• Global value chains: yesterday, today and tomorrow


    Made in Spain, Made in the USA and even Made in China labels make less and less sense in today’s world. Since firms decided to fragment their production processes and move them to other countries, the label Made in the World probably better represents the nature of most of the manufactured goods we consume. In this article we review the past, present and future of global value chains at a time when pandemic-induced restrictions on travel and supply disruptions have brought them back into the spotlight.



    Pre Titulo
    Área geográfica
    The creation of global value chains

    he 1990s saw the beginning of a far-reaching optimisation of production processes beyond the borders of a single country. Companies decided to fragment these processes and carry them out in as many countries (in order to make the most of each country’s advantages of specialisation), giving rise to what are known as global value chains (GVCs). Several factors helped to encourage the creation of GVCs but first and foremost were the advances made in information and communication technologies (ICTs), which enabled the different production stages to be coordinated perfectly. A second factor was the reduction in trade costs, helped by the important free trade agreements reached during that decade,11 as well as by improvements in transportation, especially by air.

    In fact, GVCs have boosted international trade flows to values that were unthinkable a few decades ago: exports of goods and services as a percentage of GDP rose from around 18% in the early 1990s to levels close to 30% just before the pandemic, while the relative weight of GVCs in total trade flows went from around 40% to just over 50% in the same period (see the chart below).12 

    • 11. 1994 saw the conclusion of the largest round of multilateral trade negotiations (the Uruguay Round), in which 123 countries took part. Also in 1994, the North American Free Trade Agreement (NAFTA) was concluded. Both agreements led to a substantial reduction in tariffs worldwide: from levels of around 16% in the early 1990s to 5% today (according to World Bank data, simple averages).
    • 12. The development of GVCs was particularly dynamic between 1990 and the early 2000s, just before the outbreak of the global financial crisis. Since then, the relative importance of these chains in trade seems to have stagnated.

    The importance of global value chains in trade flows

    Last actualization: 04 May 2022 - 09:16
    The pandemic: present impact and future approaches to GVCs

    The COVID crisis has raised many doubts regarding the high degree of globalisation achieved, as well as the adequacy of GVCs. At first, in countries such as Spain, we became aware of the high external dependence (beyond the EU’s borders) of goods which, at that time, were essential.

    In a second phase, with the strong recovery in demand focusing particularly on durable goods and the disruptions in some factories due to the effects of COVID,13 we have been faced with a global supply shortage problem we had not experienced since GVCs were created. And, in this world of global manufacturing, disruption in one stage of the production chain leads to major disruptions throughout the entire process. The longer the GVC, the greater the impact (the bullwhip effect).

    Such disruptions will undoubtedly change people’s minds about GVCs. Although it is still too early to know what changes the future holds, we can suggest some strategic rethinks company directors are likely to pursue in order to increase the robustness of the production chain.

    First, the chains will probably be shorter to avoid the amplifying effect of disruptions. Secondly, they will be more redundant in key components. In other words, there will be alternatives to the production of these components. Thirdly, they will be equipped with new digital technologies that will enable them to detect chain failures early on. And, in terms of logistics, investment in inventories is likely to increase: from just in time to just in case, as stated in a recent article by the Financial Times14 (see the chart below).

    • 13. See the article «Bottlenecks: from the causes to how long they will last» in the Monthly Report of December 2021.
    • 14. See the Financial Times (December 2021). «Supply chains: companies shift from ’just in time’ to ’just in case’».

    Global value chains are likely to be shorter in order to avoid the amplifying effect of disruptions.


