Cling together, swing together: The contagious effects of COVID-19 on developing countries through global value chains

Pahl, Stefan; Brandi, Clara; Schwab, Jakob; Stender, Frederik Working Paper Cling together, swing together: The contagious effects of COVID-19 on developing countries through global value chains Discussion Paper, No. 21/2020 Provided in Cooperation with: German Institute of Development and Sustainability (IDOS), Bonn Suggested Citation: Pahl, Stefan; Brandi, Clara; Schwab, Jakob; Stender, Frederik (2020) : Cling together, swing together: The contagious effects of COVID-19 on developing countries through global value chains, Discussion Paper, No. 21/2020, ISBN 978-3-96021-133-4, Deutsches Institut für Entwicklungspolitik (DIE), Bonn, https://doi.org/10.23661/dp21.2020 This Version is available at: https://hdl.handle.net/10419/227140 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Discussion Paper 21/2020 Cling Together, Swing Together Stefan Pahl Clara Brandi Jakob Schwab Frederik Stender The Contagious Effects of COVID-19 on Developing Countries through Global Value Chains Cling together, swing together The contagious effects of COVID-19 on developing countries through global value chains Stefan Pahl Clara Brandi Jakob Schwab Frederik Stender Bonn 2020 Discussion Paper / Deutsches Institut für Entwicklungspolitik ISSN (Print) 1860-0441 ISSN (Online) 2512-8698 Except as otherwise noted this publication is licensed under Creative Commons Attribution (CC BY 4.0). You are free to copy, communicate and adapt this work, as long as you attribute the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) and the author. Die Deutsche Nationalbibliothek verzeichnet diese Publikation in der Deutschen Nationalbibliografie; detaillierte bibliografische Daten sind im Internet über http://dnb.d-nb.de abrufbar. The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data is available on the Internet at http://dnb.d-nb.de. ISBN 978-3-96021-133-4 (printed edition) DOI:10.23661/dp21.2020 Printed on eco-friendly, certified paper Dr Stefan Pahl is a researcher in the research programme “Growth and Development” at the German Institute for Global and Area Studies (GIGA). Email: [email protected] Dr Clara Brandi is a senior researcher with the research programme “Transformation of Economic and Social Systems” at the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE). Email: [email protected] Dr Jakob Schwab is a researcher with the research programme “Transformation of Economic and Social Systems” at the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE). Email: [email protected] Dr Frederik Stender is a researcher with the research programme “Transformation of Economic and Social Systems” at the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE). Email: [email protected] Published with financial support from the Federal Ministry for Economic Cooperation and Development (BMZ) © Deutsches Institut für Entwicklungspolitik gGmbH Tulpenfeld 6, 53113 Bonn  +49 (0)228 94927-0  +49 (0)228 94927-130 Email: [email protected] www.die-gdi.de Abstract This paper aims at estimating the economic vulnerability of developing countries to disruptions in global value chains (GVCs) due to the COVID-19 pandemic. It uses data on trade in value-added for a sample of 12 developing countries in sub-Saharan Africa, Asia and Latin America to assess their dependence on demand and supply from the three main hubs China, Europe, and North America. Using first estimates on COVID-19-induced changes in production and sectoral final demand, we obtain an early projection of the GDP effect during the lockdowns that runs through trade in GVCs. Our estimates reveal that adverse demand-side effects reduce GDP by up to 5.4 per cent, and that collapsing foreign supply is responsible for a drop in GDP of a similar magnitude. Overall, we confirm conjecture that the countries most affected are those highly integrated into GVCs (Southeast Asian countries). We argue, however, that these countries also benefit from a welldiversified portfolio of foreign suppliers, leading to a cushioning of economic downswing from adverse supply-side spillovers, because COVID-19 stroke major hubs at different times during the first wave in early 2020. Moreover, despite expected hazardous home market effects, sub-Saharan Africa’s GDP appears to be comparatively less affected though GVCs due to a lack of intensive supplyand demand-side dependencies. Key words: COVID-19, global value chains, input-output analysis, international trade, supplyand demand-side dependency, shock spillover Contents Abstract Abbreviations 1 Introduction 1 2 Related literature 3 3 Method and data 7 4 Demand-side vulnerability 10 5 Supply-side vulnerability 16 6 Discussion and outlook 19 References 23 Appendix 27 Tables Table 1: Demand-side dependency by region (as a percentage of GDP) 11 Table 2: Demand-side dependency by product group (as a percentage of GDP generated in respective region) 13 Table 3: Sectoral downturns in final demand (in per cent) 15 Table 4: Demand-induced value-added effect (as a percentage of GDP) 16 Table 5: Supply-side dependency (as a percentage of GDP) 17 Table 6: Decline in industrial production (in per cent) 18 Table 7: Supply-induced value-added effect (as a percentage of GDP) 19 Appendix tables Table A1: Sectoral mapping 27 Table A2: Shares of value-added generated by final demand in three hubs 28 Abbreviations ACP African, Caribbean and Pacific ASEAN Association of Southeast Asian Nations COVID coronavirus disease EORA Eora Global Value Chain Database EU European Union FRED Federal Reserve Economic Data GDP gross domestic product GVC global value chain ICIO inter-country input-output ILO International Labour Organization ISIC International Standard Industrial Classification of All Economic Activities NAFTA North American Free Trade Agreement NBS National Bureau of Statistics of China OECD Organisation for Economic Cooperation and Development TiVA trade in value-added UK United Kingdom UNIDO United Nations Industrial Development Organization US United States USD United States dollar WIOD World Input-Output Database WTO World Trade Organization The contagious effects of COVID-19 on developing countries through global value chains German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 1 1 Introduction The entire world has been hit hard by the consequences of the COVID-19 pandemic and the political, social, and economic measures to contain it. While the long-term effects of the disease still remain unpredictable, some countries are currently being affected more strongly by it, and there is considerable heterogeneity in underlying causal chains. For instance, healthcare systems in developing countries are generally less prepared to cope with increasing numbers of infections. Moreover, adding to the