Tracking Industry 2020

More efforts needed
Tracking industry
In this report

Direct industrial CO2 emissions, including process emissions, declined 0.6% to 8.5 GtCO2 in 2018 (24% of global emissions), similar to the trend of relatively flat emissions in the past several years. The modest decline occurred largely in non-energy-intensive industries. To align with the SDS, industry emissions must fall by 1.2% annually to 7.4 GtCO2 by 2030 – despite expected industrial production growth. Greater energy efficiency, the uptake of renewable fuels, and research and deployment of low-carbon process routes including CCS are all critical. Governments can accelerate progress by providing innovation funding and adopting mandatory CO2 emissions reduction and energy efficiency policies.

Industry direct CO2 emissions in the Sustainable Development Scenario, 2000-2030

Tracking progress

Demand for industrial products has risen considerably in the past two decades, along with energy consumption and CO2 emissions.

Some modest improvements have been made in industrial productivity (value added per unit of energy consumed) and in renewable energy uptake, and some positive policy and innovation steps have also been taken. Nonetheless, progress is far too slow. Accelerated efforts on all fronts will be needed to get industry on track with the Sustainable Development Scenario (SDS).

The industry sector accounted for 37% (157 EJ) of total global final energy use in 2018 (including energy use for blast furnaces and coke ovens1  and feedstocks2). This represents a 0.9% annual increase in energy consumption since 2010, with 0.8% growth in 2018, following stronger growth of 1.6% the previous year.

Growth in energy consumption has been driven largely by an ongoing long-term trend of rising production in energy-intensive industry subsectors (i.e. chemicals, iron and steel, cement, pulp and paper and aluminium).

The highest rate of industrial energy consumption growth in 2010-18 occurred in India and the ASEAN countries (over 4% annual growth), while China had the largest absolute increase (accounting for 45% of the total net increase). Meanwhile, industrial energy use declined slightly in Europe and the Americas.

The industry sector’s energy mix has remained relatively unchanged overall since 2010. While solar thermal and geothermal final energy use expanded the most quickly – more than doubling from 2010 to 2018 – they accounted for less than 0.05% of total final industrial energy use in 2018. The fossil fuel share of the energy mix decreased from 73% to 69%, while electricity rose from 18% to 21%, largely owing to increasing electricity use in non-energy-intensive industry.

In the SDS, growth in energy use needs to be limited to about 0.3% per year to 2030, despite expected growth in industrial production.

Energy mix changes – particularly a shift away from coal and towards natural gas, bioenergy and electricity – contributes to a fall in the direct CO2 emissions intensity of industrial energy use. While solar thermal and geothermal energy continue to expand, they cannot yet provide high-temperature heat on a large scale, and therefore are unable to replace a significant portion of industrial process heat.

Innovation is needed to expand the potential to use renewables and electricity (both directly and indirectly via hydrogen) in a greater portion of industrial processes, particularly for high-temperature heat.

Final energy consumption and fuel shares in the Sustainable Development Scenario, 2010-2030


Industrial energy productivity (industrial value added per unit of energy used) has risen in most regions since 2000. Key contributors to the increase are the deployment of state-of-the-art technologies, operational adjustments leading to more efficient equipment use, and a structural shift away from energy-intensive industry (e.g. steel and cement) and towards a larger share of added value from higher-value-added sectors (e.g. automotive manufacturing, food and beverages, and textiles).

Historically, the greatest improvements in energy productivity have been in developed countries, which tend to focus on higher-value industrial products, while countries in which industrialisation is more recent have shown relatively little progress.

Middle Eastern industrial productivity has declined as a result of strong development in energy-intensive manufacturing subsectors (particularly cement) between 2004 and 2010, which offset the deployment of best available technologies (BATs) in several expanding industries.

In China, industrial productivity changed very little or even fell between 2000 and 2006, but it has since risen. Improvements resulted from China beginning to diversify industrial activities away from energy-intensive steel and cement production and towards high-value industries such as machinery and chemicals. Implementation of mandatory energy efficiency policies (the Top 1 000 and Top 10 000 programmes) also helped.

Energy productivity is closely connected with energy efficiency. In 2018, industrial energy efficiency investments totalled less than USD 40 billion. Although yearly investments have been relatively constant since 2015, market composition has shifted: China represented 37% of the total in 2018 (up from 25% in 2015), whereas North America’s share dropped from 17% in 2015 to less than 10% in 2018.