    However, it should be noted that these possible strategic changes, if they occur at all, may be more gradual and less far-reaching than we might have assumed after the shock of the pandemic. One of the reasons is that such changes would entail an increase in costs, with the evident impact on prices consumers would have to pay. In a globalised world, this could mean a significant loss of competitiveness compared with other countries and/or companies. Furthermore, as Harvard professor Pol Antràs has noted, the configuration of GVCs forces companies to incur large sunk costs, which leads to them being extremely rigid regarding strategic production changes.15 

    In other words, the COVID shock will indeed bring about a change in our approach to the configuration of new GVCs and may certainly lead to some rethinking of the existing chains. But, in the latter case, this rethinking might be less radical and rapid than some are predicting.

    • 15. See Antràs, P. (2020). «De-Globalisation? Global Value Chains in the Post-COVID-19 Age». National Bureau of Economic Research, no. w28115.
    The future of GVCs: plus and minus factors

    In addition to the impact of the pandemic, other factors (mostly new technologies) have the capacity to reshape GVCs and we present a brief review (see the diagram below).16


    Automation and 3D printing

    Although automation is a process that has been going on for centuries, today’s robots, equipped with artificial intelligence and at a cost that has decreased substantially over the past few decades, represent a full-fledged revolution. The improved productivity of these new robots may result in some of the manufacturing processes which had been moved to emerging countries in order to take advantage of low labour costs now returning to advanced countries. In other words, we would be shifting from an offshoring to a reshoring trend, which would entail a certain reversal in the globalisation of supply chains.

    On the other hand, 3D printing is a technology that could result in GVCs becoming shorter and, along with this, to the reshoring of part of the manufacturing activity. In fact, with this technology, it is not necessary to send physical products; all that’s required are the computer files to manufacture them! However, there is still no clear evidence in this respect. In fact, a paper published by the World Bank shows a strong increase in trade flows following the adoption of 3D technology in hearing aid production, something we would not expect with a shortening of GVCs.17 Although this is a very specific case, it does reveal some interesting effects that need to be considered. In particular, the hearing aid sector adopted 3D printing for almost all its parts when this became technologically feasible (about 10 years ago) and, since then, trade flows linked to the sector have increased by 60%. The main reason for this growth is that 3D printing has led to a huge reduction in the production cost of hearing aids and an improvement in terms of quality, resulting in a sharp increase in demand for the product. And with greater demand, international trade in hearing aids has intensified.

    • 16. Based partly on Canals, C. (2020). «Revolución tecnológica y comercio internacional 4.0». Geopolítica y Comercio en tiempos de cambio. Published by CIDOB.
    • 17. See Freund, C. L, Mulabdic, A. and Ruta, M. (2020). «Is 3D Printing a Threat to Global Trade? The Trade Effects You Didn’t Hear About». World Development Report.

    The electric car

    Another case that also warrants particular attention is that of electric cars, which have the potential to alter some of the most relevant GVCs (those of the automotive sector), as well as to considerably reduce international trade. The reason is that classic combustion-engine cars require a large variety of parts and gears that are often manufactured in different countries to maximize the competitive advantages of each location. In fact, the automotive sector is responsible for a substantial part of the world’s trade flows of intermediate goods. However, the electric car, with its much simpler mechanics (far fewer parts that are also less subject to wear and tear) could lead to a reduction in these classic intermediate flows and, consequently, to a radical change in the structure of automotive GVCs.

    The production of batteries, a key component for the new electric vehicles, will also determine the future of numerous trade flows, which in this case will focus on raw materials such as lithium, nickel and cobalt.


    Digital technologies and the emergence of new services

    The continuous evolution of ICT, hand in hand with 5G and blockchain technology, will continue to push down logistics costs and, with it, boost the trade flows of goods and services and participation in GVCs. For instance, 5G will support the development of the Internet of Things, which will enable faster and more secure tracking of shipments in the case of goods, and better connections in the exchange of services. Likewise, blockchain has the potential to greatly facilitate international payments.

    On the other hand, these digital technologies will also encourage the emergence of new products, especially services, whose organisation could be decentralised and located in different countries, creating new GVCs in the image and likeness of the chains already established for the production of manufactured goods.


    History reminds us that technological development and international trade are not independent of geopolitical developments.