immediate adverse economic consequences due to countries’ own lockdowns, many developing countries are increasingly facing additional hazards in the highly interconnected global economy stemming from their integration into global value chains (GVCs). More specifically, despite the positive developmental effects associated with participation in GVCs in normal times (see, for example, Kummritz, Taglioni, & Winkler, 2017; Pahl & Timmer, 2020; World Bank, 2020), the pandemic has transformed this channel into a menace. In this paper, we explore the economic vulnerability of developing countries to COVID-19induced demand and supply shocks occurring in the key hubs of GVCs. Using global inputoutput tables, we map countries’ value-added to final demand in specific consumer markets (as in Johnson & Noguera, 2017) and to the production within specific value chains (in accordance with Los, Timmer, & de Vries, 2015; Pahl, Timmer, Gouma, & Woltjer, 2019; and Timmer, Los, Stehrer, & De Vries, 2013). This allows us to provide first estimates on potential effects of the demand and supply shock on gross domestic product (GDP), running through trade in GVCs. As the nature of the COVID-19 crisis is different to previous global economic crises insofar as sectors of household consumption are affected very heterogeneously due to social distancing measures, and that some supply chains may be fully interrupted for certain periods of time, the expected hazardous effects from this inflicted on developing countries remain an empirical puzzle. A key characteristic of GVCs is that firms are linked to other producers (of intermediate goods) and consumers through production networks spanning multiple countries. A demand shock in a specific consumer market therefore affects all (foreign) upstream suppliers delivering to this market, which goes well beyond a country’s direct trade partners (in the spirit of Bems, Johnson, & Yi, 2011; and Johnson & Noguera, 2017). Furthermore, a shock to a specific key input supplier can cause major bottlenecks in production, which goes well beyond directly linked firms, but which can disrupt production along the entire value chain. Both types of shocks are currently looming large due to COVID-19-related lockdowns in many places around the world, in particular in the major demand and supply hubs China, Europe and North America. Starting from a local concentration on China, COVID-19 initially put a strain on the production capacities of a crucial link – but yet only one in a number of links – in international production networks. With the disease’s evolution from epidemic to pandemic proportions, however, disruptions are now also being felt across the board, from simple to complex GVCs due to COVID-19 containment policies in manufacturing hubs like the European Union (EU) and North America, leading to even more far-reaching repercussions via trade-based contagion links (Baldwin & Freeman, 2020b; Seric, Gorg, Mosle, & Windisch, 2020). While first analyses are available on the repercussions for trade in valueadded (Baldwin, 2020; Baldwin & Freeman, 2020a; Bonadio, Huo, Levchenko, & PandalaiNayar, 2020; Guan et al., 2020), we still lack a detailed understanding of how the pandemic transmits through GVCs – especially in developing countries. Stefan Pahl et al. 2 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) We implement our analysis using a new set of global input-output tables from Pahl et al. (2019), constructed for country-specific analyses of a set of lower-income countries in the world economy. These data are constructed using highly country-specific sources. In particular, the time series of national input-output tables are built up from national supply and use tables or social accounting matrices; from production data (value-added, gross output) with high sectoral detail and yearly variation; information on national accounts, such as final consumption; as well as detailed trade data. The construction strictly follows the methodologies of the World Input-Output Database (WIOD; Timmer, Dietzenbacher, Los, Stehrer, & De Vries, 2015) and can therefore be used in conjunction for global analyses. This dataset aims to represent developing countries from all world regions, while adhering to high demands in terms of data requirements for the construction of a global input-output database. For the purpose of providing an initial view of the effects of the COVID-19 crisis running through GVCs, our sample in this paper consists of all 12 lowand middle-income countries in the major emerging regions Africa, Asia, and Latin America available in that database. To identify the adverse effects of final demand contractions in hubs on developing countries, we combine these input-output data with estimates of final demand changes at the sector level and COVID-19-induced cuts in industrial production. While precise pandemic-induced economic effects are still uncertain, we provide a ballpark figure for the potential size of the trade-induced shocks to developing countries due to the ongoing crisis. With regard to the demand shock, we find that there are stark differences across the 12 developing countries, particularly in terms of the importance of hub markets for their own production and with regard to the sectors that they deliver to. In Bangladesh, for example, 6.1 (2.7) per cent of GDP depends on final demand in Europe (North America), of which more than 90 per cent stem from final demand for textiles. In Vietnam, an even greater share, namely 8.1 (9.4) per cent of GDP, depends on final demand in Europe (North America). At the same time, however, the Vietnamese export sector is more diversified, with only roughly one-quarter to one-third of the value-added being concentrated in the production of one final demand sector at maximum. Combining these dependencies with estimated changes in final demand by sector and region, we find, for instance, that 4.5 per cent of Bangladesh’s GDP are at risk alone through the demand shock and demand shifts in Europe during the starkest time of the lockdown. One per cent of GDP is at risk due to the demand collapse in North America. For Vietnam, overall 3 per cent of GDP are at risk through GVCs, almost equally due to the declining demand from Europe and North America. Sub-Saharan African countries are much more dependent on their home markets and, as such, are less affected by changes in final demand in the major hubs. On the supply-side, we take into consideration that shocks to supply within a value chain both upstream and downstream may cause the entire value chain to break down. We therefore assume for our projections that – in sectoral production for which more than a threshold share of inputs from a particular region (directly or indirectly) is needed – these inputs cannot be substituted for by inputs from other regions; the