At just over 45%, the heavy-industry share of total global industrial energy efficiency investment is lower than in 2015, when it was nearly 50%. This largely reflects the continuing slowdown in construction of new energy-intensive industrial facilities in China, which is the result of ongoing structural changes in the economies of China, Europe and North America.

In the Asia-Pacific region, India is an emerging source of industrial energy efficiency investment, with an increase of nearly 5%. Modernisation of industrial facilities, coupled with the strong government mandates of the Perform, Achieve, Trade (PAT) Scheme, stimulated higher levels of investment.

Despite these positive developments, this indicator is off track. Global industrial productivity needs to increase 2.8% per year to 2030 to get on track with the SDS – an acceleration from the 2.1% annual growth of 2010‑18.

Industry energy productivity by region in the Sustainable Development Scenario, 2000-2030


Renewable heat consumption increased 2% per year on average during 2010-18, with renewables meeting just under 10% (10 EJ) of industrial energy demand for heat in 2018. This falls short of the 3%-per-year increase needed to meet the SDS trajectory by 2030, when renewables satisfy 13% (15 EJ) of industrial heat demand.

The majority (90%) of renewable heat currently consumed in industry comes from bioenergy, equalling 8.5% of industry energy demand for heat.3 Most consumption occurs in industries that produce biomass wastes and residues on site, such as in the pulp and paper subsector and food and tobacco.

While bioenergy is used considerably in pulp and paper production (30% of energy use) and more modestly in cement (3% of energy use), its consumption is very limited in other energy-intensive industries.

To increase its use in other subsectors, biomass fuel supply chains that are competitive with fossil fuels need to be established. This can be challenging, however, because policy support for renewables in industry and carbon pricing mechanisms are not widespread. Furthermore, efforts to increase biomass use in industry should also take into account the competing demands for sustainable biomass for other critical end uses, for which finding other viable alternatives may be more difficult.

The second-largest share (nearly 10%) of renewable heat used in industry comes from renewable electricity. This share is expanding as renewables figure increasingly in national electricity generation portfolios and as more industrial processes become electrified. While renewable electricity for heat expands under the SDS, technical challenges and the high costs of using electricity directly for high-temperature heat are likely to limit its penetration.

The use of solar thermal energy for industrial processes has also been expanding, especially for processes that require low-temperature heat (below 100°C). Low-temperature needs occur in various industries such as textiles and food, for processes such as drying, bleaching, cooking and pasteurisation. In energy terms, however, the contribution of solar thermal remains very small and several barriers constrain its uptake: a lack of policy incentives, low awareness of its potential, and challenges integrating it with industrial energy demand.

Renewable heat in industry in the Sustainable Development Scenario, 2010-2030


Demand for materials is a major determinant of total energy consumption and CO2 emissions in industry subsectors. Material demand has historically been linked closely with both population and economic development: as economies develop, urbanise, consume more goods and build up their infrastructure, material demand per capita tends to increase considerably. Once industrialised, an economy’s material demand may level off and perhaps even begin to decline.

Decoupling material demand from economic and population growth can help curb growth in energy consumption and CO2 emissions from material production. In the past couple of decades, the increase in global demand for key energy-intensive bulk materials has exceeded population growth – and for many materials, GDP growth. Growth since 2000 has been particularly high, largely driven by rapid economic development in China.

Estimates suggest, however, that global demand levelled off in the past few years for cement, while GDP and population continue to grow. This levelling-off is largely the result of saturated material demand in China.

While it may be a first step towards decoupling global material demand from economic and population growth, strong growth in other emerging economies may once again drive up material demand in the coming years. Indeed, there is evidence that levelling off can be temporary: after several years of stagnation, demand for steel has begun to rise again.

Reducing material demand growth through ambitious pursuit of material efficiency strategies can contribute to emissions reductions. In the SDS in 2030, demand for steel is 5% lower than in the Stated Policies Scenario, while cement demand is 9% lower and aluminium demand is 9% less.

Opportunities for material efficiency exist at every stage of any supply value chain. In the SDS, use-phase reductions, including by extending building lifetimes through repair and refurbishment and reducing vehicle demand largely through mode-shifting, make the largest contribution (approximately 50%) to the combined reduction in demand for steel, cement and aluminium in 2030.

Product design and fabrication strategies, including vehicle lightweighting and improved building design, also make a significant contribution (about 45%). Higher metal manufacturing yields at the material production stage, as well as end-of-life reuse, contribute the remaining reductions. Additionally, rather than by reducing final material demand, increased end-of-life recycling can reduce emissions by enabling greater uptake of lower-emitting secondary production methods.

