR IDN THA KOR NLD FRA USA IND JPN ITA GER CHN 2 4 2 Export Concentration COM IRN ARE NGA OMN SAU YEM SYR KWT VEN NOR QAT ECU MLI AZE RUS GAB COG BWA COL ZMB BFA DZA JAM TCD SLE KAZ SDN MWI CRI TTO GIN ZWE CHL PER CMR NER CUB EGY GHA PRY KEN AUS MRT MOZ IRLMYS BEN VNM BHR MLT PHL ETH UGA MEX PNG MUS ARG ISL ISRZAF BDI GUY CAF SWZ CAN BOL MNG BHS LVA BRA BMU KGZ RWA GNB WSM TGO KORIND JOR HND NPLGTM ARM TZA LTU IDN GEO FJISLV BGD BLR GMB ATG MDG NZL PAK FIN THAJPN HUN FRA UKR BLZ SVK KHM NIC SEN TUN URY MDA LBN CYP SWE DOM ESP MAR STPBIH GBR PRT VUT NLD BRB EST CHE DNK TUR CHN MKD LKA BGR GRC USA CZE BEL ALB POL GER ROM SVN SRB KIR COM DJI AUT HRV AFG ERI CPV ITA 0 Export Concentration 6 Import Concentration 0 2 4 6 0 Import Concentration 2 4 Import Concentration (c) Intensive margin (d) Extensive margin Figure 11: Export versus import concentration for gross trade flows 49 6 8 6 VUT BMU AFG PAN ATG GNB ERI BHS COMWSM ARM DJI CAF SLE TCD IND NER MRT TGO BEN BFA KHM KGZKHM KGZ GMB BDI MLI GEO RWA GIN MNG PHL MWI MLT MDA UKR NPL JPN CPV ZWE YEM BLR KOR COG FJI USA SEN MOZ AZE BHR UGA CYP GUYSWZ PNG ZMB MDG ETH LTU ZAF BLZ ITAGER CHN TZA QAT THA PRY SDNCUB LBN JAM KEN MYS BRBNIC OMN PAK GHA ALB CMR BIH JOR BGR BGD TTO HND DOM ISR SVK NLD GAB SYR URY LKA HUN BRA KAZ KWT VNM IRL MUS SRB MAR SLV FIN TUR CHE FRA EST DZA IDNESP BWA SWE EGY GRC BELIRN BOL CHL ISLMKD GTM CRI CZE NGA LVA ROM NZLPER ARE ECU PRT TUN SVN AUT HRV GBR SAU POL RUS AUS MEX COL NOR DNK VEN ARG CAN 0 JPN USA GER CHN STPKIR 4 6 4 2 ITA Import Concentration NGA TCD YEM OMN GAB COGMLI IRN GNB QAT KWT ARE SAU COM WSM BDI BFA BWA AZE RWA BMU BEN SDN DZA GIN MWI MRTNER CAF ZMB VUT GMB CUB VEN JAM ECU MOZ PNG UGA ETHSYR BHS SLE ATG CMR CPV GUY ERI MNG GHA PRY NOR TTO TGO ARM BHR KAZ DJIBLZ BOL ISL MLT BRB KGZ TZA MUS CRI COL ZWE FJISWZ SEN NIC GEO KHM RUS PER PAN NPL KEN CHL EGY MDG JOR HND SLV LBNGTM BGD LVA CYP ALB VNM PHL ARGAUS IRL MDA LTU DOM BIHURY ISR MEX BLR AFG CAN NZL MYS MKD TUN MAR ZAF GRC LKA EST PAK BRA FIN UKR SRB HRV SVK HUN PRT KOR IDN ROM BGR DNKCHE SWE THA ESP SVN IND TUR POL FRA GBR CZE BEL NLD AUT STPKIR 2 8 Total Theil index − HS 6 digit Mean of the log index from 1992−2009 0 Export Concentration Total Theil index − HS 6 digit Mean of the log index from 1992−2009 −10 −5 0 −10 Log of GDP relative to the US −5 0 Log of GDP relative to the US R2=0.39 R2=0.38 (a) Overall concentration of exports (b) Overall concentration of imports Mean of the log index from 1992−2009 6 Extensive Theil index − HS 6 digit Mean of the log index from 1992−2009 6 Extensive Theil index − HS 6 digit 4 Import Concentration USA 0 4 2 KIR STP 2 TCD BDI ERI RWA BMU GMB BEN GAB DJI CAF COG MRT GIN ATG BFAZAR MWI SDN BWA BHS NER PNG MLI MOZ ETH GUY BRB UGA QAT CUBDZA YEM MNG BLZ TGO ARM ZMB JAM LAO KWT AZE CMR SEN NGA GHA PRY NIC BOL KHM TZAOMN