The Role of Global Value Chains during the Crisis: Evidence from Spanish and European Firms

by Aranzazu Crespo ** Marcel Jansen

Documento de Trabajo 2014-09

August 2014

European University Institute.

Universidad Autónoma de Madrid & FEDEA

Aranzazu Crespo ^† European University Institute

Marcel Jansen^‡ Universidad Autónoma de Madrid & FEDEA

Abstract

This paper analyzes the role that global value chains (GVC) have played during the sovereign debt crisis in Europe, and the key policy challenges that Spain faces. First, we provide a comparative analysis of the export performance and FDI activities of Spanish firms, using the firm-level information of the EFIGE dataset. Then, we use input-output tables from the WIOD database to identify the industries whose participation in GVCs should be promoted since they retain a larger share of domestic value added. Finally, we analyze the capacity of GVCs to amplify or act as built-in stabilizers following economic downturns. To do so, we compute the first round impact on Spanish gross exports and domestic value added of a 10% increase in final demand in selected areas of the world. Spain’s gross exports would be worth 2.28% of GDP while the domestic value added from export would grow by 2.34%, which suggests that GVCs act as stabilizers since the promote more the domestic value added components.

Keywords: Global value chains, FDI, exports, Great Recession JEL codes: F14, F17, F44

^†E-mail: aranzazu.crespo@eui.eu

^∗The authors gratefully acknowledge funding from the Ministerio de Economía y Competitividad grants ECO2011-30323 (Crespo) and ECO2012-37742 (Jansen) and the Fundación de Estudios de Economía Apli cada (Fedea).

^‡E-mail: marcel.jansen@uam.es

1 Introduction

The recent economic crisis in Spain began as part of the global financial crisis and continued as part of the European sovereign debt crisis, which has afected primarily the southern European states and Ireland. On top of the economic and financial crisis, the world experienced a contraction in global trade between the third quarter of 2008 and the second quarter of 2009 that is now known as the “Great Trade Collapse”. The severity of this drop in international trade, driven by the fall in internal demand around the globe, caused European trade volumes to fall abruptly to 2006 levels.

The “Great Trade Collapse” while extremely strong was also short-lived in comparison with the sovereign debt crisis in Europe. International trade started to grow again as early as 2010 and by 2011 trade volume had returned to the levels before the collapse. However, in most of Europe the recession lasted much longer. In countries like France and the United Kingdom the recession formally ended in the last quarter of 2012, while Spain and Italy did not exit the recession until the end of 2013. The aim of this paper is to shed light on the role of international trade and, in particular, the international fragmentation of production in the economic recovery of Spain and several other European countries.

The fragmentation of production is not a novel phenomena. Firms have been producing items with components sourced from around the globe for centuries. What has changed, however, is the speed, scale, depth and the extent of global interactions. Increasingly, new players have become active in what have come to be called Global Value Chains or global supply chains (GVCs). Decomposing gross trade flows in value added contributions, and mapping out how and where production activities are carried out, allows a much more accurate picture of the “real” integration of an economy with the rest of the world, provides an enhanced understanding of its implications, and leads to a better identification of barriers and bottlenecks.

The study of GVCs is also changing the way we think about some pressing policy issues. Firstly, it forces us to rethink the concept of competitiveness. Rather than the mere capacity to export goods, competitiveness should be understood as the capacity of an economy to capture income or value added from the ever more internationally fragmented production processes. Second, it is important to understand whether global supply chains amplify the trade decline following economic downturns, or whether the fixed costs of establishing production linkages and the existence of long term contracts associated with supply chains act as built-in stabilizers. Altomonte et al. (2012) document the “bullwhip efect” of trade elasticities due to the adjustment of inventories of intermediate goods within supply chains in France and point out the necessity to understand the drivers behind exporting and foreign direct investment (FDI) by firms. Finally, the study of GVCs allows us to identify the real value that the diferent industries add to production (and exports), opening up new perspectives on specialization and comparative advantages.

We address each of these issues in turn, and compare Spain with the four largest economies in Europe: France, Germany, Italy and United Kingdom. We start by analyzing the recent evolution of the unit labor costs in these five countries and the connection between this standard measure of competitiveness and the export performance of these countries during the crisis.

Afterwards, we document some of the basic stylized facts related to trade, innovation and firm size using the firm-level information contained in the EFIGE dataset. This database combines measures of firms international activities with quantitative and qualitative information on items ranging from R&D and innovation, labor organisation, financing and pricing behavior. The objective of this comparative analysis is twofold: explain the factors that account for the resilience of Spanish exports during the crisis and identify the barriers that might limit the growth potential of exports in the near future. In line with most of the previous research, this analysis highlights the importance of the strong prevalence of small firms and the relatively low investment in R&D.

Finally, in a last step we use the input-output tables from the WIOD database to identify the segments of participation in GVCs that should be promoted if Spain wants to exploit the full potential gains from international trade. Our analysis uses two diferent criteria to select the most attractive segments: industries that retain a relatively large share of domestic value added and the industries in which Spain has a value added comparative advantage but not a gross exports comparative advantage.

The size distribution of firms is a classical argument for the cross country heterogeneity in firms’ performance.¹ Our analysis reveals that Spain has a smaller number of large firms than the reference countries, but these large firms (more than 250 employees) were less afected by the financial crisis of 2008 than their European counterparts. Among these large firms, 89% either export or perform FDI or both. This suggest that Spanish international firms are competitive and that being an internationalized firm is key to overcome negative shocks.

¹See Melitz (2003), Helpman et al. (2004) or Crespo (2012) among others.

A detailed analysis of firms’ activities, reveals that Spanish firms perceive exporting and FDI to be more complex than their European counterparts — they reach fewer markets, their export intensity is smaller and there are less firms performing there. The most relevant factors for exporting are size, age, belonging to a group and being foreign owned. While the most relevant factor for FDI is being an exporter previously, at least 80% of the firms that perform FDI are also exporters in all the countries.

Next, we use the framework proposed by Koopman et al. (2014) in order to trace out the domestic value added embedded into the gross exports. Koopman’s methodology is an accounting equivalence between a country’s gross exports and the value added components by source, where the three main terms are: domestic value added content, foreign value added content, double-counted items’ value added. Wang et al. (2013) has extended this methodology to decompose gross trade flows at the sector level. To our knowledge, while there have been attempts at measuring the domestic value added content of exports of some European countries (Cappariello and Felettigh (2014) for Italy and Amador and Stehrer (2014) for Portugal) we are the first ones to do so for Spain, both at the aggregate and at the sectorial level.

An initial first glance at the decomposition for all countries reveals two interesting patterns. First, the share of non-direct exports, that is the GVC component, has been rising the past decade in all countries confirming the importance that the internationa fragmentation of production now has. Second, the share of domestic value added (DVA) embodied in gross exports is counter cyclical, and thus the change in gross exports is not a suficient statistic in order to determine the exports contribution to GDP growth, despite it being a usual assumption in business cycle analysis.

We then turn our focus to the Spanish economy. The foreign value added (FVA) at the industry level is a proxy for the extent to which industry value chains are segmented into distinct tasks and activities that generate trade, compounding the double counting efect. This along with a revised notion of a country-sector’s revealed comparative advantage based on DVA in gross exports allows us to identify what are the key industries. Chemical products and basic metals are revealed as sectors in which we have comparative advantage with respect to the world, along with textile products which is the sector we rely on in times of crisis. Finally, the refined petroleum industry stands out by its fostering the domestic value added in the exports in the last decade despite it being the industry with a largest content in foreign value added.

Although these results give some indication of the barriers to growth that Spanish firms face in comparison to the rest of the European countries considered and which are the industries we should promote, it does not say anything about the role that GVCs play in the propagation or relief to shocks. To quantify whether the increased participation in GVCs palliated the efects of the domestic crisis in Spain after the financial crisis from 2009, we compute the impact (first-round efect only) on Spanish gross exports and domestic value added when there is a 10% increase in final demand in selected areas of the world. We find that Spain’s gross exports would be worth 2.281% of GDP while the domestic value added from export would grow by 2.337%. The analysis by areas shows the dependence of Spain on the final demand coming from inside the EU15, and particularly from France and Germany who represent half of the increase in exports. At the same time, it reveals a very small impact from increased internal demand in the rest of the European countries. Finally, it is worth noticing the impact of an increase in the final demand in areas like Asia, these countries are rapidly gaining terrain in the worlds gross export and have the highest domestic value added activation ratio for Spanish exports.

