3. Moving to sustainable industrial production

This report proposes adopting a well-being lens as a framework for mitigation policies. Its first component is to define societal goals in terms of well-being and systematically reflect these goals in decision-making across the economy, putting people’s well-being at the centre of policy making. Its second component is to ensure that policymakers consider multiple well-being objectives, rather than focusing on a single issue (or a few issues) across sectors. The final component of the well-being lens is to acquire a thorough understanding of the system in which a policy intervenes, in order to grasp the broader interactions between different facets of well-being. Using this lens helps identify and design policies that can deliver two-way alignment between climate change mitigation and other well-being priorities, so that these goals are mutually reinforcing rather than antagonistic. This chapter applies a well-being lens to heavy industry, which comprises energy-intensive trade-exposed industries. It focuses on iron and steel, cement, non-ferrous metals (e.g. aluminium), pulp and paper, and chemicals (e.g. ammonia). It does not discuss refineries, which are touched upon in other chapters.

Heavy industry links to nearly every aspect of current and future well-being. It transforms the planet’s raw materials into products for society. Over the last century, the planet has experienced unprecedented urbanisation, rising standards of living and a larger population than ever before, all of which has led to increasing demand for products from heavy industry. In 1970, the world produced 22 billion tonnes of primary raw materials (i.e. materials sourced from mining or extraction in their raw form that are entering the economy for the first time) globally. This volume has grown to 70 billion tonnes by 2010,1 twice as fast as population growth during the same period (OECD, 2019[1]). Driven by living standards, population growth and urbanisation, the major focus of heavy industry has been on maximising production to meet these growing demands profitably. This meant increasing output while reducing costs, typically through improvements in production efficiency, understood as minimising inputs – labour, capital, energy and other intermediate inputs – for every unit of production.

While heavy industry has successfully met this growing demand for materials and chemicals, it has done so in a way that makes a major contribution to climate change. Heavy industry requires high temperatures for materials production, chemical feedstocks and other specialised process chemistries, rendering it very energy and emission-intensive. It is responsible for roughly 36% of annual global carbon dioxide (CO2) emissions (compared to only 5.5% for the rest of industry) (IEA, 2019[2]); these emissions are rising significantly faster than total global CO2 (Hoesly et al., 2018[3]).

The heavy industry sector emits roughly 36% of global CO2 annually (compared to only 5.5% for the rest of industry) (IEA, 2019[2]); these emissions are growing significantly faster than total global CO2 (Hoesly et al., 2018[3]).  

In terms of broader well-being, the extraction and processing of raw materials can irreversibly alter ecosystems through physical alteration of the landscape, waste and other by-products. Some of these ecosystem changes may even subsequently alter local climate conditions, as seen with rising local temperatures caused by deforestation from mining (Wolff et al., 2018[4]). Pollution of the surrounding air, water and soil damages biodiversity and threatens human life. Thus, heavy industry’s dependence on the planet’s resources (i.e. energy, land, water and raw materials) poses challenges for sustainability because of the growing competition between agriculture, energy and industry sectors. The extent of such competition depends on the speed and direction these sectors innovate.

There exists a risk of perpetuating or even worsening these well-being losses into the future. If current trends continue, models project a doubling of demand for materials in the next 50 years from 89 Gt in 2017 to 160 Gt in 2060 across all major categories of materials – metallic ores, non-metallic ores, biomass and fossil fuels (OECD, 2019[1]). In light of the well-being impacts outlined above, this calls not just for innovation but also for reassessment of heavy industry’s priorities, and further reflection on the use of resources to deliver materials and chemicals to society.

The inadequacy of the extract-process-consume-dispose economy (otherwise known as the “linear economy”) is quickly apparent when adopting a well-being lens. Heavy industry can only promote well-being if the broader economy shifts to a net-zero, circular and resource-efficient model. These broader changes will, in turn, enable heavy industry to reach net-zero, circular and resource-efficient production.

This report focuses on mitigation. It highlights a number of different strategies to decarbonise heavy industry on both the supply and demand sides (IPCC, 2018[5]). Options include improving energy efficiency; increasing the use of low-carbon electricity; using more recycled materials; modifying the existing processes to employ carbon capture, utilisation and storage (CCUS); identifying alternative heat sources for existing processes; and even switching fuels completely (e.g. through direct or indirect electrification, bio feedstocks, or hydrogen) (Bataille et al., 2018[6]; Davis et al., 2018[7]). There is limited time to meet stringent climate change mitigation goals. The multi-decadal life of industrial facilities means all new industrial facilities must be net-zero by 2030 to 2055 to achieve the 1.5° to 2°C temperature goal (Bataille et al., 2018[6]). Facilities’ investment cycles typically last 20 to 40 years, meaning firms are only one – or at most two – investment cycles away from the middle of the century (Wyns, Robson and Khnadekar, 2018[8]).

Nonetheless, compared to other end-use sectors like transport, only a subset of heavy industry processes can be cheaply and directly electrified; hence, new processes (e.g. new cement chemistries) will be required. Many of the existing options are expensive (e.g. the HIsarna process for steel) or technically difficult (Bataille et al., 2018[6]; Davis et al., 2018[7]). However, the technological possibilities for decarbonising are constantly evolving at the frontier of innovation; demonstration and deployment to establish the commercial viability of such new technologies are also vital. Reducing demand for products from heavy industry through greater materials efficiency will be just as important as supply-side measures, but caution should be exercised with regard to prioritisation. Concentrating on the demand-side can slow the deployment of low-carbon production technologies, and therefore, the rate of reduction in the emission intensity of materials production (OECD/IEA, 2019[9]).