production thus has to stand still if the required inputs are missing. The resulting shock analysis reveals that not only a lockdown in China may cause bottlenecks in production: for example, while impressive shares of Vietnamese (7.5 per cent), Indonesian (8.7 per cent), and Malaysian (11.1 per cent) GDPs are generated in GVCs in which China is a key supplier (that is, generates more than 5 per cent of inputs in value-added terms), this holds true for only 0.4 per cent of GDP in Bangladesh. Southeast Asian countries, however, not only rely heavily on Chinese intermediates but to a similar extent also on European ones, suggesting a well- The contagious effects of COVID-19 on developing countries through global value chains German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 9 recent model-based analysis of the supply-shock running through the COVID-19 crisis. One might argue that such substitution patterns occur in the longer run but are potentially less likely in the short run. Our approach of using a threshold of dependence that leads to a disruption of a GVC therefore needs to be carefully interpreted as a short-run ballpark figure if substitution of relatively important suppliers in GVCs (defined by the threshold) is not possible.5 We would further like to stress that our supply-side approach accounts for upstream as well as downstream dependencies, in contrast to supply-side applications that focus only on upstream dependencies via imported intermediates (such as Baldwin & Freeman, 2020a). For example, let us assume that Ethiopia exports cotton to China where the cotton is processed and then exported to Europe as a textiles product. Ethiopian agriculture does not require any Chinese inputs, but its production is nonetheless dependent on Chinese producers located further downstream. By decomposing the value chain by its final product, we trace all participants in that chain, independent of their relative position. However – as in the demand-side analysis – we also assume that upstream (and downstream) suppliers do not redirect their output into other value chains. To capture the effects of the supply shock, we use data on the drop in percentage of industrial production in the three hubs Europe, North America, and China in the month of the largest drop during the first months of 2020.6 The months with the largest drop were February for China, and April for Europe and North America. China was the first country globally to be affected by COVID-19, and the first to implement a drastic lockdown, with other countries around the globe following in staggered sequence as the disease emerged and spread, with the resulting effects on industrial production. For both the demandand supply-side shocks, our estimations of the effects refer to the time that the pandemic – and the measures to contain it – restrained the economies most. All three hubs initially recovered quite substantially after one to two months with regard to both their final demand and their industrial production. All estimates thus refer to this time frame and can be read as projections of what were to happen if the economic effects of the lockdown due to the pandemic prevailed or returned during a future wave. To implement the method described above to be applied to estimate both the demand and the supply shock, we need information on the global system of input-output relationships (depicted in A); information on value-added to gross output ratios (V); and a vector of final demand (F). In particular, to obtain A, one needs to turn to global input-output tables, which describe the supply and use relationships between producers within and across countries. Global input-output tables are constructed combining a large amount of information on value-added, gross output, trade flows (intermediates, final goods), and final demand categories. As this is a highly data-intensive exercise, a major bottleneck to studying the involvement of developing countries is the relatively poor coverage of less developed countries, in particular in sub-Saharan Africa. Some attempts have been made to bridge this gap. The construction of the EORA Global Value Chain database (Lenzen, Moran, Kanemoto, & Geschke, 2013) has taken a global approach covering a large amount of countries since the 1990s but naturally has to make a compromise with respect to a clear 5 In further analysis, we also used a threshold of 10 percent. The results were naturally considerably smaller in magnitude but the cross-country pattern was qualitatively similar (results available upon request). 6 For Europe, we used data from the European Union, for North America from the United States. The data came from the respective national statistical bureaus. Stefan Pahl et al. 10 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) anchoring in official statistics and simplifying assumptions. As a country-specific alternative, we use the data compiled by Pahl et al. (2019), which extend the original World Input-Output Database (WIOD) (Dietzenbacher, Los, Stehrer, Timmer, & De Vries, 2013; Timmer et al., 2015). The WIOD covers the EU 27 plus Norway, Switzerland and the United Kingdom; United States and Canada; China, India, Indonesia, Japan, Taiwan and South Korea; Brazil and Mexico; Russia, Turkey and Australia; and an aggregate for the rest of the world. The construction in Pahl et al. (2019) closely follows the approach laid out in the construction of the WIOD but adds seven new developing countries for the period 2000 to 2014 to the existing WIOD. The new countries are: Ethiopia, Kenya, Senegal, South Africa, Bangladesh, Malaysia and Vietnam. This country-specific (rather than a global) approach allows for a number of improvements, which are particularly important when studying the value-added or income effects related to GVCs. As is easy to see from equation (1), value-added to gross output ratios in V are crucial to obtain a country’s value-added in global production. An advantage of using data from Pahl et al. (2019) is the yearly variation in the input data between 2000 and 2014 in those ratios for each of the sectors and industries covered, as opposed to keeping these ratios constant with, say, 2000 values in the year 2014. Secondly, the construction in Pahl et al. (2019) provides a careful treatment of trade flows (for example, re-exports; missing trade flows; classification by use category), which is paramount to depicting the cross-country relationships in A. Moreover, to obtain the domestic supply and use relations in A, the data are built up from national supply and use tables or official input-output tables, and as such are highly country-specific. Lastly, F is consistent with national accounts, and split between household consumption, government consumption, gross fixed capital formation and inventories. This allows for the clean identification of the reductions in final demand that stem from household consumption (as depicted in the retail survey of Eurostat, 2020, the household survey by Coibion et al., 2020, and the data from NBS, 2020b). Using this dataset, we choose all 12 low and middle-income countries in the three major developing world regions: 4 countries in Sub-Saharan Africa (Ethiopia, Kenya, Senegal, South Africa); 6 in East and Southeast Asia (Bangladesh, China, India, Indonesia, Malaysia, Vietnam); and 2 in Latin America (Brazil, Mexico). The choice of countries in Pahl et al. (2019) was guided by obtaining a first overview of developing countries from the major world regions, while meeting the high data demands necessary for the data construction. We base the estimates on the final year in that dataset, that is, 2014. 