Global Value Chains

by Ayelech T. Melese and Lindsay Whitfield

Industrialization is the main driver of higher per capita incomes and a rising standard of living in low-income countries. Industrialization may be catalyzed by foreign direct investment. However, it is sustained by national firms becoming internationally competitive in range of industries, which should increase in technological complexity as firms building their technological and organizational capabilities. 

In contemporary global production and trade systems, locally owned firms in low-income countries find themselves in an increasingly difficult situation. They have limited existing knowledge of the industry and thus face a large gap in the capabilities they have and those that are required to be internationally competitive. Over the past decades, the barriers to entry in even basic manufacturing and agriculture-based export industries have increased. These barriers to entry include stringent and fast changing requirements by Western buyers, strict proprietary rights, and stiff global competition from suppliers in other low and middle-income countries. Locally owned firms must build their capabilities but face big risks in doing so as they often have scarce financial resources, weak international networks, and operate in precarious business and political environments. 

In such a context, one would wonder what can motivate local firms to enter an export industry new to their country, invest in learning, and succeed in build their technological capabilities. Seeking to answer these questions, we examined the locally owned firms in Ethiopian floriculture export industry which emerged in the 1990s, following the coming of a new government into power and the subsequent policy shift from command to mixed economy. Local firms must master significant technological and organizational skills to run a cut-flower firm, which is more like manufacturing than agriculture.

Drawing extensively on the historical evidence, scholars of industrialization in East Asia emphasized that industrial policy has been the key driver of private investment and firm level learning. This literature significantly contributed in highlighting what the industrial policies should look like and how they should be customized to the twenty-first century’s context to achieve similar effect in low income countries in Africa and elsewhere. Notwithstanding the important contributions, the industrial policy literature has given limited attention to firm-level dynamics of learning and agency of firms in (re)allocating resources based on their own business strategies. 

Our examination of the Ethiopia’s floriculture export sector largely confirms the arguments of industrial policy scholars. The sector went through exponential growth during the period 2002 to 2008. This growth was driven by sector specific industrial policies that provided generous incentive packages and infrastructural support that spurred both local and foreign investment, but notably significantly lowered the entry barriers and risks for local investors as they sought to build the required capabilities. 

However, the sector did not keep the momentum of growth and technological upgrading for long. In the 2010s, as growth in the sector stagnated, the number of local firms declined significantly. Only fifteen local firms managed to survive and thrive, most which were part of family-owned diversified business groups. Our analysis shows that the local owners of cut-flower farms often sought to develop only the minimum capabilities needed to sustain their cut-flower exports. Government industrial policies which incentivized local entrepreneurs to invest in the sector did not entail performance standards that tied the financing and other support to export performance and thus firms had little incentive to continue to invest in learning. 

Scholars of global value chains (GVC) discussed the important role and potentials of GVC configurations to bolster learning and upgrading opportunities of local firms in low-income firms. Upgrading is often confined to a certain level as local supplier firms encounter various control mechanisms exercised by lead firms in the GVC. However, local firms can exercise their agency with a much broader scope that is often (implicitly) assumed. The case of Ethiopia’s cut-flower export industry shows that local firms can pursue growth paths within and across GVCs, as they are part of diversified business groups and firm owners make decisions not only about the growth of the cut-flower firm but also about the overall profitability of their business group. 

The mapping of the floriculture GVC below shows that flowers are traded in the Dutch auction, which remains the dominant channel, and through direct sales to wholesalers, supermarkets, and other retailers, which has become more important since the 1990s.  The specification of buyers in the direct sales channel depends on their end-markets and market segments. Consumers with special demands often buy flowers from specialized outlets such as florists and web-shops. 

By the time Ethiopia joined the global cut-flower industry, the abundant year-long supply was causing stiff competition and putting downward pressure on suppliers’ profit margins. Non-price competition such as reliability and consistency in terms of product quality, quantity and delivery was becoming more important. As a result, buyers’ requirements became more stringent to differentiate products, but also address growing consumer and non-governmental organizations’ concerns about labour and environmental issues on flower farms.

Ethiopia’s national firms chose various export trajectories combining different market channels and end-markets, while protecting or improving the overall profitability of their family-owned diversified business groups. These export trajectories dictated which capabilities they needed to build and to which level to keep their flower export firm internationally competitive and thus retain buyers.

In this context, owners’ decisions for their cut-flower firms and their growth paths were influenced by the limited resources of their diversified business groups and showed significant variations. Firm owners sometimes chose to invest less in building the capabilities of their flower export firms to release resources and managerial talents to other affiliate businesses or make investments in new domestic market industries. On the other hand, some family business groups transferred resources from their domestic market-focused firms to subsidize the costs of building capabilities in their flower export firms. 