    Finally, it should not be forgotten that geopolitics has always played an essential role in international trade. In this respect, the USA’s intention to «decouple» from China, especially in the field of technology, could bring about a very significant change in world trade and in how GVCs are managed, especially in the technology sector. Even more so because the US is not alone in wanting to put more distance between itself and other economies. For instance, Europe also seems willing to reduce its external dependence in some technology segments, such as semiconductors, with the European Chips Act.

    In summary, although we do not expect any radical or abrupt change in the form taken by GVCs since they tend to be relatively stable over time, we might see a change in trend in the next few years due to the various 4.0 technologies. In addition to these ongoing trends, factors such as the Coronavirus crisis will further exacerbate certain technological dynamics. However, history reminds us that technological development and international trade are not independent of geopolitical developments. And in this respect, trade-technology tensions between the US and China will play a decisive role.

    Destacado Economia y Mercados
    Destacado Analisis Sectorial
    Destacado Área Geográfica

Have democracies been more lax with lockdowns and testing?

Content available in
Gráfico de barras formado por pilas de monedas de 1 y 2 céntimos de euro

A pandemic is a decisive test for any political system, but under the current conditions this is especially true for democracies. At a time of reduced citizen support for democracy, it is suggested that the response to the COVID-19 crisis from certain autocratic countries – China in particular – has been more effective in the fight against the coronavirus. But the fact is that this question, so raised, misrepresents the reality. With the data available to hand, it seems undeniable that China has been very effective in combating the pandemic. Testing millions of people in a matter of days is not something every country can do. However, we should also not forget that other clearly democratic countries have been able to produce great results in controlling the pandemic, such as New Zealand. Furthermore, other autocratic countries have shown little success in lowering the human cost of the crisis. In short, in order to objectively determine whether or not being democratic is a key factor in the war against the virus, we must be thorough in our assessment, in this case by using an empirical analysis.

What shape should an analysis on such a complex and ever-changing issue take? To start with, when we compare what different countries have done to combat COVID-19, it is clear that the response prior to the availability of the vaccines has essentially consisted of three elements: lockdowns (including restrictions on mobility and social distancing measures), the capacity to test and trace transmissions, and the reduction of mobility that eventually occurs. The first two are health policy tools, while the third is a result of the policies implemented.

Lockdowns and testing, as we know, are aimed at preventing the virus from spreading beyond certain hotspots. Ideally they should be used together: the source of the outbreak is identified early, contacts are detected, and quarantines or lockdowns are imposed on small groups. Unfortunately, for most of 2020, with a low testing and contact tracing capability in many countries, lockdowns were imposed en masse rather than selectively.

Having established the different components of the response, we are now in a position to begin to ask ourselves the key questions regarding the relationships between democracy and the intensity of the use of these tools (lockdowns and testing), and also between democracy and the ability to control social interactions (mobility). The first of these issues will be analysed in this article of the Dossier, while the second will be the subject of the following article.1

Thus, what we want to understand first is the relationship that exists between the political system and the stringency, or laxity, of the lockdown measures and the intensity of testing. In other words, what we first want to know is whether two countries equally affected by the pandemic (measured by the number of cases in proportion of the population)2 have responded with more or less stringent lockdowns depending on whether they were more or less democratic. Secondly, we will ask ourselves whether two countries equally affected by the pandemic have conducted different amounts of testing depending on their level of democracy.3

A first, illustrative glance at the data (see the trend lines in the charts of this article) already suggests a stronger relationship between the number of cases and the severity of the lockdown in more democratic countries. Using a panel regression analysis with controls4 statistically confirms that the response to the first question is favourable towards democracy: countries with a greater degree of democracy have responded with stricter lockdowns when affected to the same degree by the pandemic. This is an important result, as it contradicts the superficial view that democracies have been «weak» and that we must therefore consider how important it is to be more or less democratic.

