Environment at a Glance Indicators

Green transition

Measuring the green transition

Achieving decent living standards for all, improving well-being and enabling sustainable development worldwide, depends on countries’ ability to transition to an economy that uses fewer resources and generates less pollution, remaining within planetary boundaries (Rockström, 2009[1]). The green transition supports both social and economic development, by ensuring that natural assets continue to provide essential resources and ecosystem services.

The green transition requires both effective public policies and strong private sector engagement to stimulate and scale-up investment and innovation aligned with long-run sustainability objectives. Progress towards greener economies demands coordinated action across countries and within ministries, including those responsible for the economy, industry, finance, trade, and agriculture. Tracking this progress calls for a clear conceptual framework and a comprehensive and robust set of indicators.

This chapter emphasises that the pursuit of indefinite economic growth is incompatible with the planet’s finite limits. More specifically, a green transition requires new business models, sustainable consumption and production patterns, and economic and governance systems that respect the Earth’s boundaries to promote sustainable development on a global scale.

The indicators presented here apply an environmental lens to long-term drivers of productivity, competitiveness, inclusiveness, and resilience. They quantify the ‘greening’ of labour markets, education systems, consumption and production patterns, investment flows, financial markets, productivity, and public policy. As a set, these indicators provide an integrated picture of environment-economic trends, enabling the assessment of national progress in the green transition. The indicator set presented here builds on the OECD Green Growth Indicators (2011[2]), (2014[3]) and (2017[4]) (see Reader’s guide).

The chapter covers 50 countries, including OECD members, OECD Key Partners and OECD accession candidates, over the past three decades. Designed as an online publication, it sources in real time from the latest available data, updated regularly, and features dynamic data visualisation tools. It integrates recent OECD work on methodological advances, notably in the areas of employment, finance and investment. Existing indicators have been refined and enriched, including those on taxes and subsidies, technological innovation and environmentally adjusted multifactor productivity growth. Greater emphasis is placed on new indicators aligned with the System of Environmental-Economic Accounting (SEEA), enabling more granular analysis and stronger integration of environmental and economic data. Lastly, the chapter outlines priorities for developing new indicators to address key measurement gaps.

Conceptual approach

The term “green transition” is broadly defined as the economy-wide shift away from activities with high environmental impacts and natural resource dependence, toward practices that respect planetary boundaries and promote long-term sustainability, resilience, and societal well‑being globally. This definition, aligned with the OECD Green Growth Strategy, acknowledges the limits of resource-intensive growth in a world of approaching ecological tipping points. Yet, translating this concept into measurable indicators of progress presents significant challenges.

First, the green transition is a process, rather than a fixed end point. It cuts across both economic and environmental domains, making it inherently difficult to isolate and quantify. Many aspects of the transition are latent, observable only through indirect indicators or proxies. These proxies, ranging from emission intensities to resource efficiency or productivity metrics, must be interpreted with caution, as they may not fully capture all underlying structural changes, may be influenced by unrelated factors, or may depend on the socio-economic context of a country.

Second, the green transition is multifaceted. It encompasses interrelated environmental, economic and social dimensions, each evolving at different scales and influenced by various drivers, making it difficult to capture overall progress through a single indicator. In addition, both environmental pressures and socio-economic structures often exhibit inertia, resulting in time lags between policy interventions, behavioural changes, and observable outcomes. Effective measurement therefore requires analysis of both short-term and long-term effects, supported by a coherent set of indicators that reflect the multiple dimensions of the green transition.

Activities, products and services are often classified as green (clean, sustainable, or environmentally friendly) or brown (dirty, unsustainable or environmentally harmful) by assessing their environmental impact. However, these labels can be misleading. What is referred to as “green” is not an absolute category but rather a relative concept, typically dependent on time and place, indicating practices or technologies that are less harmful than conventional alternatives. In reality, all economic activity has environmental impacts. For instance, some green technology may still exert harmful effects if deployed at scale, potentially overdrawing planetary boundaries. The goal of the green transition is to reduce and maintain the cumulative impacts of human activity within the Earth’s ecological limits. Greener technologies, activities, and lifestyles should be understood as incremental steps towards reducing environmental harm, rather than absolute solutions. For these reasons, this chapter uses terms such as environmentally related or environmentally relevant when referring to specific, well-defined indicators. In contrast, simplified labels such as “green” or “environmental”, which can be ambiguous and misleading when not grounded in measurable criteria, are used as broad terms to imply environmental benefit.

Different analytical approaches can be used to quantify the green and brown components of the economy. Top-down approaches use macroeconomic modelling and econometric techniques to infer broad trends at the sectoral or national levels. In contrast, bottom-up approaches rely on detailed activity-based classifications and apply consistent criteria to tag products, services, or investments as environmentally sustainable or not. Each approach comes with advantages and limitations: top-down methods capture system-wide interactions but may rely on assumptions that overlook sectoral detail, while bottom-up methods offer granularity and transparency but often face data availability or consistency challenges. Hybrid approaches are increasingly used to balance the strengths of both.

Tracking the green transition using granular and transparent bottom-up approaches requires harmonised data based on internationally agreed classifications. Without such taxonomies, it is difficult to track and compare the scale and pace of progress across countries and over time. Classifications must also evolve regularly to reflect new technologies, products, and scientific knowledge on environmental impacts. Otherwise, outdated categories can lead to inaccurate or misleading measurements.