The methods used to decarbonise heavy industry will have great implications for other dimensions of well-being, which cannot be ignored. On the one hand, the sensitivity of heavy industry to increased production costs adds an extra dimension of complexity to competitiveness. Most of heavy industry operates at intermittent profitability and is usually highly exposed to trade, with little capacity to hand down costs to consumers (1-5% at most). Hence, extra costs linked to decarbonisation often lead to concerns over competitiveness, which in turn could lead to job losses and the “death” of communities, since heavy industry tends to be located in rural and remote areas. On the other hand, various pathways to decarbonisation can help manage the planet’s resources, by using less energy or better managing water consumption; others can help reduce pollutants and waste to maintain a healthy and safe environment. Moving forward, governments will need to navigate this tangled web of interests and unknowns with carefully crafted policy packages that create two-way alignment between all of these well-being priorities.

This chapter analyses the impacts of heavy industry on well-being, now and in the future. Section 3.1 calls for expanding policy priorities to guide decisions in the sector in a way that can secure wider well-being positive impacts, as well as anticipate and avoid potential trade-offs. Section 3.2 proposes a set of indicators to monitor the delivery of these various priorities and help countries effectively prioritise action to ensure progress in attaining them. It shows how these indicators can complement the Sustainable Development Goals (SDGs) and the OECD Framework for Measuring Well-being and Progress (henceforth the OECD well-being framework).

Despite its relatively small share of global gross domestic product (GDP), heavy industry underpins the current economy, processing almost all materials and chemicals presently in use (Wyns, Robson and Khnadekar, 2018[8]). Hence, countries often consider heavy industry strategic to economic development (Silva and Mattera, 2018[10]). Steel, in particular, is considered key not only for the economy, but also for defence. Securing domestic sources of steel is therefore often an important consideration for policy makers (Silva and Mattera, 2018[10]).

Heavy industry faces very high fixed costs upfront, leading these industries to maximise output even in periods of low demand (Silva and Mattera, 2018[10]). Rising standards of living, increasing urbanisation and population growth are also creating increasing demand for products. These combined factors provide an even greater push to increase output and improve efficiency to reduce costs, i.e. minimising inputs for every unit of output. As Figure 3.1 shows, such inputs are human – e.g. workers and their expertise (red box); natural – e.g. water, land and raw materials (blue box); and produced – e.g. energy and machinery (yellow box). In return, heavy industry creates the processed materials and chemicals needed by society (yellow box), along with wages for those workers (red box), as well as other less desirable by-products, such as pollution, waste and greenhouse gases (GHGs) (blue box).

Heavy industry is keeping up with the pace of demand. Primary aluminium production nearly doubled over the last decade, from 38 971 Mt in 2008 to 64 336 Mt in 2018, driven mainly by increased production in China (OECD, 2019[1]). Likewise, crude steel production increased from 1.3 billion metric tonnes in 2008 to 1.8 billion metric tonnes in 2018 (Mercier and Mabashi, 2019[11]). Ramping up production worldwide is adding jobs and developing regions worldwide. Even though these jobs account for a relatively small share of employment globally, they link indirectly to sectors across the economy. For example, steel employs only 6 million people globally, but links indirectly to 42 million jobs (World Steel Association, 2019[12]).

Nevertheless, automation is gradually replacing these jobs; similarly to automakers, heavy industry is shifting towards automating processes (Kherat, 2019[13]). Industrial robots will likely replace nearly 20 million manufacturing jobs (approximately 8.5%) globally by 2030 (Oxford Economics, 2019[14]). The declining costs and growing capabilities of robots (using artificial intelligence), combined with ever-increasing demand for goods, is prompting major producers like China to invest heavily in automation (Oxford Economics, 2019[14]). Since 2000, the European Union has lost 400 000 jobs to automation, China 550 000, the United States 260 000, and South Korea 340 000 (Oxford Economics, 2019[14]). The need to care for displaced workers to avoid entrenching social inequalities is increasingly recognised, as low-income and rural sparsely populated areas will be the most vulnerable to job losses from automation and the low-carbon transition (Oxford Economics, 2019[14]). The latter is an issue addressed by the 2015 ILO guidelines for a just transition (ILO, 2015[15]).

Despite this trend, heavy industry – and industrial policy – are maximising production, to the detriment of other aspects of current and future well-being. The growth in production by heavy industry over the last few decades substantially increased energy-related GHG emissions, further exacerbating climate change. Heavy industry emitted 11.8 GtCO2 (36% of global CO2) in 2016, plus 3.1 GtCO2 (9%) for fossil-fuel production (which is traditionally counted within heavy industry), making it the single-largest emitting sector when allocating CO2 emissions from electricity to consuming sectors (blue arrow in Figure 3.1) (IEA, 2019[2]). Iron and steel (31%), and cement and concrete (19%), are the top two emitters from heavy industry (see Figure 3.2 for a further breakdown).

By-products from heavy industry – i.e. waste, sludge and dust (blue arrow in Figure 3.1) – in some parts of the world may pollute the air, water and soil, damaging biodiversity through water acidification, eutrophication, and aquatic and terrestrial ecotoxicity (OECD, 2019[1]). This exposure can persist over time, e.g. sulphuric acid rain from burning coal persisted over much of North-eastern United States until the 1990s. It can also be very acute, as shown by the collapses of the Brumadinho dam in 2019 and the Mariana dam 2015, which released millions of litres of waste from mining iron ore in Brazil, killing people and destroying biodiversity. This pollution may then harm human health directly and indirectly, by contaminating food and water (Table 3.1). The use of the planet’s resources, i.e. land, water and raw materials (blue box), and energy (yellow box) – could pose future sustainability problems, owing to increasing competition for these resources between heavy industry, agriculture and energy (OECD, 2017[16]). In January 2018, Cape Town faced the stark reality of running out of water in three months; the city resolved this crisis by cutting off industry’s access to consumable water. These types of choices will become increasingly common in our future: total global water demand (i.e. the amount of water withdrawn from freshwater sources) is projected to grow by 23% between 2015 and 2060; industry accounts for 38% of this demand (OECD, 2017[16]). How to meet this increasing demand in the context of potential resource scarcity in the future is unclear.