  • 1. See the article «Have democracies achieved better control over social interactions?» in this same Dossier.
  • 2. It is interesting to consider the extent to which autocratic countries may be reporting fewer cases than are actually occurring, either for political convenience (to reduce suspicions of inefficiency in the fight against the pandemic) or because their data collection systems are less developed than in democracies (which tend to be more advanced countries with better statistical systems). If this were generally the case, then the results would still be even more favourable towards democracies, since the «lukewarm» response from autocracies would be even weaker than suggested, due to the underestimation of the severity of the pandemic.
  • 3. As a step prior to this analysis, it was studied whether the two key public health variables (i.e. how strict the lockdowns have been and the number of tests) have responded to the number of cases. The results are as expected: the authorities of the various countries have reacted by imposing stricter lockdowns and increasing the level of testing when the number of cases grows (and vice versa).
  • 4. The relationship between the stringency or laxity of lockdown measures in a given country, the impact of the pandemic there and its political system is analysed using the following panel regression with fixed effects: \(Stringency_{i,t}=\alpha_0+\alpha_1Covid_{i,t}+\alpha_2\;\lbrack Covid_{i,t}\times D_i\;\rbrack+u_i+u_t+\mu_{i,t}\) where \(Stringency_{i,t}\) is a measure, developed by Oxford University, of how strict the lockdown measures have been in country i and on day t, \(Covid_{i,t}\) is the number of daily new COVID-19 cases in proportion to the country’s population, and \(D_i\) is a measure of the country’s political system on a scale ranging from 0 to 100 from less to more democratic, based on a measure of the political rights and civil liberties in 195 countries developed by Freedom House. The coefficient \(\alpha_2\) can be interpreted as the difference in the stringency of the lockdown measures to curb the impact of the pandemic in democracies relative to authoritarian countries. The data cover 102 countries between February and November 2020. This model follows a similar approach to that used by C.B. Frey, G. Presidente and C. Chen (2020). «Democracy, Culture, and Contagion: Political Regimes and Countries Responsiveness to Covid-19», Covid Economics 18.
Relationship between the number of COVID-19 cases in a country and the stringency of restrictions in autocratic and democratic countries

Based on our estimate model, we can conclude that the institutional factor plays a rather important role. A practical example will help to understand the sensitivity. With 100 being the highest possible level of democracy, according to the indicator by Freedom House, Spain has a level of 92 and Colombia, a level of 55. Now suppose that the intensity of the pandemic (measured by the number of new cases per 100,000 inhabitants) were to multiply by a factor of 10 (as happened in Spain between 6 and 13 March, for instance). In this scenario, the result suggested by the model is that the change in the stringency of the lockdown would be 1.5 points higher in Spain than in Colombia. How should we interpret 1.5 points? It would be the increase in the stringency of the lockdown that would occur when the authorities go from recommending staying at home to imposing a compulsory residential lockdown with the exception of leaving home for work, school, exercise and essential purchases.5 This is by no means a minor jump.

The same result is obtained when using the second health policy tool, testing capacity. The model’s results suggest a greater response from democracies than from autocracies. As was the case with the strictness of the lockdowns, being more or less democratic represents a significant difference.

Ultimately, contrary to that suggested by some criticisms, democracies have not been lax in their fight against the COVID-19 pandemic compared to autocracies. A different question, and indeed a more consequential one, is whether their responses have been effective in addition to «hard». In other words, we must return to the second question mentioned at the beginning and ask ourselves whether the desired social behaviour, namely control over mobility, has been achieved. We will explore this issue in the next article, allowing us to get closer to answering the complex question of whether democracies have measured up to this great health crisis.

  • 5. The index which measures the stringency of the lockdown is the aggregate of a set of social distancing metrics, one of which is the degree of residential lockdowns. When this metric shifts from being a mere recommendation to stay at home to an obligation to do so, the aggregate stringency index increases by 1.5 points.