Demand for materials, 1990-2018


In 2018, mandatory policy-driven energy efficiency targets and standards covered less than 25% of total industrial energy use in most regions, with no major increases in coverage relative to the previous year. While a number of countries have minimum energy performance standards for electric motors, few have mandatory overall performance targets for industrial firms and sectors. 

China and India are among the strongest performers on policy coverage, as they put mandatory targets in place for energy savings in industry sectors several years ago already, which still apply today. In China, the 100, 1 000, 10 000 Programme has been included as part of the 13th Five-Year Plan (2016‑20), and it supersedes the previous Top 10 000 Programme, a component of its 12th Five-Year Plan (2011‑15). In India, the PAT Scheme began in 2012; it is now being implemented on a rolling basis with new installations being added each year, so cycles III, IV and V are now all under way.

To get on track with the SDS, it will be important to extend mandatory policy coverage to a larger portion of industrial energy use, and in more regions. Just as important is to ensure that the strength of requirements of both new and existing policies is ambitious enough.

To achieve high enough emissions reductions, policies need to cover not only energy efficiency and process optimisation, but other factors that influence industrial emissions also, such as process emissions and technological shifts. Policies targeting overall CO2 emissions reductions (e.g. a multi-sector or economy-wide emissions trading system) are therefore important.

China, for example, launched its emissions trading platform in December 2017. The technical and legal infrastructure has been established over the past two years, and the first real spot trading is expected to begin in 2020. The initial phase will cover only the power sector, but there are plans to eventually include several industry subsectors at an unspecified future date. The Chinese administration is requiring emissions reporting from industries in 2020, likely to establish better data collection for their eventual inclusion in the scheme.

In the European Union, an emissions trading system covering industry has been in place since 2005. For much of the first decade, an overabundance of allowances led to low prices, but they have begun to rise recently, with average annual prices of EUR 5.8 per tonne in 2017, EUR 15.5 per tonne in 2018, and an estimated EUR 22 per tonne in.

Additionally, for industry subsectors deemed less trade-exposed, the free allocation of allowances has been reduced from 80% in 2013 to near 30% in 2020. Highly trade-exposed industries continue to receive free allowances for emissions equivalent to production at a benchmark emissions intensity.

It appears the EU emissions trading system may be starting to have a small impact on emissions from industrial installations, with an average decrease of 0.7% registered from 2017 to 2018; this is nonetheless considerably below the 5.9% average decrease in power installation emissions.

Due to concerns about the impact of the emissions trading system on industrial competitiveness, in late 2019 the EU announced it would develop a proposal for a carbon border adjustment mechanism for certain industrial sectors by 2021, although it is expected to receive some resistance. The revised rules for phase 4 (2021-30) also include an increase in emissions cuts, with allowances declining at an annual rate of 2.2% compared with the current 1.74%, and reinforced use of a Market Stability Reserve to reduce and prevent emissions allowance surpluses.


Note: This map is 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. Source: Based on analysis from IEA (2019b).

Voluntary energy efficiency policies also exist in many regions: for instance, the number of ISO 50001 certifications for industrial energy management systems reached at least 18 000 in 2018.4 While this represents some progress, it appears the world is not on track to achieve the Clean Energy Ministerial Energy Management Campaign’s target of 50 001 industrial operations with ISO 50001 certification by 2020. Furthermore, about 75% of the certificates issued so far have been in Europe, which accounted for just 12% of global final industrial energy use in 2018, so uptake needs to accelerate in other regions.

Other energy management system standards may have higher uptake in specific regions. For example, in 2016‑17, the number of certifications under China’s GB/T 23331 standard increased by 25% (from 2 036 to 2 552). Improved data collection on these various standards, including on the resulting energy and emissions reductions, would be useful to better analyse their impact.

Two main approaches are being pursued to develop innovative low-carbon industrial processes: directly avoiding CO2 emissions by relying on renewable electricity (directly or through electrolytic hydrogen), bioenergy or alternative raw materials; and reducing CO2 emissions by minimising process energy, using fossil fuels but integrating CCUS. Finding value-enhancing uses for industrial by-products is another area of innovation, in which synergies are sought among various industrial activities, including through CCUS.