Demand for materials and chemicals from heavy industry will rise in the coming decades; exactly how much it will rise will depend on society’s response to the challenges ahead. For example, if trends continue globally (and if the economic structure remains roughly the same), growth of primary and secondary metal production – e.g. aluminium, copper, iron and steel – will most likely continue at the same rates over the next 50 years (OECD, 2019[1]). According to projections, a continuation of present trends would double demand for primary materials (from extracted raw materials) between 2017 and 2060, primarily from emerging and developing countries (OECD, 2019[1]). Although this could bring jobs and regional development, the extent to which jobs materialise will depend on advancements in automation and speed of adoption. Nevertheless, there would also be a near doubling of environmental impacts from primary material production – e.g. GHGs, acidification, eutrophication, land use, and aquatic and terrestrial toxicity (see Table 3.1) (OECD, 2019[1]). This would also place the Paris goals out of reach, undermining the prospects of well-being for future generations.

Avoiding these losses in well-being – and a future where mistakes from the last century are repeated – entails a readjustment of heavy industry’s priorities. Heavy industry will still need to maintain production while caring for workers, but this should not occur at the expense of other aspects of well-being, notably limiting climate change; maintaining a healthy and safe environment; and sustainably using the planet’s resources, such as energy, land, water and raw materials.

The priorities outlined above are unattainable for heavy industry in the context of the present emission-intensive linear economy. A holistic shift towards a net-zero, circular and resource-efficient economy is essential. Such an economy aims to reach net-zero GHG emissions by the middle of the century, and “to keep products, components and materials in the economy for as long as possible, trying to eliminate waste and virgin resource inputs,” (McCarthy, Dellink and Bibas, 2018[24]). Crucially, each of these shifts is imperative and additive: decarbonisation of production is needed in addition to circularity and resource efficiency. Transforming the economy will take a lot of time, since the behaviour of billions of consumers and producers – as well as end-use sectors for industrial products (e.g. construction standards) – will need to change. Moreover, the downgrading of recycled material brings qualitative and quantitative losses, requiring additional primary materials to meet increasing demand (e.g. airplanes require purer aluminium than food cans). In addition, some materials presently in use are not immediately available for re-use and recycling (e.g. buildings have life cycles lasting decades). New primary materials will therefore continue to be needed to satisfy continual demand (van Ewijk, 2018[25]), meaning that heavy industry needs to decarbonise production.

A number of routes exist on both the demand and supply sides to decarbonise heavy industry. With respect to the two most intensive sectors, i.e. iron/steel and concrete/cement, the Energy Transitions Commission (Energy Transitions Commission, 2018[26]) estimates that:

  • Up to 38% of iron and steel emissions could be reduced through demand management (greater and better scrap recycling, redesigning products for materials efficiency and circularity); up to 20% through improvements in energy efficiency (re-use of high-pressure gas to power other equipment, coke dry quenching, closure of inefficient plants); and up to 100% through decarbonisation technologies (scrap-based electric arc furnaces, contingent on the availability of low-GHG electricity; natural gas-based direct reduced iron (transition); hydrogen-based direct reduced iron; carbon capture and storage [CCS]; and direct molten oxide electrolysis of iron ore).

  • Up to 34% of CO2 emissions from cement and concrete could be reduced through demand management (i.e. designing buildings more efficiently, recycling un-hydrated concrete, re-using concrete and substituting timber for concrete); 10% through energy efficiency (e.g. switching to dry kilns, multistage cyclone heaters and decreasing the clinker-to-cement ratio); and the rest from decarbonisation technologies (e.g. gas, biomass/waste heat generation and kiln electrification).

In a scenario consistent with a 2°C goal, projected emissions from power generation will decrease by around 90% relative to 2010, compared to only around 50% from industry. Consequently, most of the additional emission reductions required to meet a 1.5°C goal, rather than a 2°C goal, would require more challenging emission reductions from industry and other demand sectors (buildings and transport), which are likely to have much higher marginal costs (Luderer et al., 2018[27]).

Nevertheless, changing the course of production brings a very real, short-term loss of well-being for some individuals, i.e. heavy industry employees. If all regions of the world implemented a carbon tax of 50 USD (US dollars) per tonne today, the mining and fossil-fuel sector would lose about 8% jobs in OECD countries and 6% in non-OECD countries, while construction, chemicals and other heavy industries would lose less than 5% globally (Chateau, Bibas and Lanzi, 2018[28]). In absolute terms, the total number of jobs lost in such a scenario would only amount to 21 million, on par with the expected 30 million global job losses from automation by 2030. The bigger threat to these communities is automation rather than decarbonisation, yet jobs will also be added by this shift in production. For example, almost 6 million jobs can be created globally by moving away from the linear economy to embrace the recycling, re-use, remanufacture, rental and longer durability of goods (ILO, 2018[29]).

Even though a relatively small number of jobs may be lost from this transition and it may even be possible to reallocate workers, real people work directly in these emission- and resource-intensive industries. In a typical myopic focus on mitigation, changing production often means confronting a reality where these people are out of work and communities are potentially unravelling. The key of the well-being lens lies in carefully designed policies and the right measurement tools that inform the design of these policies addressing these workers and communities. The well-being lens prompts decision makers to evaluate these costs, identifies these communities and establishes proper measures to monitor and improve policies. The next section provides an example of an indicator that can signal “at-risk” communities.