4 Demand-side vulnerability To study demand-side-related GDP effects for developing countries arising from the pandemic, we use the value-added trade data to compute how much of value-added in each of the developing countries in the sample depends on final demand in the various different regions in the world. Table 1 presents these results aggregated across sectors, where rows show individual developing countries. The first six columns list separate world regions, while the last column lists developing countries’ home markets. The values then depict how much of value-added in each country depends on final demand in each of these regions.7 7 These numbers are related to the export-to-GDP ratio, but not equal to it. Differences arise in different shares of domestic value-added to gross exports across the countries. The contagious effects of COVID-19 on developing countries through global value chains German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 11 Table 1: Demand-side dependency by region (as a percentage of GDP) Europe North America China East Asia Other emerging countries Rest of world Home market Bangladesh 6.1 2.7 0.2 0.4 0.7 2.6 87.2 China 3.2 3.6 - 2.1 2.1 7.7 81.3 India 2.1 1.7 0.9 0.5 0.8 7.3 86.7 Indonesia 2.3 2.6 2.4 3.3 1.7 7.0 80.7 Malaysia 4.7 4.8 5.3 6.5 4.9 24.1 49.7 Vietnam 8.1 9.4 5.3 6.0 3.6 15.5 52.1 Ethiopia 2.5 0.5 0.9 0.5 0.5 8.6 86.5 Kenya 3.0 0.7 0.2 0.2 0.5 14.8 80.6 Senegal 2.3 0.3 0.3 0.4 0.4 12.5 83.9 South Africa 4.7 2.5 2.6 1.5 2.0 13.9 72.8 Brazil 1.7 1.4 1.5 0.7 0.9 3.6 90.3 Mexico 1.4 13.8 0.6 0.5 0.6 2.1 81.0 Notes: Figures for 2014. GDP as the sum of value-added in 2014 USD. “Europe” refers to all 28 member countries of the European Union as of 2014 plus Switzerland; “North America” refers to the United States and Canada. Each country’s home market is included in the home market region such that columns add up to 100, except for rounding. Source: Authors’ calculations, based on method and data from Pahl et al. (2019) As shown in Table 1, Vietnam and Malaysia are most strongly dependent on foreign demand, with only around 50 per cent of domestic value-added dependent on final demand in their home markets. For other countries, GDP dependence on foreign demand ranges between 27 (South Africa) to below 10 per cent (Brazil). At the same time, for example, Bangladesh’s GDP is relatively strongly dependent on demand from Europe, at 6.1 per cent; and Mexico, unsurprisingly, on demand from North America, at 13.8 per cent. Hence, we would expect these countries to be most strongly affected by the economic downturn and plummeting demand in Europe and the United States. Considering regional differences, value-added in Asian countries is on average more dependent on foreign final demand than that in African countries.8 However, economic lockdown measures do not lead to a homogeneous decrease in demand across all sectors. Social distancing measures affect the sectors much more strongly that require direct personal interaction, besides the differentiated demand reductions due to an overall plunge in income.9 Developing countries participating in GVCs for which final demand collapsed comparatively more are therefore likely to be more vulnerable to COVID19-induced demand shocks through global production links. To reveal these sectoral dependencies, we show how much of the value-added in each developing country that 8 For Latin America, the sample is quite small and particular, with Brazil as a large country with a large home market and Mexico, with a strong dependence on US final demand, which are not necessarily representative of other countries in the region, but which, on the other hand, make for interesting contrasts in this respect. 9 The declining demand may also be due to people’s concerns rather than political measures, see Chetty, Friedman, Hendren, & Stepner (2020) and Goolsbee & Syverson (2020). We do not differentiate between various different drivers of demand downturns but viewed them jointly as a result of the pandemic. Stefan Pahl et al. 12 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) depends on final demand in the foreign region (as shown in Table 1) arises from demand in individual sectors. Table 2 presents corresponding findings, where the values depicted represent shares in per cent of total value-added in a developing country that depends on final demand in the respective foreign region, stemming from final demand in a given sector. While Table 2 is not exhaustive, it shows the most important end markets by sector grouping for the sample set of countries. The dependencies displayed reveal quite stark differences between developing countries and regions in terms of how much of domestic production for foreign demand is concentrated in production for specific sectors.10 For Bangladesh, 94.1 per cent of its value-added embedded in European final demand is for textile goods, and this value is 94.9 per cent of its production for North American final demand. Other countries, such as Vietnam, are much more diversified: for all of Vietnamese production consumed in Europe, only 24.2 per cent are for textiles whereas 26.4 per cent are for the electronics sector and 14.9 per cent go into the consumption of food. If the consumption of textiles breaks down in Europe more than in other manufacturing sectors – as has now happened during the COVID-19 pandemic (ILO [International Labour Organization], 2020) – Bangladesh is thus likely to be relatively more affected by this than a more diversified country such as Vietnam. 