The analysis also shows how the higher organizational and managerial demands of operating a firm in a competitive export industry like floriculture precipitated a move toward modern management practices in this first generation of family business groups in Ethiopia. These findings emphasize the need to consider local firms’ capabilities and position with family business groups when designing industrial policy.

Read more in our recently published article:  

Industrial policy, local firm growth paths, and capability building in low-income countries: lessons from Ethiopia’s floriculture export sector, Industrial and Corporate Change, advanced access.

by Lindsay Whitfield

The textile and apparel industry is a globalised industry characterised by a high degree of dependency and fragmentation in global supply chains. From the mid-20^th century, cotton textile and apparel manufacturing production relocated from industrialised to developing countries, particularly in Asia, and was quickly followed by the emergence of the petrochemical industry and synthetic fibre production in East Asian countries, while South Asia continued to focus on cotton and natural fibres. Changes in international trade over the past 30 years fostered the textile and apparel global value chain dynamic. These changes were marked by opening trade borders, lowering tariffs, tax breaks or subsidies offered by various governments, and free trade agreements. Low labour costs and lower production costs have shaped the decisions of global actors to relocate their operations from one country to another. 

Current global trends are set to transform the global value chains of textiles and apparel again. These trends include (1)increasing production costs in China, (2) the drive to shorten supply chains post-Covid pandemic through nearshoring and seeking countries with yarn-to-garment (or vertical) production capabilities, and (3) the sustainability shift. These trends are leading global apparel buyers to diversify their sourcing away from China and Asia more generally and to consider new supply bases. 

Global apparel buyers are no longer concerned only with cost, quality and speed to market when making sourcing decisions. They now must balance those criteria with sustainability goals. The climate crisis and the high contribution of the global fashion industry to greenhouse gas emissions are putting pressure on industry players to make radical changes in production processes. While consumers are increasingly demanding sustainability, the greatest pressure is coming from European country governments and the European Union Commission. Notably, existing and expected EU regulations have increased market demand for alternative sustainable fibres and textile production technologies.

As a result, new fibre and recycling technologies are in a phase of fast innovation to produce more sustainable and circular man-made fibres. The raw material for clothing production will become manufactured fibres that rely on advancements in chemical technologies and biofabrication. Virgin cotton will be of less importance, as it is replaced by natural fibres that are less resource intensive and by man-made cellulosic fibres that feel like cotton. In the transition stage, cotton will move to organic and regenerative as well as become blended with other natural and man-made cellulosic fibres. Future fibres will have a high technology content and require licensing technology, both man-made cellulosic fibres and the new wave of synthetics that seek to replace polymers.

Currently, African countries export very little of what is traded within apparel global supply chains and across the African continent, except in the case of cotton, but import a large amount of what the rest of the world produces. Africa’s fabric and apparel production is biased towards cotton, especially in yarn and fabric production, with little participation in production and export of man-made fiber, yarn and fabric. Most apparel exports from the continent come from North Africa, followed far behind by East and Southern Africa. Apparel exports from the African continent go largely to Europe and then North America.

Productive capabilities in the sector in Africa remain low as the continent is largely a net importer of textile and apparel products. A couple of African countries have developed industrial capabilities in textiles and apparel, but most of it has been in the apparel segment and relies heavily on imported fabrics from Asian countries. A significant share of African apparel exports comes from North Africa, followed far behind by East and Southern Africa. West and Central African countries do not have significant apparel exports, and West Africa predominantly exports cotton fibres. 

Most apparel imports on the continent come from China. Chinese imports have largely displaced imports from other Asian countries and the rest of the world but have not substantially displaced intra-African trade, which was already small in 2000 but has shrunk even further. Currently, only 8% of Africa’s textile and apparel imports are supplied by other African suppliers.

African countries have the opportunity to build sustainable textile and apparel industries from the start, which can give them a competitive advantage. The cost of renewable energy technology is falling and renewable energy technologies to power industries are evolving. New fibre and recycling technologies also offer a window of opportunity to leapfrog into the next generation of technologies. Taking advantage of this window of opportunity requires that African governments look forward and not backward, that they think in terms of building new and not rebuilding the existing textile industries. 

African textile and apparel domestic and regional value chains should be based on mastering the next generation of fibre production and recycling technologies. Such a strategy involves taking risks to invest in building the knowledge and skills required for this new technology, but not taking risks will mean that African countries miss the opportunity to move to the technological frontier. Furthermore, African countries can go beyond becoming competitive in the new textile and apparel global supply chains but also use them to drive broader green industrialization processes. 

Such a strategy also involves building diversified regional textile bases as well as regional value chains, which can harness the efforts of more economies and larger market demand as well as create a stronger platform of capabilities with which to engage in global export markets. A diversified regional textile base would make it easier for locally owned apparel firms to emerge and provide the opportunity to move into higher value products as well as move into design of fabrics.