The rest of the paper is organized as follows. Section 2 presents the main stylised facts for the five European countries under study. Section 3 analyses the diferences between Spanish and European firms. Section 4 maps out the economic relations that underlie Spanish trade with the rest of the world. In Section 5, we estimate the impact on Spanish GDP of a positive shock on foreign demand. Section 6 concludes.

2 Internationalization and firm competitiveness

The recent crisis has reopened the debate about the lack of competitiveness of the Spanish economy. The most widely-used measure of competitiveness in the European Union is the evolution of unit labor costs (ULC). The unit labor cost is a macroeconomic aggregate that measures the labor cost per unit of product and is calculated as the ratio of total labor costs to real output. A rise in labor costs higher than the rise in labor productivity may be a threat to an economy’s cost competitiveness if other costs are not adjusted in compensation.

Inspection of the data reveals, however, that a simple comparison of the evolution of prices and costs between two countries may not be informative enough to determine the competitiveness of a country. If an increase in the ULC index indicates a loss in competitiveness of the country, then we should see a decrease in a country’s export shares whenever aggregate ULC goes up. Figure 1 shows the so-called Spanish competitiveness paradox, an example of the fact that a loss in competitiveness does not imply necessarily a loss in the world’s export shares. Figure 1a shows the evolution of the ULC for Spain and the main developed economies, while Figure 1b shows the evolution of these countries world’ export share during the 2000’s. The Spanish ULC index has grown faster than that of our main trading partners in the period between 2000 and the start of the crisis, but its export shares have decreased less than those of the other countries, the only exception being Germany. After 2009, the ULC experimented a remarked improvement in Spain while the world’ export share of all countrie decreased in favor of new players like China.

(a) Unit Labor Cost (ECB)

(b) Market Share Index (WTO) Figure 1: Competitiveness Indicators Vis-á-Vis the Euro Area

Antr`as et al. (2010) show that large Spanish firms experienced both lower ULC growth and higher export growth than their counterparts in other countries, yet this diferential is not reflected in aggregate price indicators due to aggregation and dispersion bias (Altomonte et al. (2011)). In the calculation of the ULC all the firms are taken into account while to calculate the economy’s total exports, only the exporters are taken into account. The diferent relative weights of firms in the aggregate ULC and in the economy’s total exports, helps therefore to explain the Spanish paradox.

Recent literature in industrial organization and international trade has provided abundant empirical evidence (di Giovanni and Levchenko (2009) and Bernard et al. (2011)) supporting the idea that the evolution of macroeconomic aggregates is determined closely by the decisions and characteristics of the firms in the economy. For example, Crespo and Segura-Cayuela (2013) show that the evolution of the aggregate ULC is driven by the reallocation of resources within the firms of an economy, and point out the necessity of understanding the role of firms and their heterogeneity, in particular a subgroup of them:

the internationalized ones. Internationalized firms from a sector or a country, those that either export, import, are outsourcers or perform FDI, are a minority and in general those that behave better in terms of productivity, size and innovation (Altomonte et al. (2012) and Rubini et al. (2012)). The higher performance is present before these firms start any international activities (Clerides et al. (1998) and Bernard and Bradford Jensen (1999)).

Porter (1990) defines the competitiveness of a nation as the productivity with which a nation utilizes its human, capital and natural resources, while Krugman (1994) refers to competitiveness as a poetic way of speaking about productivity. Most of the definitions of competitiveness allude to the productivity of firms. Thus, in this section we analyze the diferences in firm size distribution among the European countries and the importance that the size and the international status of a firm² have on productivity using the EFIGE dataset. The EFIGE data³ consist of a representative sample at the country level for the manufacturing industry of firms owning establishments with more than ten employees in several European economies, although we will focus only on⁴: France, Germany, Italy, Spain and United Kingdom. The database combines measures of firms’ international activities (e.g. exports, outsourcing, FDI, import) as well as quantitative and qualitative information on around 150 items ranging from R&D and innovation, labor organization, financing and organizational activities to balance sheet data of the firm.⁵

2.1 Size and Internationalization

Figure 2 depicts the firm size distributions in France, Germany, Italy, Spain and United Kingdom. The vertical axis shows the proportion of firms (in natural logs) that is larger than the firm size (also in natural logs) represented on the horizontal axis. At first sight, the number of large firms is comparatively low in Italy and Spain than in Germany, France and United Kingdom. In particular, the share of firms with more than 250 employees is 4.8 % in Italy and 5.2% in Spain, compared to 11% in Germany.

²Participation of firm in diferent international activities like exporting, FDI or outsourcing.

³The database was collected by the European Firms in a Global Economy (EFIGE) project, whose objective was to examine the pattern of internationalization of European firms. Visit www.efige.org, for more information on the project.

⁴The sample includes as well Austria and Hungary, however due to missing information in several key variables, we do not include them in the analysis.

⁵Altomonte and Aquilante (2012) provides more information on the construction of the dataset and a comprehensive set of validation measures that have been used to assess the comparability of the survey data with oficial statistics.

Source: Authors’ calculations from EFIGE Dataset Figure 2: Firm Size Distribution (log-log, axes normal scale)

Figure 3a and Figure 3b present the same data, but now separating internationalized⁶ firms from non-internationalized. While the internationalized firm follow the same pattern as shown in Figure 2, the diferences between Italy and Spain and the other countries disappear for non-internationalized firms. This suggest that international trade is key for understanding the diferences in the firm size distributions in diferent countries.

(a) Internationalized

(b) Non Internationalized Figure 3: Firm Size Distribution Internationalized (log-log, axes normal scale)

⁶Internationalized firms include all firms that either exporter or importer of services or materials, are outsourcers (passsive or active) or/and perform FDI activites.

Internationalized firms are more productive and more profitable than non-internationalized firms (Altomonte et al. (2012), Rubini et al. (2012)). Figure 4a and Figure 4b document these same stylised facts. Figure 4a shows the average profits in thousands of euros for internationalized and non-internationalized firms by country.⁷ In all countries, international firms have higher average profits than non intenationalized firms. The quality of the Amadeus data varies across countries⁸, which explain the relatively high reported profits of both Germany and UK for which larger firms are over represented in this exercise. Figure 4b plots the total factor productivity (TFP)⁹ for internationalized and non-internationalized firms by country. As expected, internationalized firms are, on average, more productive than non-internationalized firms in all countries, the gap being 8.6% in Spain.

(a) Profit

(b) TFP Figure 4: Internationalized vs. non Internationalized

2.2 Econometric evidence

The relationships between firm competitiveness and internationalization activities can be further investigated by a cross-section econometric estimation, in which we regress the TFP of each firm, as measured in 2008, against the diferent categories of internationalization activities, adding country and sector fixed efects. In this way we can confirm our findings

⁷We use the Amadeus dataset, published by Bureau van Dijk, which mainly reports balance sheet data, to complement the EFIGE survey data.

⁸The coverage of the EFIGE firms within the Amadeus dataset was 92% for France, 28.5% for Germany, 94% for Italy, 88% for Spain and 22.3%

⁹TFP has been computed by Altomonte et al. (2011) using the EFIGE data

(1)

by excluding possible composition efects.

\[T F P = \beta_ {0} + \beta_ {1} I n t e r n a t i o n a l a c t i v i t i e s + \beta_ {2} F i r m s i z e + \delta_ {1} C o u n t r y + \delta_ {2} S e c t o r + \epsilon\]

Table 1 shows the results for the importance of performing any kind of international activity. Active abroad includes being an exporter, an importer of services or materials, being an outsourcer (passive or active) and performing FDI. While in Table 2 we focus exclusively on two of these activities: exporting and FDI.