10 Note that the value-added in the source country may be in other sectors, as long as they supply the respective sector of final demand. Conversely, production in a certain sector in a source country need not be for final demand in that sector in another country (or domestically). Table 2: Demand-side dependency by product group (as a percentage of GDP generated in respective region) Asia Sub-Saharan Africa Latin America Bangladesh China India Indonesia Malaysia Vietnam Ethiopia Kenya Senegal South Africa Mexico Brazil Europe Agriculture, forestry and fishing A 0.0 0.7 2.7 2.4 1.7 4.7 48.5 66.0 40.9 8.5 2.8 4.1 Food, beverages, tobacco C10t12 3.1 4.0 7.4 14.3 8.8 14.9 27.0 20.4 18.9 9.6 8.3 25.6 Textiles C13t15 94.1 13.5 20.1 13.2 4.1 24.2 3.1 1.5 2.5 1.9 1.8 3.3 Coke and refined petroleum C19 0.0 1.1 2.1 3.5 2.7 0.7 0.2 0.5 2.3 4.6 8.6 3.4 Pharmaceuticals C21 0.1 1.3 1.1 1.5 0.9 0.4 0.4 0.2 0.8 0.8 3.4 1.4 Computer, and electronics C26 0.1 15.9 1.9 7.1 18.6 26.4 0.5 0.4 1.5 3.0 5.2 1.6 Electrical equipment C27 0.0 6.0 1.3 2.1 3.4 1.5 0.5 0.1 1.9 1.8 1.6 1.3 Machinery C28 0.1 6.0 2.8 2.1 4.7 1.4 0.8 0.3 3.3 10.0 4.1 2.7 Furniture; other manufacturing C31t33 0.1 6.4 3.2 5.8 6.1 6.6 0.9 0.8 3.0 3.7 4.9 3.4 Sum 97.7 54.9 42.5 52.0 50.9 80.7 82.0 90.3 75.0 43.8 40.8 46.8 North America Agriculture, forestry and fishing A 0.1 0.3 2.0 0.8 0.5 5.1 7.7 11.6 2.7 2.2 4.9 1.8 Food, beverages, tobacco C10t12 1.3 3.2 7.4 13.0 6.0 11.6 33.2 8.7 8.6 5.6 7.0 11.0 Textiles C13t15 94.9 16.1 22.5 26.5 5.2 37.8 10.8 57.4 9.3 2.7 3.4 4.5 Coke and refined petroleum C19 0.0 0.8 2.3 2.2 1.8 1.0 1.2 0.8 3.4 5.0 4.8 4.7 Pharmaceuticals C21 0.0 0.8 1.5 0.6 0.7 0.2 1.3 0.6 1.6 1.1 0.4 1.5 Computer, and electronics C26 0.2 18.5 2.1 6.0 20.6 11.4 1.6 1.2 4.5 3.9 6.3 2.3 Electrical equipment C27 0.0 5.4 1.1 1.4 3.6 1.1 0.6 0.3 1.7 1.8 3.5 1.9 Machinery C28 0.1 6.0 3.0 2.0 3.8 1.5 1.2 0.7 3.8 7.6 5.8 5.3 Furniture; other manufacturing C31t33 0.6 6.9 6.0 7.1 7.0 10.7 3.4 2.0 21.7 4.3 4.4 5.3 Sum 97.3 58.1 48.0 59.7 49.2 80.3 61.1 83.2 57.4 34.2 40.3 38.3 Table 2 (cont.): Demand-side dependency by product group (as percentage of GDP generated in respective region) China Agriculture, forestry and fishing A 8.6 2.4 3.0 2.5 8.7 6.9 6.3 3.9 2.4 2.0 4.6 Food, beverages, tobacco C10t12 3.7 6.5 11.7 7.2 20.8 19.8 10.0 17.9 4.3 4.8 19.4 Textiles C13t15 48.6 4.6 3.6 1.7 6.5 4.1 4.4 3.0 1.5 1.5 3.3 Coke and refined petroleum C19 0.3 0.7 2.0 1.4 1.0 0.2 0.8 0.6 1.6 1.2 0.9 Pharmaceuticals C21 0.4 0.5 0.5 0.4 0.2 0.8 0.8 0.7 0.4 0.7 1.0 Computer, and electronics C26 1.7 3.3 3.7 13.2 9.9 0.9 5.2 5.1 3.2 5.3 1.5 Electrical equipment C27 1.2 2.5 3.5 4.0 2.9 1.3 3.3 3.4 3.6 4.2 2.1 Machinery C28 2.6 5.1 5.0 6.8 4.0 2.7 5.5 7.1 6.9 7.3 4.1 Furniture; other manufacturing C31t33 1.3 1.0 1.2 0.9 1.3 0.6 1.4 1.7 0.6 3.9 0.6 Sum 68.2 26.7 34.2 38.1 55.4 37.3 37.8 43.3 24.5 30.9 37.6 Notes: Figures for 2014. Shaded cells indicate 10 per cent or more in respective region. List of industries is not exhaustive but only shows industries for which we obtained demand shock figures (see Table 3). “Europe” refers to all 27 member countries of the European Union plus Switzerland and the United Kingdom; “North America” refers to the United States and Canada. Source: Authors’ calculations, based on method and data in Pahl et al. (2019) The contagious effects of COVID-19 on developing countries through global value chains German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 15 In order to provide a ballpark figure for the adverse effects inflicted on sample developing countries by the demand slumps in Europe, the United States, and China due to the COVID19 pandemic, we combine the above results with data on how much final demand fell by sector. Table 3 shows the collapse in demand by ISIC sector and region, using the sectoral mapping provided by Appendix table A1. The numbers shown in Table 3 indicate sizable heterogeneity across regions in terms of sectoral demand slumps. Textiles demand, for example, dropped by about 78 per cent in Europe but only by about 36 per cent in North America. Countries highly dependent on textiles consumed in Europe are thus likely to experience a relatively more severe shock in GDP than those dependent on textiles consumption in North America. What does this imply for those developing countries located upstream in GVCs? To get a first impression about the dimensions of what the demand slumps could imply, we assume that each sectoral downturn was passed proportionally through the value chain, thus affecting value-added in the supplying countries to the same extent.11 This produces an approximation to the loss of value-added in each developing country through its contribution to the respective final demand sectors. Table 4 shows these results. 11 This assumes that the demand shock is uniform across varieties of final goods within sectoral aggregation (for example, final good varieties from different countries), and that the production function remains unchanged (that is, cost shares remain constant); see also Pahl et al. (2019). Table 3: Sectoral downturns in final demand (in per cent) Sectors ISIC4 Europe North America China Agriculture, forestry and fishing A -1.4 -14.7 -11.3 Food, beverages, tobacco C10t12 -1.4 -14.7 -11.3 Textiles C13t15 -77.8 -35.7 -48.9 Coke and refined petroleum C19 -43.2 -28.2 -37.5 Pharmaceuticals C21 -12.4 -20.5 -33.0 Computer, and electronics C26 -41.5 -9.5 -28.6 Electrical equipment C27 -34.8 -9.5 -28.6 Machinery C28 -34.8 -9.5 -40.4 Furniture; other manufacturing C31t33 -34.8 -22.0 -66.5 Notes: “Europe” refers to all 27 member countries of the European Union plus Switzerland and the United Kingdom; “North America” refers to the United States and Canada. Source: Coibion et al. (2020), Eurostat (2020), NBS (2020b) Stefan Pahl et al. 16 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) As Table 4 illustrates, the likely contribution of demand downturns further downstream in GVCs to an overall decrease in GDP differs significantly across the countries in our sample. The countries that we found to be dependent more on foreign markets – and among those the ones specialised in sectors with forecasted sharpest demand decreases – were expected to suffer from comparatively stronger drops in own GDP through this channel. For example, our estimations suggest that Bangladesh’s GDP is experiencing a drop of about 4.5 per cent due only to falling demand in Europe and 0.9 per cent in North America. This effect is mainly because of the sharp decline in demand for textiles. Overall, Vietnam is even more dependent on foreign demand, although in different sectors, and might expect a decline of about 1.8 per cent through declining final demand in Europe and North America, respectively. Despite its comparable reliance on production in value chains for Chinese final demand rather than that in Europe or North America, it is not so much affected by declining demand in China. The reason is that Vietnam produces comparatively even more in value chains for China where final demand has not plummeted so much (food and agriculture) as in those in which it still produces significant shares for European and North American final demand (textiles, and computer and electronics). By contrast, countries in sub-Saharan Africa are much less integrated into the world economy