BHR FJI ISL KGZ TTO GEO UZB MLTALB MDG LBN NPL MUSAFG CYP JOR SLV ECUKAZ VEN SAU HND DOM SWZ MDA BIH BGD URY GTM ZWE CRI MKDPAN GRC IRN LVA KEN HRV LKA SRB CHL MAR PER SYR TUN ARE BLR EST LTU NOR EGY COL NZLVNM ARG AUSMEX PRT ISR PHL ROM PAK UKR FIN SVKIRL SVNBGR CAN HUN DNK POL RUS MYS TUR ESP AUT CHE SWE CZE BRA BEL ZAF GBR IDN THA KOR FRA NLD IND JPN ITA GERCHN GNB CPV VUT VUT GNB ERI TCD AFG COM WSM SLE DJI CAF ATG BMU ARM GMB MRT RWA KGZ COG NER CPV BDI BEN MWI MNG BFAGEO TGO MOZ MLI KHMNPL FJI GIN GUY PNGBIH AZE ZWE BLZ YEM MDA BRB ZMB ALB SWZ UGA CMR GAB SEN SDNCUB QAT MDG SYR KAZ BHR ETH BHS CHN TZA GHA BLR NIC EST GERJPN KEN PAK LVA MKD DOM HND KWTBGD MLT LTU OMN ITAIND VNM JAM MUS PRY BGR TTO UKR PAN BOL LKA URY SLV JOR USA IRN NGA LBN ISL CYP THA SVN IDN NLD FRA KOR SVK HUN BRA GTM ZAF CZE TUR PMYS PHL HL PER TUN SWE GBR POL ECU RUS ISR ROM BWA CRI BEL MAR CHE ARE EGY AUT ESP ARG DNK SRB FIN COL HRV DZA NZL IRL CHL MEX PRT VEN NOR GRC SAU AUSCAN 0 Export Concentration PLW COM −10 −8 −6 −4 −2 0 −10 Log of GDP relative to the US (d) Extensive margin of imports Mean of the log index from 1992−2009 6 Intensive Theil index − HS 6 digit Mean of the log index from 1992−2009 PAN BMU STP KIR BHS IND PHL KOR UKR MLT CYP JPN USA ZAF MYS THA BLRIRL LBN LTU SRB ISR ITAGER GRC NLD ESP PRY JAM ATG SVK BHR HUN SEN FIN MDA DZA OMN MAR JOR AFG CHL ARM BRA CHE KEN KHM BGD BWA EGY PAK BGR SWE ETH BEL TZA MDG TTO SAUTUR URY CHN NZL PRT GEO YEM TGO UGA LKA HND BFA ROM IDNMEXFRA GTM MLI DOM CZE ARE GIN NIC SLV QAT BEN PER HRV GHA AUS CRI ECU CUB VNM ISL ZMB NER MUS AUTPOL SDN KWT TUN NOR MNG MKD NGA BOL AZE VUT ZWE IRN GBR COL VEN RUS SWZ MRT SVN FJI MWI KGZ NPL CMR CAN SYR KAZDNK EST LVA ALB ARG GUY DJIBLZ GAB MOZ BDI BRB GMB PNG BIH RWA CPV CAF COG COMWSM SLE ERI TCD GNB 0 MLI RUS COG GAB BWA COL ZMB BFA JAM SDN DZA TCD SLE MWI KAZ CRI TTO GIN ZWE NER CHL PER CMR EGY AUS PAN GHA PRY KEN CUB PHL MRT MLT MOZBHR IRL MYS BEN VNM ZAF ETH UGA MEX PNG LVA MUS ARG ISL ISR BDI SWZ GUY CAF CAN BOL MNG BHS BRA KGZ RWA GNB KOR WSM BMUTGO JOR HND NPL ARM TZA IND LTUGTM IDN GEO BGD GMBBLZFJI BLR ATG MDG NZLHUN PAK SLV FIN UKR THA JPN SVK KHM NIC SEN TUN URY MDA LBN SWE NLD ESPFRA DOM MAR CYP STP GBR CHN PRT VUT BRB EST CHE DNK BIH MKD LKA BGR USA GRC CZE BEL TUR ALB GER SVN SRB ROMAUTPOL KIR COM DJI HRV AFG ERI ITA CPV 4 Import Concentration 4 IRN ARE NGA OMN SAU YEM SYR KWT VEN NOR QAT AZE ECU 2 6 Intensive Theil index − HS 6 digit 0 2 0 R2=0.56 (c) Extensive margin of exports Export Concentration −5 Log of GDP relative to the US R2=0.75 −10 −5 0 −10 Log of GDP relative to the US −5 0 Log of GDP relative to the US R2=0.01 R2=0.14 (e) Intensive margin of exports (f) Intensive margin of imports Figure 12: The relationship of export and import concentration verus GDP across 160 countries based on gross trade flows. 50