Robust standard errors in parenthesis and ^∗∗∗ denotes statistical significance at the 1 percent level. The number of observations is given by the number of purely domestic oriented firms plus the number of firms active in the selected international activity.

Table 1: Active Abroad and TFP premium

In the first column we estimate an ordered probit model, in which the internationalization status is regressed across the decile categories of TFP. The coeficient is positive and significant, indicating that the higher the productivity deciles, the more likely that a firm will be involved in some internationalisation activity. This is true for all activities though the efect is strongest for FDI. The ordered probit serves a robustness checks for the OLS

results in columns (2) and (3).

The specification of column (2) includes sector and country fixed efects. As expected, the coeficient is positive and significant, that is, the ’productivity premium’ increases with the complexity of internationalization activities. Once we focus on particular activities, like FDI and Exporting, the premia increases.

In the third column we control alsol for size of the firm, measured in terms of number of employees. While the TFP premium decreases significantly (it’s almost half than when we do not control for size), it is still positive and significant. As can be seen, coeficients tend to grow larger with firm size because more productive firms manage to grow larger than their less productive counterparts. However, for a given size class of firms, the size premium tends to be smaller in more complex international activities such as FDI. This is further evidence of tougher selectivity at the top, as more complex activities are chosen by firms which have TFPs above already quite high thresholds.

Robust standard errors in parenthesis and ^∗∗∗ denotes statistical significance at the 1 percent level. The number of observations is given by the number of purely domestic oriented firms plus the number of firms active in the selected international activity.

Table 2: International Status and TFP premium

3 Firm Level Evidence on International Activities

The size distribution of firms might be the key to explaining the cross country heterogeneity in firms’ performance. Figure 5 plots the change in profit margin between 2008 and 2009 by firm size, as measured by the number of employees, for Spain and compares these changes to the respective survey averages.

We have already reported that Spain has a larger number of small firms than our European trading partners. These small firms are also those that clearly fare worse than their European counterparts: 65.7 percent of the very small firms and 62.3 percent of the small firms in Spain report decreases in their profit margin, while the corresponding survey averages are 52.4 percent and 52.2 percent, respectively. Interestingly, for large firms this pattern is reversed. The share of firms reporting a decrease in profit margins in similar to the survey average, while the share of firms reporting an increase in their margins, 24.9%, lies substantially above the survey average of 18.4%. This result is tied to the internationalization of firms. Among the large firms, over 80% in all the countries are either FDI’s, Exporters or both, compared to 89% in Spain.

Source: Authors’ calculations from EFIGE Dataset

Figure 5: Share of firms by change in profit margin by firm size.

This results suggest that the large Spanish firms were less afected by the crisis than the small ones because the vast majority of them was involved in international activities. In this section, we explore the implications of exporting and FDI activities for Spanish firms and compare these firms to their European counterparts.

3.1 How Diferent are Spanish Exporters from European Exporters?

In the previous section, we have shown that exporting firms are more productive and bigger than non-internationalized firms, independently of country or sector. Now we seek to understand the diferences between Spanish exporters and EU exporters.

By using firm-level data it is possible to decompose a country’s manufacturing exports into two margins: the percentage of firms in manufacturing that export a fraction of their sales (the “extensive margin”) and, only for exporters, the share of the export value over total turnover (the “intensive margin”). In Table 3 we report these two figures by country. Both margins vary substantially across countries and, as expected, both numbers are relative small in a large economy like Germany, while we obtain the largest numbers for Italy. Spain, despite being a smaller economy than France and United Kingdom, has both a smaller percentage of exporting firms and an smaller percentage of export volume, very close to German’s levels.

Table 3: Extensive and Intensive Export Margin by country

Figure 6 takes a closer look at the intensive margin distribution across the exporters in each country. The distribution of the other European countries is fatter and longer tails than the distribution of Spain. This, together with the fact that there are fewer large firms in Spain, implies a relatively small export intensity for Spanish firms. An interesting exception is the case of Italy. This country has even fewer large firms than Spain but at the same time has the fattest tail in the export intensity distribution, which explains the high percentages reported above.

Figure 6: Intensive Margin distribution

This evidence suggests that size is not the only relevant firm characteristic for internationalization. Therefore, we perform a regression analysis to determine how important are other firm characteristics to determine export performance by country. Specifically, we estimate a linear probability model, where the dependent variable is a dummy which is equa to 1 if a firm exports and 0 otherwise, and a probit model for robustness, controlling for the number of employees, the age of the firm, the firm belonging to a group, the firm being foreign owned, and the innovation activities (product innovation and R&D) performed by the firm.

\[\begin{array}{r c l} P r (E x p o r t e r) & = & \beta_ {0} + \beta_ {1} l n (E m p l o y e e s) + \beta_ {2} l n (F i r m a g e) + \beta_ {3} G r o u p \\ & + & \beta_ {4} F o r e i g n o w n e d + \beta_ {5} P r o d u c t I n n o v a t i o n + \beta_ {6} R D s h a r e + \epsilon \end{array}\]

The results are reported in Table 4. In the first column of each country are the LPM coeficients, while in the second column of each country are the Probit coeficients.

Probability of E^xporting by Countr

The estimated coeficient of firm size is visibly similar across countries, with the exception of United Kingdom it is smaller; the same is true for the firm being foreign owned and for performing R&D. The age of the firm has a particularly strong efect for Spanish firms, which is confirmed by the results for the Probit specification, with marginal efects twice as large as in any of the other countries. Product innovation is not as important for French and Spanish firms, but it is still significant and afects positively the probability of exporting. Finally, whether or not a firm is part of a group of firms has very diferent efects across the countries in our sample. While it is relevant in France, Germany and United Kingdom, it seems to have little to no efect in Italy and Spain.

Finally, an indicator of the complexity of the exporting activity is the number of foreign destination markets of firms. Eaton et al. (2004) found that the number of French exporters is strongly decreasing in the number of destination countries.¹⁰ Table 5 shows that this is the case also in our sample.

In all countries, only a small share of firms export to more than 20 destinations. However, we can see some diferences across countries. For each export percentile (according to their revenues), German exporters reach more destinations than their counterparts in the rest of the countries in the sample, while Spanish exporters are those who reach the lowest number of destinations. More importantly, the median exporter in Germany reaches twice the number of destinations that the median exporter in Spain does, which emphasizes the unevenness between two similar firms in both countries.

Table 5: Number of Export Destinations

¹⁰The number of firms selling to multiple markets falls with the number of destination areas. Using more recent data (2000-2006), Fontagn`e and Gaulier (2010) show that most French exporters are involved in only one foreign market. In addition, they show that the number of served countries increases with firm size and productivity.

3.2 How Diferent are Spanish FDI makers from European FDI makers?

By using firm-level data it is also possible to decompose a country’s manufacturing FDI activities into two margins: the percentage of firms in manufacturing that perform FDI (the “extensive margin”) and, for FDI makers, the share of the FDI value over total turnover (the “intensive margin”).

In Table 6 we report these two figures by country, along with the number of firms that both export and perform FDI. Again we find that both margins vary substantially across countries. Both France and United Kingdom stand out. France because it has the highest FDI volume and the number of French FDI makers is also one of the highest in the sample, whereas United Kingdom present one of the smallest FDI volumes despite having the largest number of FDI markers. More importantly, at least 80% of the firms that perform FDI are exporters, which indicates a complementarity between both activities.

Table 6: Extensive and Intensive FDI Margin by country

Figure 7 takes a closer look at the intensive margin distribution across the FDI makers. The distribution of Spanish FDI makers is similar to that of the other EU countries, indicating that despite having a smaller percentage of firms doing FDI, the intensity of the FDI makers is up to par with the firms in the other countries.

We now perform a regression analysis to determine the relative importance of other firm characteristics as determinants of FDI performance, just as we did in the case of exporting firms. To be more precise, we estimate a linear probability model where the dependent variable is a dummy which is equal to 1 if a firm performs FDI and 0 otherwise, and a probit model for robustness, controlling for the number of employees, the age of the firm, the firm belonging to a group, the firm being foreign owned, and the innovation activities (product innovation and R&D) performed by the firm.