The capital requirements of establishing textile mills and vertically integrated factories from spinning to fabric to garment are very high, and there is limited knowledge in most African countries on how to operate the most modern textile equipment and produce export-quality fabric. Therefore, the first wave of new investments to create a textile base will have to be led by foreign investment. Industrial policy should not be just about attracting foreign investment by providing what they want but also attracting the kind of foreign direct investment that can build globally competitive textile and apparel industries. For example, the Togolese government has started implementing its strategy centres on an eco-industrial park (PIA) with vertically integrated knit factories established through a public-private partnership with Arise Integrated Industrial Platform.

But foreign investment must lead to technology transfer in the form of skilled textile technicians and managers as well as the business acumen side of organising and managing textile and apparel firms. A key part of the industrial policy approach should include assisting local firms in leveraging technology from these foreign firms. Technology is not actually ‘transferred’ but rather must be ‘leveraged’, and to do so; local firms must build their capabilities. The actual process of technology leverage takes place through various forms of investments and contractual relations between foreign and local firms. 

For more information and analysis of African textile and apparel industries, industrial policy, and implications for on-going negotiations in the African Continental Free Trade Area, see:

Lindsay Whitfield, “Current Capabilities and Future Potential of African Textile and Apparel Value Chains: Focus on West Africa, CBDS Working Paper 2022/3, Centre for Business and Development Studies, Copenhagen Business School. 

CBDS Working Paper, Policy Brief & Podcast

Tobias Wuttke and Lindsay Whitfield

When trying to come up with lessons for developing countries that want to industrialize today, people usually refer back to the success of East Asia. But 40 years have passed since South Korea and Taiwan industrialized through exporting and import substitution. Countries trying to industrialize today face a world of globally fragmented production, a world of global value chains (GVCs). The question that has come to the forefront both in academic discussions as well as in policy circles is whether GVC participation can deliver for low-income countries what exporting did for South Korea and Taiwan. South Africa’s integration into the automotive GVC over the past 25 years makes for an interesting case study to shed some light on this question. The development of the South African auto industry has been a story of success and failure at the same time, not least because of the difficulties associated with industrializing in a world of GVCs.

South Africa is the only country in Africa, next to Morocco, that has an automotive industry worth speaking of. It produces 600,000 cars per year, 70% of which are exported, 3 out of 4 of them to Europe. The South African Department of Trade, Industry, and Competition consistently implemented an automotive industrial policy that provided investment support to the foreign-owned automotive assemblers and later support to foreign and locally owned component manufacturers as well. These industrial policies made support for foreign direct investment in the sector conditional on auto assemblers exporting cars and on using some degree of local content.

On the surface, the South African government seems to have followed the industrial policy playbook of the developmental states of the second half of the 20th Century, as illustrated by Alice Amsden. However, the automotive industry has failed to play the role of a leading sector for economic development that it did in South Korea and Taiwan. This is because South Africa’s industry lacks locally owned firms participating at the technological frontier in the industry and it lacks significant linkages with other industries in the country. These limitations result from insufficient industrial policy to support local firms and from the ways in which the automotive global value chain has evolved since the 1990s.

The automotive industry in South Africa makes up 6-7% of the country’s GDP, roughly a quarter of manufacturing output, and employs more than 100,000 people in vehicle and component production. There are three automotive hubs in the country: Gauteng with Ford, Nissan and BMW; the Eastern Cape with VW, Mercedes-Benz and Isuzu; and Durban with Toyota. In addition to the assemblers, South Africa has attracted investment by many of the largest multinational component suppliers, such as Continental, ZF, Bosch, MA Automotive, Dana Spicer, Benteler, etc. There are also around 200 locally owned component manufacturing firms engaged in the automotive value chain.

The automotive industry has lots of potential for backward linkages. A car is made of 10,000 different parts, from various materials – steel, aluminium, plastics, glass, leather, textiles, rubber. Automotive production is characterized by just-in-time (JIT) and just-in-sequence (JIS) requirements, which means that some local footprint is always established once volumes go beyond pure kit assembly. But on the other hand, the automotive GVC is also characterized by follow design and follow sourcing, meaning, for example, that Volkswagen designs their car in Wolfsburg, produces that exact car in its plants all over the world without much local design adaptation, and asks its established large suppliers from developed countries to set up subsidiaries next to its plants in developing country locations to fulfil the JIT and JIS requirements.