Other priorities for heavy industry – maintaining production while caring for workers, increasing resource efficiency or reducing pollution – can help limit climate change and achieve two-way alignment. For example, Kalundborg Symbiosis, the first functioning example of industrial symbiosis (i.e. greater resource efficiency), includes several (public and private) facilities that exchange energy, water and materials in closed loops. In 2015, Kalundborg reduced its use of drinking water by 3.6 million m3 and 87 000t of materials (gypsum, fly ash, sulphur, sand and ethanol), while reducing emissions by 635 MtCO2 (equivalent to the per capita CO2 of 75 000 Danes) (Ellen MacArthur Foundation, 2017[30]). Table 3.2 explains further how these other priorities can align with climate change mitigation, and how potential trade-offs with climate change mitigation may arise when pursuing these other priorities.

Section 3.2 called for expanding heavy industry’s priorities beyond maintaining production and caring for workers. Failure to do so creates a risk of perpetuating – and even worsening – future losses of well-being. This expanded set of priorities includes limiting climate change, maintaining a healthy and safe environment, and sustainably using the planet’s resources. Achieving these priorities means the upcoming era needs net-zero, circular and resource-efficient production. In this shift, policy makers need indicators to track whether heavy industry actually attains these priorities and is producing what society needs, without undermining well-being, limiting climate change, etc.

Although this report is far from the first to propose a set of indicators that depict a fuller picture of well-being, it is one of the first to do so at a sectoral level. The SDGs and the OECD well-being framework laid the groundwork for well-being priorities and indicators at level of the whole economy. In fact, the priorities defined here noticeably link with those in these existing frameworks. Table 3.3 lists these priorities, mapping them to relevant goals in the OECD well-being framework and SDGs. This section walks through these well-being priorities for heavy industry step by step and proposes indicators that can help translate these goals into measurable outcomes. It also discusses the relationship between the indicators proposed and the existing indicators in both the SDGs and the OECD well-being framework. This list of indicators is not exhaustive; rather, it suggests the types of indicators that could be useful and the type of data enhancements needed.

The present section differs from a traditional industrial mitigation report. The indicators proposed do not provide detailed measures for drivers of emissions, which would simply repeat many years of work performed by other institutions. Instead, the section focuses on indicators that can measure outcomes relative to different policy goals to grasp the interactions between different facets of well-being, along with their impacts. Using these types of indicators – especially simultaneously – will help, for instance, identify opportunities for two-way alignment, by better capturing the potential positive and negative impacts of different decarbonisation pathways on multiple well-being priorities.

The previous section called for heavy industry to shift towards net-zero, circular and resource-efficient production. This is echoed in SDG 9.2, “promote inclusive and sustainable industrialization and, by 2030, significantly raise industry’s share of employment and gross domestic product, in line with national circumstances, and double its share in least developed countries”, and the OECD well-being framework dimension on “future resources”, which includes economic and natural capital (Table 3.3). To incentivise “sustainable industrialisation”, current measures of success need to be adjusted (as argued in Chapter 1). Productivity measures need to value environmental quality and environmental inputs explicitly, which would create two-way alignment (at least in terms of measurement) between the goals of production on the one hand, and limiting climate change, sustainably using the planet’s resources, and maintaining a healthy and safe environment on the other hand. Historically, increases in productivity lead to significant increases in CO2 and other pollutants (Empora and Mamuneas, 2011[31]; Kalaitzidakis, Mamuneas and Stengos, 2018[32]), signalling that productivity measures do not fully capture well-being.

Neither the SDGs nor the OECD well-being framework offers a viable indicator as an alternative to traditional measures of productivity. The key indicator for SDG 9.2, value added of manufacturing as a proportion of GDP, measures a given industry’s contribution to the economy using the System of National Accounts, although there is ongoing effort to adequately account for the environment in this through the System of Environment and Economic Accounts. However, there exists an incongruity in the existing SDG framework between the stated goal of SDG 9.2 and the chosen indicator. Therefore, if this indicator is used in decision-making, it will lead to unsustainable choices that perpetuate losses of well-being. The OECD well-being framework does not propose an alternative measure of productivity.

The environmentally adjusted multifactor productivity (EAMFP) indicator in the Green Growth Database (covering 51 countries since 1990) fills gaps in existing frameworks and addresses the shortcomings of existing measures. Traditionally, multifactor productivity (also known as total factor productivity) measures the share of output that cannot be explained by either labour or capital inputs. Economists view this as a measure of efficiency, i.e. technological innovation, even though it has been criticised. By contrast, the EAMFP indicator measures the share of output that cannot be attributed to a given set of inputs, while accounting for the consumption of natural resources and environmental outputs (Cárdenas Rodríguez, Haščič and Souchier, 2018[33]). EAMFP is not the only indicator that adjusts productivity measures; it is merely an example of the types of indicators that could be useful.

First, the EAMFP indicator includes the private cost to firms to extract the natural capital (by including resource rents) of 14 subsoil assets of fossil fuels (hard coal, soft coal, gas, oil) and minerals (gold, iron ore, lead, nickel, phosphate, bauxite, copper, silver, tin and zinc), encompassing many of the key raw materials used by heavy industry. Valuing the extraction of raw materials enables policy makers to compare the costs of primary materials versus secondary materials more accurately, leading to better management of finite natural resources. Using this valuation, it is feasible to calculate the amount of GDP growth due to the extraction of natural capital. As a result, the indicator helps manage the planet’s resources sustainably, fostering two-way alignment between these priorities. Other forms of natural capital, such as land or water, could also be included. Cárdenas Rodríguez, Haščič and Souchier (Cárdenas Rodríguez, Haščič and Souchier, 2018[33]) provide further details on methodology.