Source: Authors’ calculations from EFIGE Dataset Figure 7: Intensive Margin distribution

\[\begin{array}{r c l} P r (F D I) & = & \beta_ {0} + \beta_ {1} l n (E m p l o y e e s) + \beta_ {2} l n (F i r m a g e) + \beta_ {3} G r o u p \\ & + & \beta_ {4} F o r e i g n o w n e d + \beta_ {5} P r o d u c t I n n o v a t i o n + \beta_ {6} R D s h a r e + \epsilon \end{array}\]

The results are reported in Table 7. In the first column of each country are the LPM coeficients, while the second column of each country reports the Probit coeficients. The estimated coeficient of firm size is visibly similar across countries; the same is true for the firm being foreign owned. The age of the firm and belonging to a group have diverse efects across countries.

robability of perform^ing FDI by count

In this section we have explored the characteristics of Spanish firms that export and perform FDI and compare them to their European counterparts. Spanish firms perceive these activities to be more complex than their European counterparts and most of the firms who perform FDI are exporters beforehand. The most relevant factors for exporting are size, age, belonging to a group and being foreign owned, although belonging to a multinational is not as significant for Spanish firms as for European ones. While the most relevant factors for FDI are size, being foreign owned and R&D.

Exporting and FDI are two complementary but alternative ways of internationalization, which are jointly reflected by the globalization of value chains. Production processes are becoming increasingly fragmented geographically, and thus multinational firms play a prominent role as they have a global reach that allows them to co-ordinate production and distribution across many countries and shift their activities depending on changing demand and cost conditions. In the next section, we explore the position of Spain in the international fragmented production process.

4 Reconciling aggregate and firm-level evidence: the role of global value chains.

Trade and production networks are not new phenomena. Firms have been producing items with components sourced from around the globe for centuries. Businesses have continuously sought out new markets for their products. What have changed, however, are the speed, scale, depth and breadth of global interactions. Increasingly, new players have become active in what have come to be called global value chains or global supply chains. This process of organization has brought entirely new issues for consideration.

GVCs pose measurement challenges to the evaluation of a country’s exposure to foreign shocks. As intermediates travel to their final destination by an indirect, possibly multi-country route, it becomes more complex to associate a country’s production (and its domestic-value added content) with the final demand that activated it.

The propagation of GVCs has contributed to the growth of international trade in intermediate inputs, as sequential stages of production (“tasks”) are often performed at several locations all over the world before assembly into the final product. As production becomes more and more internationally fragmented, conventional indicators based on gross exports are no longer informative and measurement challenges are posed to full evaluation of a country’s exposure to foreign shocks. First, the development of multi-country production linkages has made it more dificult to associate a country’s production with the final demand that activated it since intermediates produced in one country can be processed in many other locations before they are ultimately exported and consumed (or invested) in the final destination country. Second, traditional gross trade statistics are increasingly afected by the well-known problem of “double counting” since intermediates crossing the border back and forth as they are being processed get recorded multiple times in trade statistics. This holds even for bilateral trade flows.

The main purpose of this section is to analyze the Spanish capacity to generate value added through exports, which depends critically on our position in the GVC and map out the economic relations that underlie Spanish trade with the rest of the world.

To this end, we use the global input-output database WIOD¹¹ and the approach developed by Koopman et al. (2014) in order to trace out how final internal demand around the world is difused along global value chains and ultimately afects the creation of domestic value added across Spanish sectors. Essentially, WIOD tables match national inputoutput (supply and use) tables so that the foreign sector in each national table is broken down among partner countries both on the export (use) and on the import (supply) side. The outcome is a global input-output table where productive sectors are distinguished by their“country of residence”. The papers by Koopman et al. (2014) and Wang et al. (2013) lay out the appropriate algebra in order to trace out the contributions of final internal demand in each country in activating Spanish exports and the value added contained therein.

The rest of the section is organized as follows. First, we describe the Koopman methodology. Then, we analyze the domestic value added embedded in each of the European countries’ exports. After that, we focus on Spain and perform an industry analysis. Finally, we propose a new measure for comparative advantage and compare the diferences with respect to the traditional estimators.

4.1 Decomposition of the Gross Exports

Koopman et al. (2014) is the first paper in the vertical specialization literature that develops a fully coherent accounting identity that breaks up a country’s gross exports into value added components by source. The authors’ methodology decomposes gross exports into three main terms: domestic value added, foreign value added, double-counted value added.

¹¹See Timmer et al. (2014) for a description of the data or visit www.wiod.org for more information.

Figure 8 breaks down the gross exports into the nine sub-components specified by Koopman et al. (2012) depending on the use (final vs intermediate) of the exported goods and services and on the geographical origin (foreign vs. domestic) of the final demand that activated them. These sub-components can be grouped in two distinctive ways. On the one hand, we can distinguish between domestic value added (DVA), namely the Spanish GDP embodied in Spanish gross exports (components (1) to (5)), foreign value added (FVA) embodied via imports of intermediate inputs in Spanish gross exports (components (7) and (8)), and double-counted terms (components (6) and (9)). On the other hand, we can distinguish between stand-alone exports, that measures how much of a country’s exports is created outside any supply chain (SE ), given by sub-components (1) and (2), and the exports belonging to a global value chain (GVC ), which measures exports generated by participation in an international fragmented production process, given by sub-components (3) and (9).

4.2 Domestic Value Added embedded in Gross Exports

Because not all exports constitute domestic value added, the share of value added trade captured by a country can be quite diferent from its share in global exports. Figure 9 depicts the domestic value added (DVA), the foreign value added (FVA) and the doublecounted items (DC) as obtained from the Koopman decomposition for several European countries for 2011.

The domestic value added embodied in Spanish gross exports is the lowest among the countries considered, while the largest belongs to United Kingdom (70% and 78%) and double counting items are around 4% of the gross exports of all countries.

Figure 9: Breakdown of European Gross Exports in 2011

Factors that influence the share of domestic value added in exports include:

1. Size of the economy. Large economies tend to have significant value chains and to rely less on foreign inputs.

2. Composition of exports and position in GVCs. Countries with significant shares of natural resources, oil or other commodities in their exports, tend to have a higher relative value added trade share, as such exports are at the beginning of GVCs and require little foreign inputs. In contrast, countries with significant shares of exports in highly segmented industries may need to import more to generate exports.

3. Economic structure and export model. For example, countries with important processing trade sectors will capture less domestic value added.

The combination of these three factors explains the bulk of the diferences in the domestic value added shares (net of policy factors). Furthemore, the average GVC share of exports of an industry provides a rough indication of the extent to which industries rely on internationally integrated production networks, as it proxies the extent to which intermediate goods and services cross borders until final consumption of the industry’s output. Thus, Table 8 provides a breakdown of the domestic value added content of the exports from 2007 to 2011 of France, Germany, Italy, Spain and United Kingdom.¹² Particularly we provide, the total value of the gross exports in millions of dollars, and the percentages associated to each component according to the breakdown in Figure 8. For example, in 2009, 2.73% of the spanish gross exports is actually double counted items.

The fluctuations in 2009 indicate that the majority of the aggregates presented in Table 8 are sensitive to the business cycle. The share of DVA in gross exports is countercyclical, as for the complementary share, both FVA and double-counting are pro-cyclical. Similarly, since stand-alone exports (SE) constitute the greater part of DVA, the GVC (complementary share of the SE) is pro-cyclical. Since the DVA embodied in gross exports is counter-cyclical, the change in gross exports is not a suficient statistic in order to determine their contribution to GDP growth, despite it being a usual assumption in business cycle analysis.