A number of key innovation efforts are under way around the world, including the following (see subsector pages for additional examples):

  • In February 2019, the European Commission announced EUR 10 billion in funding for the demonstration of low-carbon technologies. The Innovation Fund, initially proposed in 2015 and largely funded by revenue from the EU ETS, will support large-scale demonstrations of low-carbon technologies and processes in energy-intensive industries, CCUS, renewable energy and energy storage. The first call for proposals will be in 2020, with subsequent calls until 2030.
  • Mission Innovation is a global initiative of 23 countries and the European Commission to accelerate global clean energy innovation. Four of the seven Innovation Challenges launched in 2016 are relevant to the industry sector: carbon capture, clean energy materials, sustainable biofuels and converting sunlight. In 2018, an eighth Innovation Challenge was launched on renewable and clean hydrogen, also relevant to industry.
  • In the iron and steel sector, the HISarna project has successfully demonstrated steel production using an enhanced smelt reduction technology, which has the possibility of incorporating CCS, at a pilot plant in the Netherlands. Plans are under way to develop a second large-scale pilot plant in India, which could open during 2025-30. Meanwhile, the HYBRIT project is constructing a pilot plant in Sweden to investigate steel production through hydrogen-based direct reduced iron production, with start-up scheduled for summer 2020.
  • Multiple projects are under way around the world to test CCUS applications in the cement sector, using at least five different capture technologies. For example, the CLEANKER project’s pre-commercial demonstration plant in Italy applying calcium looping is expected to begin operation in 2020; the CO2ment project in Canada launched trials of a novel physical adsorption technology in 2019; and Dalmia Cement announced in 2019 it will undertake large-scale demonstration of chemical absorption capture at its plant in Tamil Nadu, India.
  • Related to chemical production, a large-scale demonstration plant to produce ammonia with hydrogen produced from solar power in Australia is expected to begin operation in 2021. A number of projects in China are demonstrating CCUS in high-value chemical production. Innovation is also under way on improved plastic recycling.
  • In the aluminium industry, progress has been made in recent years towards commercialising inert anodes, which – unlike conventional carbon anodes used in aluminium smelting – do not degrade and do not release CO2 as process emissions. In 2019, construction began on the Elysis research centre in Canada (a joint venture between Alcoa and Rio Tinto), with the aim of bringing inert anodes to market by 2024. In Russia, RUSAL is now producing aluminium with inert anodes at an industrial scale and is targeting mass-scale production by 2023.

While innovation efforts of recent years are promising, accelerated action will be needed to develop and deploy technologies for medium- to long-term CO2 emissions reductions in industry. Most of the new low-carbon processes in industry that will be key to long-term emissions reduction in the SDS are on track to become commercially available by 2030-35. Ensuring these milestones are achieved, or possibly even reaching them ahead of schedule, will be critical to put the industry sector on a net-zero-emissions trajectory. 

Deployment of BATs can help improve industry energy efficiency and should be pursued when economical, keeping in mind the longer-term need to transition to breakthrough near-zero-emissions technologies. Adopting waste heat/gas recovery and cogeneration technologies could be expanded in subsectors such as iron and steel and pulp and paper.

Furthermore, ensuring efficient equipment operations and maintenance will help guarantee optimal energy performance. This can be reinforced by implementing energy management systems. Developing plant-level action plans and sharing best practices can also help improve energy efficiency, while governments can accelerate the process by adopting energy efficiency targets and regulations.

Shifting increasingly to secondary production methods – i.e. using recycled inputs – will be important to improve energy efficiency and reduce CO2 emissions in metals, chemical products (including plastics) and paper manufacturing.

Governments and industries can work together to improve collection avenues for recycled products and increase co‑operation among stakeholders involved in the production and end-of-life stages of the value chain. Government-mandated recycling requirements, waste disposal fees, recycled content requirements and extended producer responsibility can also help increase recycling.

Reducing overall demand through material efficiency strategies at all stages of the value chain can avoid CO2 emissions from industrial production. Industry can help by considering lifecycle emissions when designing products and construction projects; by reducing waste during manufacturing and construction; and by developing sharing- and circular-economy-based business models.

Policies that favour durability and the refurbishing of buildings over demolition will be pivotal to reduce demand for bulk materials. Governments can also encourage material efficiency by moving from use-phase to life-cycle-based CO2 emissions regulations, and from prescriptive to performance-based design standards.

Industries should also take advantage of opportunities for industrial symbiosis – including using the waste or by-products from one process to produce another product of value – to help close the material loop, reduce energy use and reduce emissions in the case of carbon capture and utilisation.

Examples include using steel blast-furnace slag in cement production, carbon from steel waste gases to produce chemicals and fuels, and waste from other industries as alternative fuels for cement production. Industrial symbiosis can also involve sharing energy utilities, infrastructure and services. Policy support can facilitate these endeavours.