This used to be different. Up to the late 1980s, cars were locally designed, and significant and distinct local supplier networks supplied materials and parts to the automotive assemblers in developing countries. This situation has been superseded by the almost total ubiquity of the ‘global car’, with follow design and follow sourcing, since the 1990s. In the case of South Africa, in the early 1990s, several of the assemblers and most of the component firms were partially or fully locally owned. This had already changed by the mid-2000s with all seven assemblers as well as many of the component firms being in full foreign ownership.

The automotive assemblers in South Africa do not undertake local R&D. They assemble vehicles that have been designed in their R&D headquarters. The same is true for multinational component suppliers. In addition to that, none of the locally owned supplier firms in South Africa conduct their own design or product development. Instead, they produce subcomponents based on design drawings either from the assemblers directly or from subsidiaries of multinational suppliers also operating in South Africa. The most common components that locally owned firms produce are plastic injection mouldings, metal pressings, and trim components. The total lack of local R&D is very concerning, given the historical importance of the development of innovation capabilities in local companies in successful catch-up industrializers. It is not unique to South Africa. As Richard Doner and colleagues show in their recent book on the automotive industry, not even Thailand – which has an auto industry more than three times the size of South Africa – has managed to develop local R&D capabilities.

Locally owned suppliers in South Africa complain about low-profit margins, difficulties to access funding, excessive demands from their buyers regarding certifications, audits, open-book-costing, and production flexibility. They still decide to service the auto industry because it has high and steady demand, as vehicle exports mean production volumes are much higher than they would be if firms were producing only for the domestic market. Domestic demand for vehicles is limited by low incomes and high inequality and has stagnated since 2008.

Looking further upstream, the backward linkages from automotive assembly and component production to the local material industries in South Africa have also been disappointing. South Africa has a local steel mill, two large local polymer producers, and a local aluminium smelter. Nevertheless, by far the majority of the materials for automotive production are imported. It is a consequence of follow sourcing and the fact that material requirements for automotive are incredibly high. BMW does not use the same polymeric material for a car bumper as Mercedes-Benz does. They all have their own standards, formulations, blends, and intellectual property around this. This hyper-diversity of materials reduces the scale of production and thus makes firm-level investments into the local production of materials less attractive.

The scale of production in the South African automotive industry – albeit exporting 70% of final output – is still insufficient to localize more sophisticated components like engines and electronics, materials production, as well as tooling and machinery (almost all tooling is imported; all machinery is fully imported). In contrast to exports of vehicles, exports of components are low (with the exception of catalytic converters based on locally available platinum-group metals). The government tries to tackle this scale problem. The South African Automotive Masterplan aims at upping vehicle production volumes from 600,000 to 1.4 million by 2035. Whether this can be achieved remains to be seen. It is certainly true that if production volumes could be increased, this would help achieve further localization. Some countries with larger production volumes like Thailand have for example managed to localize the production of engines and transmissions (for a comparison of localization achievements of different countries, see here).

The South African picture is not unique. The requirements to get some of the historically important elements of industrialization going, like inter-industry linkages and nurturing innovation capabilities in locally owned firms, have increased substantially compared to the time when South Korea and Taiwan started their automotive industries. The gap between incumbents and followers has widened. Doner and his colleagues show that out of the top 100 global automotive suppliers in 2019, only three came from countries other than from Western Europe, the US, Japan, South Korea and China (Motherson from India, Nemak from Mexico and Iochpe-Maxion from Brazil). Incumbent lead firm power in both assembly and component manufacturing is huge. They undertake most R&D and orchestrate global sourcing of materials and components. Other research demonstrates that the denationalization of automotive industries is commonplace across developing countries, see e.g. here and here.

While it has been emphasized in policy circles that the emergence of GVCs has made it easier for developing countries to enter export markets for manufactured goods and led to increases in local firms’ productivity, the case of the automotive GVC illustrates two major limitations of GVC-based industrialization. The potential for the development of inter-industry linkages and innovation capabilities in developing countries participating in the automotive GVC is undermined by excessive incumbent power, follow design, and follow sourcing.

This blog is based on the PhD research carried out by Tobias Wuttke. The PhD project is on the South African automotive industry, researching the technological capabilities of local component manufacturing firms, linkages from the industry to other domestic economic sectors, and the role of industrial policy. Primary data collected throughout 2021 both online and in South Africa includes semi-structured interviews with vehicle assemblers operating in South Africa, materials producers, multinational automotive component manufacturers, locally owned South African automotive component manufacturers, as well as several interviews with policymakers and industry experts. Some of the research was conducted together with Lorenza Monaco of GERPISA, ENS Paris Saclay.

Tobias Wuttke is a Ph.D. Fellow at Roskilde University, Department of Social Sciences and Business.

Lindsay Whitfield is Professor of Business and Development at the Centre for Business and Development Studies, within the Department of Management, Society and Communication at Copenhagen Business School.