During the pre-crisis period, the share of domestic value added embodied in every country’s gross exports declined over time. In particular, for Spain, it dropped from 74,18% to 10,39% between 2001 and 2008, rebounded to 75.75% in 2009, and dropping down below 2008 levels (70.06 %) by 2011. The pattern is similar across all the European countries considered, which indicates an increasing use of intermediates produced abroad and a strengthening of European producers’ position in the international production process. However, there are some diferences worth remarking. On the one hand, the average DVA embodied in UK’s gross exports is almost 10% higher than any other, which might be explained by strong internal value chains.¹³ On the other hand, Spanish DVA is the lowest among the five countries, which indicates we rely more on foreign inputs.

In the next subsection we center on the Spanish economy and identify which sectors are more integrated in the GVC process.

¹² A full decomposition of the nine components in Figure 8 and the complete description of the participation in fragmented production for each country for the 2001-2011 period can be found in the Appendix.

¹³Large economies tend to have significant internal value chains and rely less on foreign inputs, although according to the UNCTAD (2013) report there are important exceptions, including China and Germany.

Table 8: Participation in Fragmented Production in Europe

4.3 An industry analysis of Spain’s gross exports

We first look at the decomposition for the gross exports of the textile industry and the sector of refined petroleum products. The first sector is one of the traditionally most relevant export sectors for Spain and represents around 4% of total gross exports. The second sector is of particular interest because of Spains dependence on energy imports. Figure 10 provides a graphical presentation of the decomposition of gross exports. The four components clearly exhibit diferent levels and trends in the two sectors.

On the one hand, the evolution of the components in the textile product sector (Figure 10a) is fairly constant over the last ten years. On average, domestic value added that is ultimately absorbed abroad (DVA) is about 75% of gross exports, while foreign value added (FVA) that is embedded in Spanish textile exports is around 20% of the gross exports. The remaining parts consist of returned domestic value added (RDV) and pure double counting (DC), which together are around 5%.

On the other hand, the evolution of the components in the refined petroleum sector (Figure 10b) fluctuates over the whole sameple. On average, domestic value added that is ultimately absorbed abroad (DVA) is between 20% and 40% of gross exports, with a clear decreasing pattern, which highlights the dependence of Spain in these products. Complementary, foreign value added (FVA) presents an increasing pattern and represents between 40% and 60%of the gross exports. Interestingly, this sector presents the highest rate of gross exports being double counted. The intermediates crossing the border back and forth as they are being processed represent around 20% in the refined petroleum sector.

(a) Textile Products

(b) Refined Petroleum Figure 10: Evolution of the Gross Exports Composition

While FVA in exports is not a fully fledged indicator of the GVC complexity of industries, its analysis can provide insights on the extent of cross-industry production of goods and services. At the country level, foreign value added in exports indicates what part of country’s gross exports consist of inputs that have been produced by other countries, or the extent to which a country’s exports are dependent on imported content. At the industry level, the average foreign value added is a proxy for the extent to which industry value chains are segmented into distinct tasks and activities that generate trade, compounding the double counting efect.

Figure 11 plots the evidence for the foreign value added content and its contribution to the total Gross Exports of the diferent sectors. We have highlighted several sectors. The Textile Products (in green), is the sector whose FVA content and contribution to gross exports are equal to the mean of the economy. The sectors whose FVA content and contribution to exports are above the mean of the economy in red, and those whose FVA content is below the mean but their contribution to exports are above the mean in blue.

Figure 11: FVA vs. importance for the Gross Exports

The red sectors could be considered the bad or unattractive sectors of the economy, but they are also those who have been traditionally considered to be at the forefront of value chain segmentation and of associated trends such as outsourcing and ofshoring. The electronics and automotive industries, where products can be broken down into discrete components that can be separately produced, easily transported, and assembled in low-cost locations, have led the way in shaping GVCs and consequently rank highest by share of foreign value added in trade. A number of industries that incorporate and process outputs from extractive industries and traded commodities (e.g. petroleum products, plastics, basic chemicals) follow closely behind. The blue sectors, could be considered the good or attractive sectors of the economy, but particularly some of them like telecommunications, business services and utilities, rank low in terms of imported content of exports as they use fewer intermediate inputs and their involvement in FVCs typically occurs through value added incorporated in manufactured goods.

The identification of the industries which not only contribute more to the gross exports, but also that are less susceptible of sufering foreign shocks is important for policy design in fields such as industrial development, or trade and investment promotion.

4.4 Revealed Comparative Advantage of Spanish Industries

The previous discussion naturally leads to a revised notion of a country-sector’s revealed comparative advantage. The traditional definition of a country-sector’s revealed comparative advantage (TRCA) is based on a country’s gross exports (e) and is defined as as the share of that country-sector’s gross exports in the country’s total gross exports relative to that sector’s gross exports from all countries as a share of the world total gross exports.

\[T R C A _ {i} ^ {c} = \frac {\frac {e _ {i} ^ {c}}{\sum_ {i = 1} ^ {N} e _ {i} ^ {c}}}{\frac {\sum_ {c = 1} ^ {G} e _ {i} ^ {c}}{\sum_ {i = 1} ^ {N} \sum_ {c = 1} ^ {G} e _ {i} ^ {c}}},\tag{2}\]

where indicates sector and indicates country. When the RCA exceeds one, the country is said to have a revealed comparative advantage in that sector; when the RCA is below one, the country is said to have a revealed comparative disadvantage in that sector.

The traditional RCA ignores both domestic production sharing and international production sharing. To be more specific, it ignores the fact that a country-sector’s value added may be exported indirectly via the country’s exports in other sectors. Second, it also ignores the fact that a country-sector’s gross exports partly reflect foreign contents (which show up both FVA and a portion of DC items). A conceptually correct measure of comparative advantage needs to exclude foreign-originated value added and pure dou ble counted terms in gross exports but include indirect exports of a sector’s value added through other sects of the exporting country. Taking this into account, we define a new measure of a country sector’s revealed comparative advantage (NRCA) as the share of a country-sector’s domestic content dc in exports in the country’s total domestic content in exports relative to that sector’s domestic content in exports from all countries as a share of global domestic value added in exports.

\[N R C A _ {i} ^ {c} = \frac {\frac {d c _ {i} ^ {c}}{\sum_ {i = 1} ^ {N} d c _ {i} ^ {c}}}{\frac {\sum_ {r = 1} ^ {G} d c _ {i} ^ {r}}{\sum_ {i = 1} ^ {N} \sum_ {r = 1} ^ {G} d c _ {i} ^ {r}}}\tag{3}\]

In Figure 12 we depict the evolution of both measures for selected sectors of the Spanish economy¹⁴: Textiles, Paper and Printing, Refined Petroleum, Chemical Products, Basic Metals and Transport Equipment.

Paper and Printing, Chemical Products and Basic Metals are three cases where the traditional relative comparative advantage would indicate that we have a comparative advantage in them while the new indicator would indicate the opposite. At the same time, the three of them follow very diferent paths. For Paper and Printing the paths of both indicators are divergent, particularly, the traditional indicator remains more or less constant over the last decade while the new indicator decreases, indicating a rapid loss in competitive advantage that would have been lost were we not to able to distinguish between the gross exports and the domestic content in the gross exports. For Chemical Products and Basic Metals the paths of both indicators are similar, and rising. Particularly, the new indicators of comparative advantage which indicated a disadvantage at the beginning of the 2000, have kept rising despite the crisis and are near to indicating comparative advantage in these sectors. This, together with the evidence on the contribution to the Exports while not depending excessively on the foreign components, suggest that they are attractive sectors on which we should focus more in the future.

Refined Petroleum and Transport Equipment are two sectors of particular interest since they are two of the largest contributors to Gross Exports, but their exports contain a high percentage of foreign value added, that is, value that does not remain in Spain. On the one hand, the comparative advantage indicators of Refined Petroleum is growing without interruption since 2005, which indicates that despite the high presence of foreign value content in the exports, the participation in the global value chain is also fostering the domestic value added in the exports, and hence why the new indicator is not falling below the traditional indicators. On the other hand, the comparative advantage indicators of Transport Equipment present a decrease since 2004. Furthermore, the new indicator being lower than the traditional one indicates that the presence of foreign value content in exports is not fostering the domestic value content in them. We should aim to decrease the foreign content in both while maintaining our export quotas.