and therefore only experience minor economic effects through GVCs. 5 Supply-side vulnerability With the role for some of the poorest developing countries in GVCs remains restricted to the supply of commodities to be processed abroad, others have managed to become important pillars further downstream in value chains – for example, in the assembly of final goods (World Bank, 2020). As such, the maintenance of output capacities in developing countries for both commodity exporters as well as downstream assemblers often relies on Table 4: Demand-induced value-added effect (as a percentage of GDP) Europe North America China Bangladesh -4.46 -0.93 -0.06 China -0.52 -0.31 - India -0.37 -0.20 -0.04 Indonesia -0.36 -0.36 -0.13 Malaysia -0.38 -0.25 -0.21 Vietnam -1.78 -1.75 -0.45 Ethiopia -0.10 -0.05 -0.05 Kenya -0.09 -0.17 -0.01 Senegal -0.11 -0.03 -0.02 South Africa -0.25 -0.12 -0.08 Brazil -0.09 -0.08 -0.08 Mexico -0.10 -0.73 -0.02 Notes: Figures for 2014. “GDP” is the sum of value-added in 2014 USD. “Europe” refers to all 27 member countries of the European Union plus Switzerland and the United Kingdom; “North America” refers to the United States and Canada. Source: Authors’ calculations, based on Table 2 (including all sectors) and demand-side estimates as shown in Table 3 The contagious effects of COVID-19 on developing countries through global value chains German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 17 intermediate inputs from foreign sources. We would like to emphasise that this dependency can consequently be critical for production in GVCs. Analogously to Table 1 and for an aggregation across sectors, Table 5 presents how much of value-added in each developing country in our sample depends on critical inputs from the three hubs. Values indicate the share of value-added in the row country (as per cent of GDP) generated in value chains with a minimum supply-side contribution of 5 per cent at any stage in the production process according to the column regions. Defining a threshold contribution implies that substituting existing supply-side relations appears fairly difficult (at least in the short run) and, with this, unlikely. This emphasises the displayed dependencies. Table 5 reveals that the dependency of developing countries is far from being homogeneous with respect to supplying countries: indeed, geographical proximity appears to be a salient determinant (see also Baldwin & Lopez-Gonzalez, 2015; Johnson & Noguera, 2017). For example, while Mexico generates nearly 50 per cent of domestic value-added through value chains which depend on inputs from the United States and/or Canada, Indonesia (8.7 per cent), Malaysia (11.1 per cent), and Vietnam (7.5 per cent) exhibit natural dependencies on China. At the same time, however, all three are not exclusively tied to Chinese inputs. Instead, their value-added appears to depend equally on European inputs, suggesting a welldiversified portfolio of suppliers. In view of the uneven temporal distribution of production bottlenecks across major GVC-hubs induced by COVID-19 that is still affecting Europe and North America while Chinese production capacities started to be ramped up again in the late spring of 2020, Southeast Asian countries do not seem to have to bear the full impact of supply shortages at the same time. Instead, it appears that their supply-side diversification at least partly contributes to a cushioning of the adverse spillover effects originating in major hubs. What is more, depending on both its duration and extent, Southeast Asian countries not only benefit disproportionally from Chinese economic recovery; at the same time, their supply-side diversification potentially allows them to partly circumvent adverse effects originating in Europe if Chinese supply growth outbalances declines in Europe. Table 5: Supply-side dependency (as a percentage of GDP) Europe North America China Bangladesh 0.6 0.1 0.4 China 5.4 1.9 93.2 India 7.1 1.1 2.6 Indonesia 7.7 1.5 8.7 Malaysia 13.4 3.8 11.1 Vietnam 8.4 2.8 7.5 Ethiopia 2.6 0.4 1.4 Kenya 5.8 0.3 4.5 Senegal 3.3 0.2 1.8 South Africa 8.5 2.2 5.8 Brazil 5.4 1.7 3.4 Mexico 7.6 2.5 Notes: Shares indicate value-added in row country generated in value chains with contributions of 5 per cent or more of column region. “Europe” refers to all 28 member countries of the European Union as of 2014 plus Switzerland; “North America” refers to the United States and Canada. Row countries are included in the respective column. Source: Authors’ calculations, based on Pahl et al. (2019) and method as described in the main text Stefan Pahl et al. 18 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) Bangladesh constitutes an exceptional case in Asia as the country shows extraordinarily little dependence on inputs from any of the major hub regions. In sharp contrast to the country’s deep integration into GVCs through significant demand-side dependencies especially vis-à-vis the European textile industries (see Table 1), Bangladesh’s limited supply-side dependence makes it a natural candidate for low vulnerability in terms of foreign supply-side shocks arising from COVID-19-induced bottlenecks. In other words, without ignoring Bangladesh’s high degree of vulnerability due to its obvious concentration on both sector (textiles) and downstream destination (Europe), and the country’s domestic supply difficulties due to high future infection rates and economic lockdowns, the country is exposed to surprisingly little vulnerability due to COVID-19-induced supply constraints in the three major global economic hubs.12 Similarly, pronounced “shock-resilience” towards adverse supply-side spillovers from global GVC-hubs can be found for three sub-Saharan countries, namely Kenya, Senegal, and to a lesser extent, Ethiopia, where in all cases the most important value chains seem to depend primarily on domestic sources. The above-mentioned sectoral assessment already provides a partial explanation for this finding as all three reveal a sharp concentration in value-added generation in “agriculture, forestry and fishery” for example, which is characterised by limited international production fragmentation (see, for example, Johnson & Noguera, 2017). Underpinning our a priori findings, we use data from the national statistical bureaus of the European Union, the United States, and China. The peaks in industrial production declines since the outbreak of the pandemic amount to 27 per cent (Europe in April), 16.6 per cent (United States in April) and 28.7 per cent (China in February). These numbers are shown in Table 6. Assuming, for simplicity, that sectoral export activities were hit proportionally, this implies that the same share of value-added that uses more than 5 per cent of Chinese inputs as overall intermediates cannot produce anymore for this time span. As the above numbers