Additionally, the EAMFP indicator values undesirable outputs such as air emissions, including three GHGs – CO2, CH4, N2O – and five air pollutants – sulphur oxides, nitrogen oxides, PM10, carbon monoxide and non-methane volatile organic compounds. Calculating a shadow price of the number of foregone units needed to reduce one unit of pollution allows an estimation of pollution-adjusted GDP growth, a critical facet of maintaining a healthy and safe environment, and limiting climate change. This contrasts with traditional measures of multifactor productivity, which do not explicitly value pollution or emissions because these frequently unpriced.

If productivity measures fully value environmental quality and the natural environment, then production will shift away from emission-intensive and resource-intensive facilities, and jobs will be lost. Therefore, an accompanying indicator to the EAMFP – or similar measures – identifies regions at risk of “losing” from sustainable production to better target policies aiming to help these communities and workers. The indicators for SDG 8.5 (relative to the unemployment rate) and SDG 9.2 (relative to the share of employment in manufacturing) lack the granularity to evaluate effectively which policies should be used to gauge the impacts of net-zero, circular and resource-efficient production. One indicator to track this is the U.S. Cluster Mapping Project, a national initiative that amalgamates over 50 million open-data records on industry clusters in the United States. Led by Harvard University, the U.S. Department of Commerce and the U.S. Economic Development Administration, the project groups industries into clusters, e.g.  Biopharmaceutical cluster (US Cluster Mapping, 2019[34]). A Cluster Dashboard provides data on economic performance, geographic presence, and sub-cluster and industry composition. The geographic presence of any sub-cluster can be shown at the level of the state, economic areas, metro/micropolitan areas and counties. The economic indicators include level of specialisation, absolute level of employment, employment growth rate, job creation, annual wage, annual wage growth rate, number of establishments, establishment growth rate, establishment formation, patent count (indicator for innovation) and patent growth rate. Figure 3.3 is an example of changes in employment (number of jobs) between 2010 and 2016 for iron and steel forging (a sub-cluster of upstream metals manufacturing) in the United States, by state using data from US Cluster Mapping.

Table 3.4 summarises the key points from the section above. The blue columns show the well-being priority for heavy industry, the corresponding SDGs, and the well-being dimension and domain (introduced in Table 3.3), as well as the indicators attached to these frameworks, to reveal how the EAMFP indicator and the Cluster Map can complement them.

Heavy industry needs to reach net-zero emissions by the middle of the 21st century. This is echoed in the framework for SDG 13, “take urgent action to combat climate change and its impacts”, and the OECD well-being framework, which includes natural capital under the “future resources” dimension. Neither of these frameworks provides indicators on emissions from heavy industry. This section, therefore, suggests indicators of emissions from: a) energy use; and b) the processes used in the chemical and physical transformations undertaken by heavy industry. As a complement, a vast line of work explores the drivers of heavy industry emissions (which lies outside the scope of this chapter), e.g. Tracking Industrial Energy Efficiency and CO2 Emissions (OECD/IEA, 2007[35]).

Estimating process emissions is challenging, since they vary according to the technology used during production and the plant’s location. To calculate process GHGs per unit of GDP for each sub-sector and/or unit of physical output, the Intergovernmental Panel on Climate Change (IPCC) presents three methodologies: Tier 1, Tier 2 and Tier 3; accuracy increases with each tier. Tier 1 and Tier 2 use an output-based methodology of multiplying production volumes with emission factors. Tier 1 uses the default emission factors of the IPCC (IPCC, 2006[36]); Tier 2 adjusts these emission factors to country-specific values. Tier 3 is an input-based methodology that calculates emissions based on carbon inputs; this is a more demanding task, as it requires a material flow analysis of the entire production supply chain. Total GHGs per unit of physical output per industry helps gauge whether its emission intensity is decreasing over time. If possible, breaking emissions into non-CO2 and CO2 gases per unit of output would be valuable, given their different lifetime in the atmosphere (for further explanation, see Box 1.2, Chapter 1).

Emissions from energy use can be calculated based on electricity use per unit of GDP for each sub-sector and/or physical output (percentage of end-use energy if fossil-fuel data are lacking), multiplied by the emission intensity of electricity production (e.g. tonnes of GHG per kWh).

Even in isolation, the electricity use per unit of GDP for each sub-sector and/or physical output is useful to identify industries that are vulnerable to loss of competitiveness owing to increasing electricity prices. This indicator is also useful for utilities to balance demand response: it allows groups to collect information on existing prevention and control techniques, while integrating variable renewables into the grid and the electrification of end uses (e.g. heating and transport). Any significant imbalance between consumption and generation could cause grid instability or severe voltage fluctuations, and failures within the grid, affecting well-being (discussed in detail in Chapter 2 on electricity). The electricity consumption of heavy industry is already a useful tool to help balance consumption and generation needs.

Table 3.5 summarises the key points from the section above: the well-being priority for heavy industry, its corresponding SDGs, and the well-being dimension and domain (introduced in Table 3.3), as well as the indicators attached to these frameworks.

A future well-being priority for heavy industry will be to maintain a healthy and safe environment in order to protect human health and biodiversity. This aligns with SDG 3.9, “substantially reduce the number of deaths and illnesses from pollution and contamination”; SDG 15.5, “take action to reduce the degradation of natural habitats, SDG 9.2; “promote inclusive and sustainable industrialization”; and SDG 12.4, “ensure the environmentally sound management of chemicals”. The indicators for these SDGs and the OECD well-being framework measure mortality rates from pollutants or the absolute levels of pollution. Their drawback is the missing link to heavy industry – e.g. how much pollution comes from which facilities? This section proposes additional indicators to assess the quality of heavy industry facilities and their contribution pollution.