¹⁴See the appendix for the evolution of both measures for all the sectors of the economy

Finally, Textile Products is a sector where both indicators have similar values until the start of the crisis, when the new indicator starts to rise much faster than the traditional one, indicating an strengthening of the domestic content in the gross exports. This combined with the fact that it happened during the crisis years and the contribution of the exports are above the sectorial mean, imply that it is a sector on which we rely in times of crisis and that can help us grow, and therefore should be promoted.

(a) Textiles

(b) Paper & Printing

(c) Refined Petroleum

(d) Chemical Products

(e) Basic Metals

(f) Transport Equipment Figure 12: Evolution RCA Sectors Spain

5 GVCs as an stabilizer in times of crisis

In the previous section, we have argued that since the DVA embodied in gross exports is counter-cyclical, the change in gross exports is not a suficient statistic in order to determine their contribution to GDP growth.

Table 9 provides the growth rates of output, gross export, domestic value added and foreign value added from 2007 to 2011 of France, Germany, Italy, Spain and United Kingdom. The year 2009 is a striking example; according to WIOD data, gross exports contributed for 19.88% to the 2009 contraction of nominal GDP. The correct computation uses the domestic value added content of exports (DVA) in the place of gross export, and results in a negative contribution to GDP growth worth 13.63% percentage points. The contraction in Spanish DVA was contained thanks to the increase in the DVA share. The latter (5.36%) was mainly mirrored in a reduction of the foreign value added embodied in gross exports (4.24%). This is consistent with Spanish exporters adjusting their production function in favour of relatively inflexible inputs (domestic labour and capital) and at the disadvantage of more flexible inputs such as imported intermediates (and inventories thereof). Part of the explanation is also the sharp contraction in prices of imported oil and other raw materials.

The increased participation in GVCs implies we are more open to foreign shocks, but at the same time it provides relief in cases of domestic shocks. In 2009 the financial crisis hit most of the developed economies, which shows up not only in the contraction of al the economies considered, but also in the heavy contraction through the FVA component. The crisis, while temporary for some countries, was more permanent for some others such as Spain, as can be seen by the negative output growth in 2010. Exports, however, grew by 9.70%, thanks to our presence in GVCs and the growth in FVA exports. The presence in GVCs has palliated the efects of the domestic crisis felt in Europe after 2009 as can be seen by the more rapid growth of gross exports than output, and the increased growth in foreign value added.

Table 9: Growth in Output, Exports, Domestic VA and Foreign VA across Europe (%)

In Table 10 we present the impact (first-round efect only) on Spanish gross exports and domestic value added when there is a 10 percent increase in final demand in selected areas of the world. The last column is the ratio of the first two, namely the domestic value added intensity (the Spanish GDP content of one dollar of exports). The year of reference utilized has been the last one in the sample, 2011. Since composition efects have not been taken into account, the results have to be interpreted with caution.

A 10 percent increase in world final demand in 2011 would led to an increase of Spanish gross exports worth 2.281 percentage points of GDP whereas the domestic value added in GDP grew by 2.337 percent. The analysis by areas shows the dependence of Spain on the final demand coming from inside the EU15, and particularly from our close neighbours who represent half of the increase in exports. At the same time, it reveals a very small impact from increased internal demand in the rest of the European Countries, where despite having a good DVA/Gross Exports ratio, the increase in exports is only 0.1%. Finally, it is worth noticing the impact of an increase in the final demand in areas like Asia, countries which are rapidly gaining terrain in the worlds gross export that have a high activation ratio for Spanish exports.

Table 10: Impact on Spanish Exports of a 10% increase in selected areas final internal demand

6 Conclusions

As production becomes more and more internationally fragmented, countries become more vulnerable to foreign shocks but also more resilient to domestic shocks. This paper has analyzed the role that global value chains (GVC) have played during the sovereign debt crisis in Europe, and the key policy challenges that Spain faces.

The analysis confirmed earlier findings about the contrasting performance of Spanish firms. Spain’s largest firms are remarkably competitive and well managed to maintain their share in world exports during the crisis, but the overall export share of Spain is low by international standard due to the abundance of small firms that only serve the domestic market. Moreover, a careful analysis of the latest available input-output data reveals that the share of domestic value added (DVA) is low compared to the corresponding shares for our main European trading partners. This evidence points at a relatively unfavorable position of Spanish exporters in the global value chains.

GVCs have the capacity to act as built-in stabilizers following economic downturns if they are promoted in the right industries. In the second part of the analysis we compute the first round impact on Spanish gross exports and domestic value added of a 10% increase in final demand in selected areas of the world, concluding that Spain’s gross exports would be worth 2.28% of GDP while the domestic value added from export would grow by 2.34%. Even though this exercise is only a partial equilibrium analysis, it provides us with insight of how Spanish GDP can be activated through Spanish exports by foreign demand. The geographical analysis shows the dependence Spain has on the final demand coming from inside the EU15, and particularly from France and Germany who represent half of the potential increase in exports, but also highlight the potential impact of an increase in the final demand in Asia, which would be the most efective in activating domestic value added exports in Spain.

A sustained improvement in the export performance of Spanish firms requires both an increase in the number of large firms and a reorientation of production towards the sectors in which Spanish producers manage to capture a relatively large share of total value added. In the second part of the analysis we propose novel indicators of revealed comparative advantage based on the domestic value-added content of exports at the industry level. A comparison between these novel indicators and standard indicators of revealed comparative advantage (based on the gross value of exports) reveal significant diferences, suggesting substantial gains from an appropriate reorientation of production. Our results provide a clear indication that future reforms should focus on measures that stimulate firm growth and remove barriers to the sectoral reallocation of firms and workers.

References

1. Altomonte, C. and T. Aquilante (2012). The EU-EFIGE/Bruegel-Unicredit Dataset. Bruegel working paper.
2. Altomonte, C., T. Aquilante, and G. Ottaviano (2012). The triggers of competitiveness: The EFIGE cross-country report. Blueprint (17).
3. Altomonte, C., F. di Mauro, G. Ottaviano, A. Rungi, and V. Vicard (2012, January). Global value chains during the great trade collapse: a bullwhip efect? Working Paper Series 1412, European Central Bank.
4. Altomonte, C., G. B. Navaretti, F. di Mauro, and G. Ottaviano (2011). Assessing competitiveness: How firm-level data can help. Policy Contributions (643).
5. Amador, J. and R. Stehrer (2014). Portuguese Exports in the Global Value Chains. Economic Bulletin and Financial Stability Report Articles.
6. Antr`as, P., R. Segura-Cayuela, and D. Rodríguez-Rodríguez (2010). Firms in international trade, with an application to spain. In SERIEs Invited Lecture at the XXXV Simposio de la Asociación Española de Economía.
7. Bernard, A. B. and J. Bradford Jensen (1999). Exceptional exporter performance: cause, efect, or both? Journal of International Economics 47 (1), 1–25.
8. Bernard, A. B., J. B. Jensen, S. J. Redding, and P. K. Schott (2011). The empirics of firm heterogeneity and international trade. CEP Discussion Papers (dp1084).
9. Cappariello, R. and A. Felettigh (2014). How does Foreign Demand activate Domestic Value Added? A Dashboard for the Italian Economy. mimeo.
10. Clerides, S. K., S. Lach, and J. R. Tybout (1998). Is learning by exporting important? micro-dynamic evidence from colombia, mexico, and morocco. The Quarterly Journal of Economics 113 (3), 903–947.
11. Crespo, A. (2012). Trade, innovation and productivity: A quantitative analysis of Europe. Working Paper EFIGE (62).
12. di Giovanni, J. and A. A. Levchenko (2009). International trade and aggregate fluctuations in granular economies. Working Papers (585).
13. Eaton, J., S. Kortum, and F. Kramarz (2004, May). Dissecting Trade: Firms, Industries, and Export Destinations. American Economic Review 94 (2), 150–154.
14. Fontagn`e, L. and G. Gaulier (2010). Performance `a l’ Exportation de la France et de l’ Allemagne. Conseil d’ analyse economique, paris.
15. Helpman, E., M. J. Melitz, and S. R. Yeaple (2004). Export versus fdi with heterogeneous firms. American Economic Review 94 (1), 300–316.
16. Koopman, R., Z. Wang, and S.-J. Wei (2014, February). Tracing Value-Added and Double Counting in Gross Exports. American Economic Review 104 (2), 459–94.
17. Krugman, P. (1994). Competitiveness: A dangerous obsession. Foreign Afairs, vol 73(2).
18. Melitz, M. J. (2003). The Impact of Trade on Intra-Industry Reallocations and Aggregate Industry Productivity. Econometrica 71 (6), 1695–1725.
19. Porter, M. E. (1990). The Competitive Advantage of Nations. Free Press, New York.
20. Rubini, L., K. Desmet, F. Piguillem, and A. Crespo (2012). Breaking down the barriers to firm growth in Europe: The fourth EFIGE policy report. Number 744 in Bruegel Blueprints. Bruegel.
21. Timmer, M. R. S., G. J. de Vries, B. Los, and H. Dietzenbacher (2014). The World Input-Output Database: Content, Concepts and Applications. GGDC Research Memorandum GD-144.
22. UNCTAD (2013). Global value chains: Investment and trade for development. World Investment Report.
23. Wang, Z., S.-J. Wei, and K. Zhu (2013, November). Quantifying International Production Sharing at the Bilateral and Sector Levels. NBER Working Papers 19677, National Bureau of Economic Research, Inc.