suggest, the stricter lockdowns in Europe and China resulted in a larger production drop than in North America. Thus, being integrated into GVCs dependent on those hubs has a more severe effect on the developing countries’ own production. A back-of-the-envelope calculation can give a rough estimate of what effect this will have on GDPs in our sample developing countries, given the results presented in Table 5. Table 7 shows the results of this exercise. Resulting from its enormous supply-side dependency on the United States, 11 per cent of Mexico’s GDP are at risk. While Bangladesh’s GDP may drop by a slight 0.3 per cent following adverse supply-shocks originating across the three major hubs, the GDPs of Indonesia (-4.8 per cent), Malaysia (-7.4 per cent), Vietnam (-5.0 per cent) are threatened 12 Bangladesh does rely on imports from countries other than the hubs, particularly India. While initially, agricultural production in India did not plummet as much as other Indian sectors, the overall effects of past and potential future lockdown measures in India for developing countries is beyond the scope of this paper. This would be an important extension of our work. Table 6: Decline in industrial production (in per cent) Europe North America China Decline in industrial production -27 -16.6 -28.7 Note: Figures refer to April in Europe and North America, and February in China. Source: Eurostat (2020) for Europe, FRED [Federal Reserve Economic Data] (2020) for North America, NBS (2020a) for China The contagious effects of COVID-19 on developing countries through global value chains German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 25 ILO (International Labour Organization). (2020). COVID-19 and the textiles, clothing, leather and footwear industries. ILO Sectorial Brief. Retrieved from https://www.ilo.org/wcmsp5/groups/public/---ed_dialogue/--- sector/documents/briefingnote/wcms_741344.pdf Javorcik, B. (2020). Global supply chains will not be the same in the post-COVID-19 world. In R. Baldwin & S. Evenett (Eds.), Covid-19 and trade policy: Why turning inward won’t work (pp. 111-116). London: CEPR Press. Johnson, R. C., & Noguera, G. (2017). A portrait of trade in value-added over four decades. Review of Economics and Statistics, 99(5), 896-911. Kilic, K., & Marin, D. (mimeo). A new era of world trade: Global value chains and robots. Munich: Technische Universität München, TUM School of Management. Kummritz, V., Taglioni, D., & Winkler, D. (2017). Economic upgrading through global value chain participation: Which policies increase the value added gains? (World Bank Policy Research Working Paper 8007). Washington, DC: World Bank. Lenzen, M., Moran, D., Kanemoto, K., & Geschke, A. (2013). Building Eora: A global multi-region input– output database at high country and sector resolution. Economic Systems Research, 25(1), 20-49. Levchenko, A. A., Lewis, L. T., & Tesar, L. L. (2010). The collapse of international trade during the 2008– 09 crisis: In search of the smoking gun. IMF Economic Review, 58(2), 214-253. Los, B., Timmer, M. P., & de Vries, G. J. (2015). How global are global value chains? A new approach to measure international fragmentation. Journal of Regional Science, 55(1), 66-92. Meyn, M., & Kennan, J. (2009). The implications of the global financial crisis for developing countries' export volumes and values (Working Paper 305). London: Overseas Development Institute (ODI). Miller, R. E., & Blair, P. D. (2009). Input-output analysis: Foundations and extensions. Cambridge: Cambridge University Press. Miroudot, S. (2020). Resilience versus robustness in global value chains: Some policy implications. In R. Baldwin & S. Evenett (Eds.), Covid-19 and trade policy: Why turning inward won’t work (pp. 117-130). London: CEPR Press. Nagengast, A. J., & Stehrer, R. (2016). The great collapse in value added trade. Review of International Economics, 24(2), 392-421. NBS (National Statistical Bureau of China). (2020a). Industrial production operation in the first two months of 2020. Retrieved from http://www.stats.gov.cn/english/PressRelease/202003/t20200317_1732640.html NBS. (2020b). Monthly data, domestic trade, retail sales of enterprises above designated size. Retrieved from https://data.stats.gov.cn/english/easyquery.htm?cn=A01 OECD (Organisation for Economic Cooperation and Development). (2020). Trade interdependencies in Covid-19 goods. Retrieved from https://read.oecd-ilibrary.org/view/?ref=132_132706-m5stc83l59&title=Policy-Respone-TradeInterdependencies-in-Covid19-Goods. Pahl, S., & Timmer, M. P. (2020). Do global value chains enhance economic upgrading? A long view. The Journal of Development Studies, 56(9), 1683-1705. Pahl, S., Timmer, M. P., Gouma, R., & Woltjer, P. J. (2019). Jobs in global value chains: New evidence for four African countries in international perspective (World Bank Policy Research Working Paper 8953). Washington, DC: World Bank. Park, Y., Hong, P., & Roh, J. J. (2013). Supply chain lessons from the catastrophic natural disaster in Japan. Business Horizons, 56(1), 75-85. Pentecôte, J. S., & Rondeau, F. (2015). Trade spillovers on output growth during the 2008 financial crisis. International Economics, 143, 36-47. Stefan Pahl et al. 26 German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) Rose, A. K., & Spiegel, M. M. (2010). Cross‐country causes and consequences of the 2008 crisis: International linkages and American exposure. Pacific Economic Review, 15(3), 340-363. Seric, A., Gorg, H., Mosle, S., & Windisch, M. (20 April 2020). Managing COVID-19: How the pandemic disrupts global value chains. Retrieved from https://iap.unido.org/articles/managing-covid-19-howpandemic-disrupts-global-value-chains Seric, A., & Winkler, D. (2020). COVID-19 could spur automation and reverse globalisation – to some extent. VoxEU.org column. Retrieved from https://voxeu.org/article/covid-19-could-spur-automation-and-reverse-globalisation-some-extent Timmer, M. P., Dietzenbacher, E., Los, B., Stehrer, R., & De Vries, G. J. (2015). An illustrated user guide to the world input–output database: The case of global automotive production. Review of International Economics, 23(3), 575-605. Timmer, M. P., Los, B., Stehrer, R., & De Vries, G. J. (2013). Fragmentation, incomes and jobs: An analysis of European competitiveness. Economic Policy, 28(76), 613-661. UNIDO (United Nations Industrial Development Organization). (2020). World manufacturing production: Statistics for quarter I, 2020. Vienna: Author. Retrieved from https://www.unido.org/sites/default/files/files/202006/World_manufacturing_production_2020_Q1.pdf World Bank. (2020). World Development Report 2020: Trading for development in the age of global value chains. Washington, DC: World Bank. WTO (World Trade Organization). (2019). World trade statistical review 2019. Geneva: Author. Retrieved