Best available techniques (BATs, not to be confused with best available technology) are a key tool to prevent and control pollution from industrial facilities. A growing number of governments use BATs to establish legally binding emission-limit values in industrial permits for emissions to air, water and soil. BATs are state-of-the-art techniques for emission prevention and control, developed at a scale that allows implementation under technically and economically viable conditions. A BAT-based approach to environmental permitting for industrial installations allows setting conditions for environmental permits that are rooted in evidence and based on participatory decision-making, and are thus more likely to result in a high level of human health and environmental protection. To establish BATs, governments typically set up sector-specific technical working groups involving stakeholders from government, industry and environmental non-governmental organisations. The groups collect information on existing prevention and control techniques, and conduct a thorough assessment of these techniques according to environmental, economic and technical criteria. This process results in a set of BATs and associated emission levels (presented as a range), which are published in best available techniques reference documents (BREFs). The key information contained in the BREFs serves as a basis for setting emission-limit values and other permit conditions for individual industrial installations (OECD, 2018[37]), measuring the level of compliance with BAT-based emission-limit values for a given heavy industry by monitoring the percentage of industrial facilities that meet these values. The assumption is that these values are stringent. For further information, see Measuring the Effectiveness of BAT Policies (OECD, 2019[38]) and Best Available Techniques for Preventing and Controlling Industrial Pollution (OECD, 2018[37]).

The next set of indicators measures the pollutants from heavy industrial facilities. The EU Pollutant Release and Transfer Register (E-PRTR)2 is a front runner in this respect. Facilities must report annual data to a national repository; these data are then recorded in the E-PRTR. The database includes more than 30 000 industrial facilities, covering 65 economic activities within 9 industrial sectors: energy, production and processing of metals, mineral industry, chemical industry, waste and wastewater management, paper and wood production and processing, animal and vegetable products from the food and beverage sector, and other activities. The register tracks 91 pollutants released into air and water, including GHGs and other gases, heavy metals, pesticides, chlorinated organic substances, other organic substances and inorganic substances. This granularity helps identify particular facilities that harm the environment, and potentially harm human health and biodiversity indirectly, signalling where to target policies.

Table 3.6 summarises the key points from the section above: the well-being priority for heavy industry, its corresponding SDGs, and the well-being dimension and domain (introduced in Table 3.3), as well as the indicators attached to these frameworks.

Heavy industry is one among many users of the planet’s resources, including raw materials, land, water and energy. Indicators on heavy industry’s use of these resources is beneficial, on the one hand to avoid competition between sectors in a future of scarce resources, and on the other hand to improve the circularity and resource efficiency of production. These ambitions overlap with the goals of the existing frameworks and more specifically SDG 8.4, “improve progressively, through 2030, global resource efficiency in consumption and production”, and the OECD well-being framework’s “natural capital” domain under the “future resources” dimension. This subsection proceeds by resource to fill in gaps in these existing frameworks.

Part of the shift that needs to occur in heavy industry is to decrease the use of raw materials as much as possible (SDG 8.4) while recognising it will not drop to zero in the near future, and increase the use of secondary materials (SDG 12.5, “by 2030, substantially reduce waste generation through prevention, reduction, recycling and re-use”). Producing outputs based on secondary materials is less emission-intensive than outputs from primary materials, which explains why secondary materials are candidates for decarbonisation (as mentioned in the previous section). In addition, processing of secondary materials causes less pollution, maintaining environmental quality and creating two-way alignment across heavy industry’s priorities (OECD, 2019[1]). While the indicators adequately assess the use of raw materials and the production of waste, they do not capture the circularity of resources, or whether some waste could be repurposed.

A number of institutions, including UN Environment, the OECD and the European Union, collect data for indicators related to material flow analysis, expanding on the indicators already proposed for SDG 8.4. The aim is to describe the interaction of the domestic economy with the natural environment in terms of the flow of materials, seen as material “inputs” and “outputs”. The Global Material Flows Database3 produced by UN Environment’s International Resource Panel calculates a set of indicators, definitions and existing data. The exhaustiveness of these (readily available) datasets extends beyond domestic material consumption and the material footprint.

SDG 12.5.1 uses indicators for the national recycling rate and tonnes of material recycled. However, none of these indicators tell us whether waste is being re-used (a measure of circularity in the economy); what kinds of waste are being produced, and by whom (e.g. heavy industry); or whether some waste that is being disposed of could be re-used for other purposes.

A novel indicator used by the European Union is the circular material use (CMU) rate,4 which is “the share of material recovered and fed back into the economy – thus saving extraction of primary raw materials – in overall material use” on an annual basis. This indicator can be further disaggregated by material. It is especially valuable as it actually evaluates the capacity of the economy to re-use these materials and signals which heavy industries use more scrap (since there is disaggregation by material).

Indicators on the quantity and type of waste produced could help heavy industry and other industries minimise and re-use waste. A best practice is California's Department of Resources Recycling and Recovery (CalRecycle), which runs a database5 that expands on the SDGs by measuring disposal amounts (e.g. landfilled, imported, exported) over time at varying levels of disaggregation at the county, facility and jurisdiction levels. The database also measures exports of recycled materials and biomass.

In addition, CalRecycle tracks what exactly characterises the waste, to understand what various actors could re-use. The “Waste Characterisation Tool”6 helps jurisdictions understand the types and amounts of materials disposed in and diverted from California's waste stream. The State of California collects waste samples from three sources – residential, commercial/industrial and self-hauled – that would have been put in landfills. It then sorts samples into components in order to understand what is actually being thrown away. These data are then used to estimate the potential disposal and diversion rate (for recycling) by business group, material type and residence. Material types include several originating from heavy industry, including metal (e.g. aluminium, other ferrous, other non-ferrous); special waste (e.g. ash); inert and other (e.g. concrete, gypsum board, other wood waste); household hazardous waste (e.g. batteries, vehicle and equipment materials); glass; electronics; and paper. After characterising the waste streams, jurisdictions can try to divert this waste into secondary material usage for heavy industry, when applicable.