The share of renewables in industry can be increased by several means: first, by ensuring that biomass waste and residues are used at as high an efficiency as possible by industries that have access to them. Examples include shifting to higher-efficiency co‑generation technologies in the pulp and paper and sugar and ethanol industries.

The cement industry offers scope to utilise greater amounts of low-value biomass residues and MSW to offset thermal demand currently met by coal. MSW consumption especially can be encouraged by increasing refuse-derived fuel availability through best-practice waste management – i.e. highly efficient waste sorting and collection, combined with mechanisms that place a cost on landfill disposal (e.g. landfill taxation).

With increasing shares of renewables in national electricity generation portfolios, the electrification of industrial processes (both directly and indirectly via hydrogen), when possible, can also raise renewable energy consumption. In countries with high amounts of direct irradiation, ESCO business models could boost solar thermal use in industry.

Decarbonising current industrial processes is challenging for a number of reasons. For example, emissions resulting from chemical reactions during industrial processes (process emissions) cannot be mitigated by greater energy efficiency and fuel switching alone. The high-temperature heat required in many industrial processes makes it difficult to switch completely from fossil fuel-based energy to low-carbon electricity and fuels.

Innovation over the next decade will therefore be critical to develop and reduce the costs of industrial processes and technologies that could enable substantial emissions cuts post-2030, including, for example, hydrogen-based production methods and CCUS.

Increased support for RD&D is needed from governments and financial investors, particularly to advance the large-scale demonstration and deployment of technologies that have already shown promise. Private-public partnerships can help, as can green public procurement, contracts for difference, and near-zero-emission materials quotas, which can generate early demand and enable producers to gain experience and bring down costs. Government co‑ordination of stakeholder efforts can also direct focus to priority areas and avoid overlap.

It will also be important to begin planning and developing infrastructure for eventual deployment of innovative processes, such as CCUS pipeline networks to transport CO2 for use or storage, and electricity transmission grids and near-zero-emission electricity generation to enable low-carbon hydrogen production. Gaining social acceptance for building this infrastructure, particularly COtransport and storage facilities, and ensuring affordable access to infrastructure and energy inputs will also be necessary.

Governments can promote CO2 emissions reduction efforts by adopting mandatory CO2 emissions reduction policies, such as a gradually increasing carbon price or tradeable industry performance standards that require average CO2 intensity for production of each key material to decline across the economy and permit regulated entities to trade compliance credits.

Adopting these policies at lower stringencies in the short term (i.e. within the next three to five years) would provide an early market signal, enabling industry to prepare and adapt as stringency increases over time. It could also help reduce the costs of low-carbon production methods, softening the impact on materials prices in the long term. Complementary measures may be useful in the short to medium term, such as differentiated market requirements – i.e. a government-mandated minimum proportion of low-emission materials in targeted products.

Ideally, mandatory policies would be applied globally at similar levels of ambition. Since many industrial products are widely traded, if the strength of efforts differs regionally, measures will be needed to help ensure a level global playing field – e.g. border carbon adjustments or free allocation of allowances for emissions below a targeted benchmark in an ETS. Another option may be downstream carbon pricing or regulations on the lifecycle emissions of end products rather than on material production – for instance, regulating vehicle manufacturing plants to reduce vehicle lifecycle emissions could raise the competitiveness of lower-emissions steel and aluminium.

Governments can extend the reach of their efforts by partaking in multilateral forums to facilitate low-carbon technology transfer and to encourage other countries to also adopt mandatory CO2 emissions policies.

Improving collection, transparency and accessibility of energy performance and CO2 emissions statistics on industry would facilitate research, regulatory and monitoring efforts. Industry participation and government co‑ordination are both important to improve data collection and reporting.

Governments also need to clarify avenues for greater data sharing in a way that will not put industries at risk of breaching compliance with competition laws.


External reviewers: Andrew Purvis (World Steel Association), Asa Ekdahl (World Steel Association), Florian Ausfelder (Dechema), Henk Reimink (World Steel Association), Markus Steinhäusler (voestalpine), Nicola Rega (Cepi)

  1. Listed in the transformation and energy industry own-use sections in the IEA energy balance.

  2. Listed under non-energy use in the IEA energy balance.

  3. Including bioenergy used for district heating.

  4. The ISO gauges the number of certifications using a voluntary survey of certification bodies, so this estimate should be regarded as conservative. Note that the ISO changed the methodology of accounting for sites in 2018.

Next Chemicals