Appendix A — The Algebra of Koopman et al. (2014) Decomposition

In this section we briefly describe the application of the decomposition of gross exports developed by Koopman et al. (2014) to the WIOD Data. In our analysis there are three countries, Spain (H) , the EU15¹⁵ (F) and the Rest of the World¹⁶ (RW) and in each of the three countries there are industries.

All gross exports of country s are used as an intermediate and final good abroad, according the following definition:

\[E ^ {s *} = \sum_ {r \neq s} ^ {C} E ^ {s r} = \sum_ {r \neq s} ^ {C} \left(A ^ {s r} X ^ {r} + Y ^ {s r}\right)\]

where is a block input-output coeficient matrix, for any is the N×1 gross output vector of country s; is a N×1 vector that gives the demand in country in countries s and for goods produced in country is the vector of N products exported by country s to the 3 regions; and is the N×1 vector of gross exports from country s to r.

These exports can be fully decomposed into various value added and double counted components described in Figure 8 as follows

\[\begin{array}{r c l} E ^ {H *} & = & V ^ {H} B ^ {H H} Y ^ {H F} + V ^ {H} B ^ {H H} Y ^ {H R W} \qquad (1) \\ & + & V ^ {H} B ^ {H F} Y ^ {F F} + V ^ {H} B ^ {H R W} Y ^ {R W R W} \qquad (2) \\ & + & V ^ {H} B ^ {H F} Y ^ {F R W} + V ^ {H} B ^ {H R W} Y ^ {R W F} \qquad (3) \\ & + & V ^ {H} B ^ {H F} Y ^ {F H} + V ^ {H} B ^ {H R W} Y ^ {R W H} \qquad (4) \\ & + & V ^ {H} B ^ {H F} A ^ {F H} X ^ {H} + V ^ {H} B ^ {H R W} A ^ {R W H} X ^ {H} \qquad (5) + (6) \\ & + & V ^ {F} B ^ {F H} Y ^ {H F} + V ^ {F} B ^ {F H} Y ^ {H R W} + V ^ {R W} B ^ {R W H} Y ^ {H R W} + V ^ {R W} B ^ {R W H} Y ^ {H F} \qquad (7) \\ & + & V ^ {F} B ^ {F H} A ^ {H F} X ^ {F} + V ^ {F} B ^ {F H} A ^ {H R W} X ^ {R W} + V ^ {R W} B ^ {R W H} A ^ {H R W} X ^ {R W} + V ^ {R W} B ^ {R W H} A ^ {H F} X ^ {F} \quad (8) + (9) \end{array}\]

The terms 5 and 7 on the RHS of the equation, they can be decomposed easily into terms (5) and (6), and (8) and (9) respectively, using the following identity: . Then it is easy to show that

¹⁵Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, Netherlands, Portugal, Sweden and United Kingdom

¹⁶Brazil, Bulgaria, Canada, China, Cyprus, Czech Republic, Estonia, Hungary, India, Indonesia, Japan, Korea, Latvia, Lithuania, Malta, Mexico, Poland, Romania, Russia, Slovak Republic, Slovenia, Taiwan, Turkey, United States and RoW

\[\begin{array}{l l l} V ^ {H} B ^ {H F} A ^ {F H} X ^ {H} & + & V ^ {H} B ^ {H R W} A ^ {R W H} X ^ {H} = \\ & = & V ^ {H} B ^ {H F} A ^ {F H} (I - A ^ {H H}) ^ {- 1} Y ^ {H H} + V ^ {H} B ^ {H R W} A ^ {R W H} (I - A ^ {H H}) ^ {- 1} Y ^ {H H} \\ & + & V ^ {H} B ^ {H F} A ^ {F H} (I - A ^ {H H}) ^ {- 1} E ^ {H F} + V ^ {H} B ^ {H R W} A ^ {R W H} (I - A ^ {H H}) ^ {- 1} E ^ {H R W} \\ & + & V ^ {H} B ^ {H F} A ^ {F H} (I - A ^ {H H}) ^ {- 1} E ^ {H R W} + V ^ {H} B ^ {H R W} A ^ {R W H} (I - A ^ {H H}) ^ {- 1} E ^ {H F} \end{array}\tag{5}\]

(6)

\[\begin{array}{l l l} V ^ {F} B ^ {F H} A ^ {H F} X ^ {F} & + & V ^ {F} B ^ {F H} A ^ {H R W} X ^ {R W} + V ^ {R W} B ^ {R W H} A ^ {H R W} X ^ {R W} + V ^ {R W} B ^ {R W H} A ^ {H F} X ^ {F} = \\ & = & V ^ {F} B ^ {F H} A ^ {H F} (I - A ^ {F F}) ^ {- 1} Y ^ {F F} + V ^ {F} B ^ {F H} A ^ {H R W} (I - A ^ {R W R W}) ^ {- 1} Y ^ {R W R W} \\ & + & V ^ {R W} B ^ {R W H} A ^ {H R W} (I - A ^ {R W R W}) ^ {- 1} Y ^ {R W R W} + V ^ {R W} B ^ {R W H} A ^ {H F} (I - A ^ {F F}) ^ {- 1} Y ^ {F F} \\ & & \\ & + & V ^ {F} B ^ {F H} A ^ {H F} (I - A ^ {F F}) ^ {- 1} E ^ {F H} + V ^ {F} B ^ {F H} A ^ {H R W} (I - A ^ {R W R W}) ^ {- 1} E ^ {R W H} \\ & + & V ^ {R W} B ^ {R W H} A ^ {H R W} (I - A ^ {R W R W}) ^ {- 1} E ^ {R W H} + V ^ {R W} B ^ {R W H} A ^ {H F} (I - A ^ {F F}) ^ {- 1} E ^ {F H} \\ & + & V ^ {F} B ^ {F H} A ^ {H F} (I - A ^ {F F}) ^ {- 1} E ^ {F R W} + V ^ {F} B ^ {F H} A ^ {H R W} (I - A ^ {R W R W}) ^ {- 1} E ^ {R W F} \\ & + & V ^ {R W} B ^ {R W H} A ^ {H R W} (I - A ^ {R W R W}) ^ {- 1} E ^ {R W F} + V ^ {R W} B ^ {R W H} A ^ {H F} (I - A ^ {F F}) ^ {- 1} E ^ {F R W} \end{array}\tag{8}\]

(9)

where:

is the N×N Leontief inverse matrix, which is the total requirement matrix giving the amount of gross output produced in country s required for a one-unit increase in final demand in country r.

is the 1xN direct VA coeficient vector, . Each element of gives the share of direct domestic VA in total output.