Heavy industry uses water, which could be problematic in a resource-scarce future. For example, during a drought in South Africa, local municipalities cut off industries from the local water supply, which prompted these facilities to pursue greater resource and explore alternative ways to access and re-use water. To ensure a steady supply of materials and fuels, and sustainably use the planet’s resources, indicators tracking water consumption can signal industries where greater resource efficiency is needed. This aligns with SDG 6.4, which aims by 2030, substantially increase water-use efficiency across sectors by tracking two indicators: change in water-use efficiency and percentage of freshwater withdrawal as a proportion of resources. A valuable indicator from the Food and Agricultural Organization is the value added divided by the water used (USD per m3) over time by ISIC code.7

In the future, sectors will increasingly compete for land: energy needs land for solar panels and wind farms, agriculture needs it to feed our rising populations and for bioenergy crops, and industry – and extractive industries – need it to produce materials and fuels. Extractive industries – which provide the inputs for heavy industry – use land for different purposes. A particular concern is their conversion of forested land (e.g. to extract Bauxite in Malaysia), since this exacerbates climate change by both the resulting direct emissions and by removing a carbon sink.

Table 3.7 summarises the key points from the section above: the well-being priority for heavy industry, its corresponding SDGs, and the well-being dimension and domain (introduced in Table 3.3), as well as the indicators attached to these frameworks.

This chapter argued that heavy industry should broaden its priorities to safeguard current and future well-being. It should maintain production while addressing the needs of workers and communities, limiting climate change, ensuring a healthy and safe environment, and sustainably managing the planet’s resources. To achieve these priorities, heavy industry needs to adopt net-zero, circular and resource-efficient production. The last section identified a number of indicators – e.g. the EAMFP and the CMU rate – that reflect these multiple priorities, in order to evaluate the synergies and trade-offs of different actions and strategies.

The foundation of any policy package for decarbonisation is the creation of transition plans, set within the context of wider economic transformations and progress towards a net-zero, resource-efficient and circular economy. Constructing these plans with relevant stakeholders from heavy industry builds on their expertise to identify feasible pathways towards this (Bataille et al., 2018[6]; Davis et al., 2018[7]). These transition plans can then direct needed policies and further investments to facilitate heavy industry’s transition. However, to do this, they must be built on a set of core policies, including carbon pricing; targeted research, design and development; and resource efficiency programmes. Finally, they must be accompanied by a number of enabling policies, including removal of market distortions and trade barriers, and better classifications of trade, BATs and waste management/reduction.


[6] Bataille, C. et al. (2018), “A review of technology and policy deep decarbonization pathway options for making energy-intensive industry production consistent with the Paris Agreement”, Journal of Cleaner Production, Vol. 187, pp. 960-973, https://doi.org/10.1016/J.JCLEPRO.2018.03.107.

[33] Cárdenas Rodríguez, M., I. Haščič and M. Souchier (2018), “Environmentally Adjusted Multifactor Productivity methodology and empirical results for oecd and G20 countries”, OECD Green Growth Papers, No. 2018/02, OECD, Paris, http://www.oecd.org/greengrowth (accessed on 13 March 2019).

[28] Chateau, J., R. Bibas and E. Lanzi (2018), “Impacts of Green Growth Policies on Labour Markets and Wage Income Distribution: A General Equilibrium Application to Climate and Energy Policies”, OECD Environment Working Papers, No. 137, OECD Publishing, Paris, https://dx.doi.org/10.1787/ea3696f4-en.

[7] Davis, S. et al. (2018), “Net-zero emissions energy systems.”, Science, Vol. 360/6396, https://doi.org/10.1126/science.aas9793.

[30] Ellen MacArthur Foundation (2017), Effective Industrial Symbiosis, https://www.ellenmacarthurfoundation.org/case-studies/effective-industrial-symbiosis.

[31] Empora, N. and T. Mamuneas (2011), The Effect of Emissions on U.S. State Total Factor Productivity Growth, https://ideas.repec.org/a/ren/journl/v3y2011i2p149-172.html.

[26] Energy Transitions Commission (2018), Reaching zero carbon emissions from Steel, Energy Transitions Commission, http://www.energy-transitions.org/sites/default/files/ETC_MissionPossible_FullReport.pdf.

[22] Farooq, M., F. Anwar and U. Rashid (2008), “Appraisal of heavy metal contents in different vegetables grown in the vicinity of an industrial area”, Pakistan Journal Botany, Vol. 40/5, pp. 2099-2106, https://s3.amazonaws.com/academia.edu.documents/30910209/PJB40%285%292099.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1550489659&Signature=lE5KHGez%2FwCDRul0tMRrLShA1dc%3D&response-content-disposition=inline%3B%20filename%3DAppraisal_of_heavy_metal_co (accessed on 18 February 2019).

[23] Giam, X., J. Olden and D. Simberloff (2018), “Impact of coal mining on stream biodiversity in the US and its regulatory implications”, Nature Sustainability, Vol. 1/4, pp. 176-183, https://doi.org/10.1038/s41893-018-0048-6.

[19] Gunatilake SK (2015), Methods of Removing Heavy Metals from Industrial Wastewater, http://www.jmess.org (accessed on 18 February 2019).

[3] Hoesly, R. et al. (2018), “Historical (1750–2014) anthropogenic emissions of reactive gases and aerosols from the Community Emissions Data System (CEDS)”, Geoscientific Model Development, Vol. 11/1, pp. 369-408, https://doi.org/10.5194/gmd-11-369-2018.

[2] IEA (2019), World Energy Balances Database, https://www.iea.org/statistics/balances/.

[29] ILO (2018), World Employment and Social Outlook 2018 – Greening with jobs, ILO , Geneva, http://www.ilo.org/publns. (accessed on 13 March 2019).

