Is the Global Value Chain also a Global Pollution Chain?

Yuwan Duan
Jun, 2024
#Trade and FDI
#Environment and climate change
One of the most well-known theories in international trade and environment is the “Pollution Haven Hypothesis” (PHH). According to the theory, high-income countries with strong environmental regulations will have a comparative disadvantage in pollution-intensive industries, and will tend to offshore their polluting industries to poorer countries. Hence, developing countries will become pollution havens, while the developed ones will specialize in and export clean goods.

Whereas several studies have empirically tested the PHH, they fail to account for the prevalence of global value chains (GVCs)—whereby goods that used to be produced within one country are now fragmented and distributed across global networks of production—as a new way of organizing production. Indeed, available evidence suggest that more than 50 percent of global production are now organized through GVCs (Sampath & Vallejo, 2018; World Bank, 2019). In this case, an empirical test on PHH that ignores GVCs may greatly obscure the impacts of the pollution haven effect and generate unreliable results. This blog discusses results and draws policy implications from our recent study (Duan, Ji and Yu, 2021) where we account for GVCs in an empirical test of the PHH. One of the key findings is that that GVCs are evolving into global pollution chains. Dropping them from the analysis masks much of the pollution haven effect.

GVCs and the Pollution Haven Effect

In principle, the emergence of GVCs as a new way of organizing global production may change prior understanding about the PHH for at least two reasons. First, the pollution content embedded in intermediate goods would be wrongly attributed to the country where downstream production occurs, even though that country is neither responsible for producing the pollution nor actually suffers from it (Duan and Yan, 2019) (see Figure 1). Consider an imaginary GVC with 3 countries: China, Germany, and the U.S. Germany produces smart phones using iron produced in China as intermediate input, and then exports the phones to the U.S. In this GVC, China exports to the U.S. indirectly and generates emission for US consumers by providing intermediates to Germany. However, gross trade statistics would show that China didn’t export anything to the US and didn’t generate any emissions to cater for consumers’ demand in the US. Moreover, the traditional statistics would exaggerate the true trade value and the induced emissions that Germany exports to the US by ignoring that part of the phone value and corresponding emissions were actually created in China. In this scenario, the PHH is not valid. Importing inputs with high carbon footprint from China was largely based on “traditional” comparative advantage in these inputs rather than on a German strategy to outsource polluting production stages.


Second and more importantly, GVCs may allow firms from countries with stringent environmental regulations to outsource some of their production stages - instead of the whole production process- to countries with less stringent regulations to avoid environmental costs. As a result, the so-called “dirty” industries are not easily identified in the context of GVCs. A country may be responsible either for a “clean” or a “dirty” stage of a product. Moreover, a typically “dirty” industry may be clean for a country that specializes only in the “clean” stage. Therefore, studies that do not properly account for GVCs may seriously distort the true environmental consequences of international trade.
Figure 1, How GVC reshape emission patterns.

Accounting for GVCs in PHH empirical test

In a recent study, Duan et al., (2021) re-examined the PHH by uncovering the implicit trade flows and emissions that go from original producers to final consumers. We distinguished two types of trade data: i) traditional trade, referring to the traditional trade statistics from the Customs; ii) value-added trade, which fully accounts for GVCs by referring to the value-added in one country to satisfy final consumption in another country, introduced by Johnson and Noguera (2012). We computed the emission intensity (emission induced by per unit of trade, following Duan and Jiang (2018), and singled out the composition effect (the effect of industry composition on emission intensity) by removing the influence of heterogonous technologies across countries in the two types of bilateral trade. Based on the panel data from the World Input-Output Database (WIOD) for 40 major economies from 1995 to 2009, we investigate whether higher-income countries, all things equal, import pollution intensive products from poorer countries. The results show –inter alia- that the larger the per capita income gap between importing and exporting economies, the more pollution-intensive the value-added exports. This validates the existence of the PHH when GVCs are properly accounted for using value-added trade data. However, when using traditional gross trade data (reflecting trade and specialization based on “traditional” comparative advantage rather than fragmentation within industries), the PHH disappears. Thus, our findings suggest that high-income countries are likely to offshore their pollution to developing countries by outsourcing only those “dirty” stages of production instead of the entire production process to low-income countries. GVCs are therefore evolving into global pollution chains.

Policy Implications

Our results carry some important policy implications for a cleaner global production. Although GVCs broadened the international division of labor and enabled poorer countries to participate in manufacturing high-end products, they also made it easier for wealthier countries to selectively outsource the most-polluting production stages to poorer ones. This calls for a new collective action at the multilateral level. First, multi-nationals (who are often the main players in GVCs) should strengthen Green Supply Chain Management to reduce the environmental footprint of their production. By investing in eco-design, green purchasing, green manufacturing, and reverse logistics, they could effectively decrease the emissions embodied in the product value chain. Governments and firms could collaborate to improve the environmental monitoring capability for each production stage and make the relevant emissions data publicly available.

Secondly, developing countries should design their economic and environmental policies more carefully. They should benefit from opportunities brought about by GVCs, but also attempt to balance their specialization in production stages matching the country’s development needs with a reduced impact of this production on the environment by adopting cleaner production technologies. Given that developed countries usually possess cleaner production technologies, multilateral efforts and technology transfer could be an effective tool to fight the climate crisis together.

Finally, intensifying environmental provisions in trade and investment treaties and ensuring their implementation can be a promising way forward when countries seek to promote economic cooperation (Brandi et al., 2020). Environmental provisions could encourage green technology transfer or foreign investment in renewable energy production and green industries. These treaties, while promoting trade and investment, can create a level playing field with respect to environmental standards and prevent the so-called “race to the bottom.”
Brandi, C., Schwab, J., Berger, A., & Morin, J. F. (2020). Do environmental provisions in trade agreements make exports from developing countries greener?. World Development, 129, 104899.

Duan, Y., Jiang, X. (2017). Temporal change of China's pollution terms of trade and its determinants. Ecological Economics 132, 31–44.

Duan, Y., Yan, B. (2019). Economic gains and environmental losses from international trade: a decomposition of pollution intensity in China's value-added trade. Energy Economics 83, 540–554.

Duan, Y., Ji, T., Yu, T. (2021). Reassessing pollution haven effect in global value chains, Journal of Cleaner Production, 284 (15), 124705.

Johnson, R.C., Noguera, G. (2012). Accounting for intermediates: production sharing and trade in value added. Journal of international Economics 86(2), 224–236.

Sampath, P. G., & Vallejo, B. (2018). Trade, global value chains and upgrading: what, when and how? The European Journal of Development Research, 30(3), 481-504.

World Bank (2019). World Development Report 2020: Trading for development in the age of global value chains. Washington, DC: World Bank.

Yuwan Duan

Associate Professor, School of International Trade and Economics, Central university of Finance and Economics, China

Scroll to Top