While the algebra to obtain the previous equation may be a bit tedious, expressing a country’s gross exports as the sum of these nine terms is very useful. We try to explain briefly their economic interpretations.

Terms (1) and (2) are the direct value added exports, i.e., the source country value added absorbed by direct importer, country r. Term (3) is, instead, its value added exported to country r and, after some processing in finally absorbed in a third country t. Terms (4) and (5) include source country’s value added which is first exported but return in both final and intermediate imports to be consumed or re-exported by country s. The sum of the first five terms corresponds to the definition of “ domestic value added exports”. Terms (7) and (8) represent foreign value added in the source country’s exports, including foreign value added embodied in both final and intermediate products. Term (6) and (9) are the two pure double counted terms that sum up the double counted share of two way intermediate trade from all bilateral routes.

Appendix B — Tables and Figures

Continued on next page

Table B.1: List of WIOD Industries

Table B.2: Gross Export Decomposition in France

Table B.3: Participation in Fragmented Production in France

Table B.4: Gross Export Decomposition in Germany

Table B.5: Participation in Fragmented Production in Germany

Table B.6: Gross Export Decomposition in Italy

Table B.7: Participation in Fragmented Production in Italy

Table B.8: Gross Export Decomposition in Spain

Table B.9: Participation in Fragmented Production in Spain

Table B.10: Gross Export Decomposition in UK

Table B.11: Participation in Fragmented Production in UK

(a) c1

(b) c2

(c) c3

(d) c4

(e) c5

(f) c6

(g) c7

(h) c8 (i) c9

(j) c10

(k) c11

(l) c12 Figure B.1: Evolution RCA Sectors Spain

(a) c13

(b) c14

(c) c15

(d) c16

(e) c17

(f) c18

(g) c19

(h) c20 (i) c21

(j) c22

(k) c23

(l) c24 Figure B.2: Evolution RCA Sectors Spain

(a) c25

(b) c26

(c) c27

(d) c28

(e) c29

(f) c30

(g) c31

(h) c32

(i) c33

(j) c34 Figure B.3: Evolution RCA Sectors Spain

References

1. 2014-09: “The Role of Global Value Chains during the Crisis: Evidence from Spanish and European Firms”, Aranzazu Crespo y Marcel Jansen.

References

1. 2014-08: “Can Fixed-Term Contracts Put Low Skilled Youth on a Better Career Path? Evidence from Spain”, José I. García Pérez, Ioana Marinescu y Judit Vall Castello.

References

1. 2014-07: “Gender Peer Efects in School, a Birth Cohort Approach”, Antonio Ciccone y Walter Garcia-Fontes.

References

1. 2014-06: “Delaying the Normal and Early Retirement Ages in Spain: Behavioural and Welfare Consequences for Employed and Unemployed Workers”, Alfonso R. Sánchez, J. Ignacio García-Pérez y Sergi Jiménez-Martín.

References

1. 2014-05: “FEDEA and IAE (CSIC)”, FEDEA and IAE (CSIC), Jesús Fernández-Huertas Moraga.

References

1. 2014-04: “The Incentive Efects of Minimum Pensions: extended version”, Sergi Jiménez-Martín.

References

1. 2014-03: “A Practitioners' Guide to Gravity Models of International Migration”, Michel Beine, Simone Bertoli y Jesús Fernández-Huertas Moraga.

References

1. 2014-02: “L'auberge Espagnole y el Apartamento Francés: los Determinantes del Aprendizaje del Francés en España”, Brindusa Anghel y Maia Güell.

References

1. 2014-01: “Temporary Intergenerational Mobility and the Informational Content of Surnames” Maia Güell, José V. Rodríguez Mora y Christopher I. Telmer.

References

1. 2013-25: “Informal Care and Intergenerational Transfers in European Countries”, Sergi Jiménez-Martín y Cristina Vilaplana Prieto.

References

1. 2013-24: “Inference Based on SVARs Identied with Sign and Zero Restrictions: Theory and Applications”, Jonas E. Arias, Juan F. Rubio-Ramírez y Daniel F. Waggoner.

References

1. 2013-23: “Estimating Dynamic Equilibrium Models with Stochastic Volatility”, Jesús Fernández-Villaverde, Pablo Guerrón-Quintana y Juan F. Rubio-Ramírez.

References

1. 2013-22: “Perturbation Methods for Markov-Switching DSGE Models”, Andrew Foerster, Juan Rubio-Ramirez, Dan Waggoner y Tao Zha.

References

1. 2013-21: “Do Spanish informal caregivers come to the rescue of dependent people with formal care unmet needs?”, Sergi Jiménez-Martín y Cristina Vilaplana Prieto.

References

1. 2013-20: “When Credit Dries Up: Job Losses in the Great Recession”, Samuel Bentolila, Marcel Jansen, Gabriel Jiménez y Sonia Ruano.

References

1. 2013-19: “Efectos de género en las escuelas, un enfoque basado en cohortes de edad”, Antonio Ciccone y Walter Garcia-Fontes.

References

1. 2013-18: “Oil Price Shocks, Income, and Democracy“, Markus Brückner , Antonio Ciccone y Andrea Tesei.

References

1. 2013-17: “Rainfall Risk and Religious Membership in the Late Nineteenth-Century US”, Philipp Ager y Antonio Ciccone.

References

1. 2013-16: “Immigration in Europe: Trends, Policies and Empirical Evidence”, Sara de la Rica, Albrecht Glitz y Francesc Ortega.

References

1. 2013-15: “The impact of family-friendly policies on the labor market: Evidence from Spain and Austria”, Sara de la Rica y Lucía Gorjón García.

References

1. 2013-14: “Gender Gaps in Performance Pay: New Evidence from Spain”, Sara de la Rica, Juan J. Dolado y Raquel Vegas.

References

1. 2013-13: “On Gender Gaps and Self-Fulfilling Expectation: Alternative Implications of Paid-For Training”, Juan J. Dolado, Cecilia García-Peñalosa y Sara de la Rica.

References

1. 2013-12: “Financial incentives, health and retirement in Spain”, Pilar García‐Gómez, Sergi Jiménez‐Martín y Judit Vall Castelló.

References

1. 2013-11: “Gender quotas and the quality of politicians”, Audinga Baltrunaite, Piera Bello, Alessandra Casarico y Paola Profeta.

References

1. 2013-10: “Brechas de Género en los Resultados de PISA: El Impacto de las Normas Sociales y la Transmisión Intergeneracional de las Actitudes de Género”, Sara de la Rica y Ainara González de San Román.

References

1. 2013-09: “¿Cómo escogen los padres la escuela de sus hijos? Teoría y evidencia para España”, Caterina Calsamiglia, Maia Güell.

References

1. 2013-08: “Evaluación de un programa de educación bilingüe en España: El impacto más allá del aprendizaje del idioma extranjero”, Brindusa Anghel, Antonio Cabrales y Jesús M. Carro.

References

1. 2013-07: “Publicación de los resultados de las pruebas estandarizadas externas: ¿Tiene ello un efecto sobre los resultados escolares?”, Brindusa Anghel, Antonio Cabrales, Jorge Sainz e Ismael Sanz.

References

1. 2013-06: “DYPES: A Microsimulation model for the Spanish retirement pension system”, F. J. Fernández-Díaz, C. Patxot y G. Souto.

References

1. 2013-05: “Vertical differentiation, schedule delay and entry deterrence: Low cost vs. full service airlines”, Jorge Validoa, M. Pilar Socorroa y Francesca Medda.

References

1. 2013-04: “Dropout Trends and Educational Reforms: The Role of the LOGSE in Spain”, Florentino Felgueroso, María Gutiérrez‐Domènech y Sergi Jiménez‐Martín.

References

1. 2013-03: “The European Crisis and Migration to Germany: Expectations and the Diversion of Migration Flows”, Simone Bertoli, Herbert Brücker y Jesús Fernández-Huertas Moraga.