[15] ILO (2015), Guidelines for a just transition towards environmentally sustainable economies and societies for all, http://www.ilo.org/publns (accessed on 8 August 2019).

[5] IPCC (2018), Global Warming of 1.5°C. An IPCC Special Report on the impacts of global warming of 1.5°C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty, IPCC, https://report.ipcc.ch/sr15/pdf/sr15_spm_final.pdf.

[36] IPCC (2006), Emission Factor Database, IPCC, https://www.ipcc-nggip.iges.or.jp/EFDB/find_ef.php?ipcc_code=2&ipcc_level=0.

[32] Kalaitzidakis, P., T. Mamuneas and T. Stengos (2018), “Greenhouse Emissions and Productivity Growth”, Journal of Risk and Financial Management, Vol. 11/3, p. 38, https://doi.org/10.3390/jrfm11030038.

[13] Kherat, S. (2019), “The Future of AI in Heavy Industry – Agriculture, Construction, Mining, and Beyond”, https://emerj.com/ai-podcast-interviews/future-ai-in-heavy-industry/.

[20] Leung, A. et al. (2008), “Heavy Metals Concentrations of Surface Dust from e-Waste Recycling and Its Human Health Implications in Southeast China”, Environment, Science and Technology, Vol. 42/7, https://doi.org/10.1021/es071873x.

[27] Luderer, G. et al. (2018), “Residual fossil CO2 emissions in 1.5–2 °C pathways”, Nature Climate Change, Vol. 8/7, pp. 626-633, https://doi.org/10.1038/s41558-018-0198-6.

[24] McCarthy, A., R. Dellink and R. Bibas (2018), “The Macroeconomics of the Circular Economy Transition: A Critical Review of Modelling Approaches”, OECD Environment Working Papers, No. 130, OECD Publishing, Paris, https://dx.doi.org/10.1787/af983f9a-en.

[11] Mercier, F. and D. Mabashi (2019), “Steel Market Developments”, STI Working Paper Series, OECD, https://doi.org/JT03447063.

[18] Mogren, C. and J. Trumble (2010), “The impacts of metals and metalloids on insect behavior”, Entomologia Experimentalis et Applicata, https://doi.org/10.1111/j.1570-7458.2010.00967.x.

[38] OECD (2019), Best Available Techniques (BAT) for Preventing and Controlling Industrial Pollution Activity 3: Measuring the Effectiveness of BAT Policies, OECD, https://www.oecd.org/chemicalsafety/risk-management/measuring-the-effectiveness-of-best-available-techniques-policies.pdf.

[1] OECD (2019), Global Material Resources Outlook to 2060: Economic Drivers and Environmental Consequences, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264307452-en.

[37] OECD (2018), Best Available Techniques for Preventing and Controlling Industrial Pollution, OECD, https://www.oecd.org/chemicalsafety/risk-management/approaches-to-establishing-best-available-techniques-around-the-world.pdf.

[16] OECD (2017), The Land-Water-Energy Nexus: Biophysical and Economic Consequences, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264279360-en.

[9] OECD/IEA (2019), Material efficiency in clean energy transitions, OECD/IEA, https://webstore.iea.org/material-efficiency-in-clean-energy-transitions.

[35] OECD/IEA (2007), Tracking Industrial Energy Efficiency and CO2 Emissions, International Energy Agency, Paris, https://www.iea.org/publications/freepublications/publication/tracking_emissions.pdf.

[14] Oxford Economics (2019), How robots change the world: what automation really means for jobs and productivity, Oxford Economics, http://resources.oxfordeconomics.com/how-robots-change-the-world.

[10] Silva, F. and G. Mattera (2018), “State enterprises in the steel sector”, OECD Science, Technology and Industry Policy Papers, No. No. 53, OECD, Paris, https://www.oecd-ilibrary.org/docserver/2a8ad9cd-en.pdf?expires=1556613534&id=id&accname=guest&checksum=723964F9DAF5593087A0F61553ADD591 (accessed on 30 April 2019).

[17] Tang, W. et al. (2014), “Heavy metal accumulation by periphyton is related to eutrophication in the Hai River Basin, Northern China.”, PloS one, Vol. 9/1, https://doi.org/10.1371/journal.pone.0086458.

[34] US Cluster Mapping (2019), Iron and Steel Mills and Forging Subcluster, https://clustermapping.us/cluster/upstream_metal_manufacturing/subclusters/upstream_metal_manufacturing_iron_and_steel_mills_and_forging (accessed on 21 May 2019).

[25] van Ewijk, S. (2018), “An Introduction to Resource Efficiency: Concepts and Definitions”, in Flachenecker, F. and J. Rentschler (eds.), Investing in Resource Efficiency, Springer International Publishing, https://doi.org/10.007/978-3-319-78867-8.

[4] Wolff, N. et al. (2018), “Impacts of tropical deforestation on local temperature and human well-being perceptions”, Global Environmental Change, Vol. 52, pp. 181-189, https://doi.org/10.1016/j.gloenvcha.2018.07.004.

[12] World Steel Association (2019), “Working in the steel industry”, World Steel Association, https://www.worldsteel.org/en/dam/jcr:597ab555-3b9e-4177-bcd3-604e4d6e5a2b/fact_Employment_2019.pdf.

[8] Wyns, T., I. Robson and G. Khnadekar (2018), A Bridge Towards a Carbon Neutral Europe, Institute for European Studies, https://www.ies.be/files/Industrial_Value_Chain_25sept.pdf.

[21] Zheng, N. et al. (2010), “Health risk assessment of heavy metal exposure to street dust in the zinc smelting district, Northeast of China”, Science of The Total Environment, Vol. 408/4, pp. 726-733, https://doi.org/10.1016/J.SCITOTENV.2009.10.075.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.

© OECD 2019

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.