from https://www.wto.org/english/res_e/statis_e/wts2019_e/wts19_toc_e.htm Yamamoto, S. (2014). Transmission of US financial and trade shocks to Asian economies: Implications for spillover of the 2007–2009 US financial crisis. The North American Journal of Economics and Finance, 27, 88-103. The contagious effects of COVID-19 on developing countries through global value chains German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) 27 Appendix Table A1: Sectoral mapping ISIC Rev 4 Classification Europe: sectoral classification in Eurostat (2020) North America: sectoral classification in Coibion et al. (2020) China: sectoral classification in NBS (2020b) Agriculture, forestry and fishing Food, drinks, tobacco Food Grain and oil, foodstuffs, beverages, tobacco Food, beverages, tobacco Food, drinks, tobacco Food Grain and oil, foodstuffs, bseverage, tobacco Textiles Textiles, clothes, footwear Clothing, footwear, personal care Garments, footwear, hats, knitwear Coke and refined petroleum Automotive fuel Gasoline Petroleum and related products Pharmaceuticals Pharmaceutical and medical goods Medical Traditional Chinese and Western medicine Computer, and electronics Computer equipment, books Durable goods Communication appliances Electrical equipment Electrical goods and furniture Durable goods Communication appliances Machinery Electrical goods and furniture Durable goods Cultural and office appliances Furniture; other manufacturing Electrical goods and furniture Furniture, jewellery, small appliances and other small durable goods Furniture Notes: Column 1 lists the sectors of final demand used in this study. Columns 2-4 list the sectors from the classifications in Eurostat (2020), Coibion et al. (2020) and NBS (2020b) that we map with the respective ISIC Rev. 4 sector. Source: Authors Table A2: Shares of value-added generated by final demand in three hubs Shares Bangladesh Brazil China Ethiopia Indonesia India Kenya Mexico Malaysia Senegal Vietnam South Africa Agriculture A 0.3 5.4 0.8 42.8 2.9 3.4 57.8 8.1 3.1 43.9 8.6 10.6 Mining B 0.0 0.3 0.4 0.1 0.1 0.1 0.0 1.0 0.1 0.1 0.1 0.8 Food C10t12 2.7 32.2 5.9 33.1 20.5 10.8 19.2 12.4 15.5 23.2 22.4 14.7 Textiles C13t15 94.9 6.1 24.8 5.5 23.3 26.8 13.0 5.6 7.6 4.1 38.6 4.1 Wood C16 0.0 0.1 0.4 0.5 0.3 0.5 0.1 0.1 0.2 0.1 0.3 0.1 Paper C17 0.0 0.5 0.5 0.2 0.4 0.3 0.1 0.6 0.3 0.1 0.1 0.4 Printing C18 0.0 0.1 0.1 0.0 0.1 0.1 0.0 0.1 0.1 0.0 0.0 0.1 Coke and refined petroleum C19 0.0 4.9 1.5 0.4 4.0 2.8 0.6 8.8 4.1 2.9 1.3 8.0 Chemicals C20 0.0 2.3 1.9 0.5 1.9 2.5 0.3 2.7 2.6 0.8 0.5 2.8 Pharmaceuticals C21 0.1 1.6 1.2 0.6 0.9 1.3 0.2 0.9 1.0 0.6 0.3 1.1 Rubber C22 0.1 0.5 1.8 0.2 1.4 1.1 0.1 1.2 4.4 0.3 0.9 0.7 Non-metallic mineral products C23 0.2 0.2 0.8 0.1 0.4 0.2 0.1 0.2 0.3 0.1 0.3 0.3 Basic metals C24 0.0 0.2 0.9 0.0 0.1 0.2 0.0 0.1 0.2 0.1 0.1 0.4 Fabricated metal products C25 0.0 0.6 1.6 0.1 0.4 0.5 0.1 0.7 0.4 0.5 0.3 1.0 Computer C26 0.0 1.0 8.7 0.3 3.5 1.1 0.3 5.6 8.0 0.9 2.2 2.0 Electrical equipment C27 0.0 0.9 5.1 0.2 1.3 0.7 0.1 2.4 2.8 0.5 0.6 1.6 Machinery C28 0.0 0.3 1.0 0.1 0.2 0.3 0.1 0.5 0.4 0.3 0.3 0.7 Automotives C29 0.1 3.1 4.6 0.7 3.5 3.9 0.4 10.7 4.6 1.3 2.1 12.1 Transport equipment C30 0.5 0.5 1.1 0.1 0.6 0.6 0.1 1.2 0.9 0.4 0.5 1.1 Furniture; other manufacturing C31t33 0.2 1.7 9.1 1.0 6.2 4.5 0.9 3.4 6.2 4.9 9.3 4.4 Electricity DtE 0.0 2.2 1.3 0.4 1.6 1.3 0.3 2.4 2.2 1.8 0.7 4.2 Construction F 0.0 0.3 0.3 0.1 0.2 0.3 0.1 0.1 0.3 0.4 0.1 0.4 Wholesale & retail, food & restaurants GnI 0.3 11.4 7.1 6.8 8.8 7.4 2.8 10.3 9.7 4.7 3.1 8.2 Transportation & storage; telecomm. HnJ61 0.1 4.1 3.8 0.8 3.3 3.1 0.5 4.7 4.4 1.6 1.2 4.3 Business services JtNexJ61 0.2 9.5 8.2 2.0 6.1 9.3 1.2 5.6 9.7 2.9 2.5 7.1 Public services OtQ 0.2 6.7 4.8 2.4 5.2 5.5 1.0 8.4 7.5 2.5 2.5 6.4 Other services RtT 0.1 3.1 2.3 1.0 2.4 11.5 0.5 2.3 3.2 0.9 1.0 2.4 Notes: Figures for 2014. Shares are all value-added generated in respective country by delivering to final demand in three hubs by consumption by sector aggregate. Source: Authors’ calculations, based on data in Pahl et al. (2019) Publications of the German Development Institute / Deutsches Institut für Entwicklungspolitik (DIE) Discussion Papers 20/2020 von Haldenwang, Christian. Digitalising the fiscal contract: An interdisciplinary framework for empirical inquiry (37 pp.). ISBN 978-3-96021-132-7. DOI:10.23661/dp20.2020. 19/2020 Fischer, Roger A. How the G7 reviews its work on development: A case study of internal accountability (47 pp.). ISBN 978-3-96021-130-3. DOI: 10.23661/dp19.2020. 18/2020 Römling, Cornelia, & Sarah Holzapfel. Monitoring in German bilateral development cooperation: A case study of agricultural, rural development and food security projects (49 pp.). ISBN 978-3-96021-129-7. DOI:10.23661/dp18.2020. 17/2020 Laudage, Sabine. Corporate tax revenue and foreign direct investment: potential tradeoffs and how to address them (46 pp.). ISBN 978-3-96021-128-0. DOI:10.23661/dp17.2020. 16/2020 Cavatorta, Francesco, & Fabio Merone. Never-ending reformism from above and dissatisfaction from below: The paradox of Moroccan post-Spring politics (32 pp.). ISBN 978-3-96021-127-3. DOI:10.23661/dp16.2020. 15/2020 Eppler, Mirko, Stella Gaetani, Francy Köllner, Jana Kuhnt, Charles Martin-Shields, Nyat Mebrahtu, …Carlotta Preiß. Information and communication technology in the lives of forcibly displaced persons in Kenya (31 pp.). ISBN 978-3-96021-126-6 DOI:10.23661/dp15.2020. 14/2020 Hackenesch, Christine, Julia Leininger, & Karina Mross. What the EU should do for democracy support in Africa: Ten proposals for a new strategic initiative in times of polarisation (45 pp.). ISBN 978-3-96021-125-9. DOI:10.23661/dp14.2020. 13/2020 Never, Babette, Jose Ramon Albert, Hanna Fuhrmann, Sebastian Gsell, Miguel Jaramillo, Sascha Kuhn, & Bernardin Senadza. Carbon consumption patterns of emerging middle classes (42 pp.). ISBN 978-3-96021-124-2. DOI:10.23661/dp13.2020. 12/2020 Wehrmann, Dorothea. Transnational cooperation in times of rapid global changes: The Arctic Council as a success case? (29 pp). ISBN 978-3-96021-123-5. DOI:10.23661/dp12.2020. 11/2020 Erforth, Benedikt. The future of European development banking: What role and place for the European Investment Bank? (36 pp.). ISBN 978-3-96021-122-8. DOI:10.23661/dp11.2020. 10/2020 Kaplan, Lennart. Systemic challenges and opportunities of Franco-German development cooperation (64 pp.). ISBN 978-3-96021-121-1.DOI:10.23661/dp10.2020. 9/2020 Stoffel, Tim. Socially responsible public procurement (SRPP) in multi-level regulatory frameworks: Assessment report on policy space for SRPP regulation and implementation in Germany and Kenya (55 pp.). ISBN 978-3-96021-120-4. DOI:10.23661/dp9.2020. [Price: EUR 6.00; publications may be ordered from the DIE or through bookshops.] For a complete list of DIE publications: www.die-gdi.de