Creating transparency in global value chains and their environmental impacts to support sustainability policies

Climate change, air pollution, water stress, and biodiversity loss are the most important global environmental impacts that need to be addressed in the coming decades. This thesis shows that most of these impacts are caused by the extraction and processing of materials, food, and fuels, summarized as “materials” here. With the demand for materials expected to double by 2050, improved sustainability policies are critical. As many materials are produced in another country than ultimately consumed, such policies require detailed information on global value chains and their environmental impacts. Multi-regional input-output (MRIO) analysis plays a key role in providing this information, but several research gaps exist. One gap is the lack of an accurate method for assessing scope 3 impacts of materials, industries, and nations, including cumulative upstream and direct impacts (for any impact category). Also, no method exists for analyzing downstream impacts, which is a particular issue for greenhouse gas (GHG) and particulate matter (PM) emissions of fuels, such as coal. Another gap is the limited spatial and sectoral resolution and the incomplete coverage of sustainability indicators in current MRIO databases. This includes the lack of regionalized assessment of water and land use impacts. Due to these gaps, an accurate and extensive environmental assessment of materials is missing both globally and nationally. The objective of this thesis was to provide an improved MRIO method and database for creating transparency in global value chains and their impacts, to support sustainable policy-making. For this purpose, a method was developed that allows assessing the scope 3 impacts of any sector and region of an MRIO database (Chapter 2), tracking them along the global value chain (for GHG emissions and any other impact category), and analyzing downstream impacts (for GHG and PM emissions of fuels, Chapter 4 and 5). Furthermore, an automated, transparent, and time-efficient approach was developed to improve the resolution and quality of an existing MRIO database (Chapter 3). It was applied to merge the global MRIO databases EXIOBASE3 and Eora26 and add data from FAOSTAT and previous studies to create an MRIO database with high spatial (189 countries), sectoral (163 sectors), and temporal resolution (year 1995–2015). Finally, a set of sustainability indicators was implemented into the database: Climate change from GHG emissions, health impacts from PM emissions (primary and secondary particles), water stress and land-use-related biodiversity loss (both regionalized), value added, and number of workers (Chapter 2–5). The importance, versatility, and broad applicability of the improved method and database was illustrated by several application examples. These include a case study on material production globally (Chapter 2) and for the G20 (Chapter 4). An in-depth analysis of the role of coal combustion is provided in Chapter 4 for the production of metals and construction materials, and in Chapter 5 for global plastics production. A detailed analysis of the food supply chain and the related water and land footprint is shown in Chapter 3 for the European Union (EU). The case study on global material production (Chapter 2) showed that previous MRIO methods either underestimated or overestimated the environmental impacts of material production by 20–60%. The improved method found that material production causes half of global GHG emissions, one-third of global PM health impacts, and, because of biomass production, more than 90% of global water stress and land-use related biodiversity loss. Since 1995, global material-related impacts have increased by 52% (GHG emissions), 56% (PM health impacts), and 22% (water stress). While high-income regions mainly use materials for private consumption, emerging economies use a large share of materials for infrastructure build-up. Although the latter was the main driver of the rising material-related GHG emissions, material-related carbon footprints of high-income regions are still several times higher than those of emerging economies on a per-capita level (year 2015). This underscores the need to decouple environmental impacts from economic growth and to promote sufficiency measures. Material production for building infrastructure in emerging economies, mainly China, has also driven the increase in the G20's overall carbon footprint (Chapter 4). Since 1995, China’s carbon footprint of metals and construction materials has quadrupled, causing more than 10% of global GHG emissions in 2015. Similarly, the case study on plastics (Chapter 5) showed that plastics-related carbon footprints of China’s transportation, Indonesia’s electronics industry, and India’s construction sector have increased more than 50-fold. Thus, measures to reduce, reuse, recycle, and substitute high-impact materials are critical to mitigate the environmental impacts of the expected economic growth in developing countries. Reliance on coal to produce materials has been another key driver of the G20’s rising carbon footprint (Chapter 4). In 2015, half of global coal was used for the G20’s production of metals and construction materials, the majority in China and India. Thus, 85% of India’s total domestic coal was used for the production of these materials in 2015. This points to the need for a rapid phase-out of coal and a shift to renewables in the G20’s material production chain. Similarly, it was found that due to the growth in plastics production in coal-based economies, the carbon and PM health footprint of plastics has doubled since 1995 (Chapter 5). In 2015, 6% of global coal electricity was used for plastics production. Moreover, plastics accounted for 4.5% of global GHG emissions. This is higher than expected, as previous studies did not account for the increased reliance on coal energy in the plastics sector. It was also assumed that equal amounts of oil were used as fuel and feedstock in plastics production, while this thesis shows that twice as much fossil carbon is combusted as fuel than contained as feedstock. Even in a worst-case scenario where all plastics were incinerated, the production stage would still contribute most to plastics-related GHG and PM emissions. This means that previous studies have underestimated the relative significance of the production versus the disposal phase, and thus the enormous potential to reduce the carbon and PM health footprint of plastics by renewable energy investments. High-income regions have significantly contributed to the rising environmental impacts by outsourcing the extraction of resources and processing into materials to lower-income regions with less stringent environmental policies, more water stress, and high biodiversity (Chapter 2–5). Due to increasing imports of plastics from coal-based economies, the share of the plastic-related carbon footprint generated abroad increased to 67% in the EU, 79% in the USA, 90% in Canada, and 95% in Australia in 2015 (Chapter 5). Similarly, the case study on the EU’s water-stress and land-use related biodiversity loss footprint found that most of the associated impacts are caused abroad (Chapter 3). This is mainly attributed to food imports from emerging and developing countries where water is scarce (e.g. Egypt) and biodiversity is high (e.g. Madagascar). The improved spatial resolution (189 countries instead of 49 regions) and regionalized impact assessment led to a significant increase in the EU’s water and land impact footprint induced abroad. These results highlight the need for expanding environmental policy initiatives (e.g., the Paris Agreement and the EU’s Green Deal) from production-based to consumption-based accounting to foster improved supply chain management. This includes the investment in clean energy production throughout the supply chain and the use of regional comparative advantage for reducing water stress and biodiversity loss. In addition to environmental impacts, the value added and workforce associated with material production are also unequally distributed around the world. Trade in materials reinforces this imbalance (Chapter 2–5). It was shown that although high-income regions strongly rely on low-paid work abroad due to material imports (mainly food), they generate most of the associated value added inland (e.g., due to food processing). The extent of this imbalance was highlighted, e.g., in the G20 case study: Since 2011, the number of workers employed globally to meet Australia’s material demand is greater than the number of workers employed in the entire Australian economy (Chapter 4). Similarly, the plastics case study found that although 70% of the workforce required for plastics consumption in the EU was employed abroad, 80% of the associated value added was generated domestically (year 2015), as only the low-paid steps in the plastics value chain have been outsourced (Chapter 5). The method and database of this thesis are open access and can be applied by researchers, industries, and policy makers for a more accurate impact assessment of various materials, commodities, industries, and nations. The method is available as a software tool that can be used to track the scope 3 impacts of industries and nations for a range of sustainability indicators along the global value chain (Chapter 2). Future work can apply the approach of Chapter 3 to improve the spatial, sectoral, and temporal resolution and quality of the database by integrating further MRIO databases and data sources. Also, future work is needed to incorporate detailed bottom-up inventories and use remote sensing data to improve the resolution and coverage of life-cycle inventories