Progress towards sustainability is rarely linear or unidirectional. Transition implies continuous movement rather than fixed endpoints, and there are no universally agreed targets that define a “fully green” economy. Compounding this, many conventional economic models assume easy substitution of natural capital, a premise that is increasingly implausible, given the degradation of natural assets and the limited substitutability of essential ecosystem services, and ultimately unrealistic in a finite world. As scientific evidence about tipping points and systemic risks grows, it is urgent to integrate environmental limits and resource constraints explicitly into economic analysis.

Adding to the complexity, much of nature’s contribution to the economy remains unmeasured or underestimated. Ecosystem services, biodiversity loss, and climate feedback loops are often omitted from national accounts, despite their substantial economic and environmental significance. As data gaps persist, decision-makers must make decisions based on incomplete, but improving information, while being guided by precautionary and transparency principles.

Finally, green transition indicators must be interpreted within a broader socio-economic context. Environmental sustainability cannot be pursued in isolation from issues like income inequality, energy poverty, inflation, unemployment, and demographic pressures. The Green Growth Indicators database (OECD, n.d.[5]) includes a comprehensive set of socio-economic indicators that complement this chapter, including GDP structure, energy price volatility, tax systems, education, population dynamics, and trust in institutions. Together, they provide essential context for assessing the feasibility and fairness of transition policies.

Indicator grouping

The green transition is not limited to an environmental shift, it represents a broader economic and societal transformation. To reflect this multidimensional perspective, the chapter groups indicators into five sections:

• Employment and education

• Markets of goods and services

• Finance and investment

• Multifactor productivity

• Policies and regulations

The green transition reshapes employment and education systems. Labour markets must adapt as demand grows for green occupations and new skill sets. This section tracks shifts in employment structures, workforce preparedness, and academic systems. It highlights both opportunities and challenges in equipping human capital with the capabilities needed for a low-carbon, resource-efficient economy.

In markets of goods and services, indicators assess the diffusion of green technologies as enablers of the transition, the economic relevance of environmentally significant sectors, both green and brown, and track trends in trade patterns, production shifts and the emergence of green businesses. Understanding these developments helps identify how markets are adjusting in response to changing consumer preferences, policy incentives, and biophysical limits.

Finance and investment indicators shed light on the alignment of capital flows with environmental goals. This includes access to green finance, the development of sustainable infrastructure, and the growing role of green financial instruments such as green bonds. This section also explores how financial markets contribute to systemic change, and the importance of risk management in the face of environmental uncertainties.

The section on multifactor productivity captures how efficiently economies combine labour, capital, and natural resources in the production process. As environmental pressures intensify, improving resource-use efficiency and decoupling growth from environmental degradation becomes essential. Tracking changes in productivity growth, especially when adjusted for environmental inputs and outputs, helps assess progress toward more sustainable economic performance.

Policies and regulations are the institutional backbone of the green transition. This section examines the use and design of environmental policies and regulatory frameworks. It also considers broader governance aspects, including policy stringency, coherence across sectors and institutional capacities for implementation.

This indicator grouping mirrors the sequence of change in a green transition. It begins with people: the skills, jobs, and knowledge that form the human capital base. It then moves to markets of goods and services, where sustainable production and consumption take shape. Finance and investment follow, as the flow of capital enables low-carbon biodiversity-friendly infrastructure and technologies. Multifactor productivity captures how efficiently economies combine people, capital and natural resources in this new context. Finally, public policy accelerates the process, providing the incentives and enforcement needed to sustain the transformation.

Indicators in this chapter complement those focused on natural capital (e.g., materials, energy, land, water) and environmental quality (e.g., air pollution, climate-related hazards) in other chapters of Environment at a Glance Indicators. Taken together, these chapters support a more comprehensive and coherent assessment of how countries are transitioning to a greener economy. By integrating environmental concerns with key structural components of economic performance included in the green transition chapter, the OECD Environment at a Glance Indicators help policymakers identify trade-offs, synergies, and opportunities for action, supporting effective decision-making across priority areas of the transition.

Employment and education

Key messages

• Green employment is rising across OECD countries but remains uneven and limited. Green jobs are concentrated in industrialised and high-skilled regions. Promoting a just green transition will require addressing inequalities in access to these jobs, including gaps in mobility and education, through inclusive training, reskilling and targeted support programmes.

• Jobs in GHG-intensive sectors are declining but only gradually. In some countries, carbon-intensive sectors still account for a large share of employment. Stronger government action is needed to better align labour markets with environmental goals, reduce structural mismatches and support worker mobility.

• Compulsory instruction time in natural sciences varies widely across OECD countries, with some allocating less than half the time of others. Increasing early exposure to environmental and natural sciences can strengthen environmental literacy and support long-term development of green skills.

• Environmentally related scientific output is rising globally. However, relative specialisation and share of global output in OECD countries are declining, especially compared to Key Partner countries. Greater support for environmental research, particularly in larger OECD economies, is needed to sustain knowledge generation for the green transition.

Policy context

Adapting labour markets is key to a successful green transition. A well-managed shift supports job retention and mobility while enabling green skills development. It involves retraining workers from declining sectors and preparing them for emerging roles. This transition draws parallels with previous structural changes in labour markets driven by globalisation and digitalisation, which also required rapid adjustments in skills and employment structures across OECD countries. Education and training play a central role is this shift, both in supporting workforce transitions and in equipping future generations with the knowledge, values, and competencies needed for a resilient, environmentally sustainable economy.

Indicators on green and brown jobs offer a useful lens for analysing labour market dynamics during the green transition. While the definitions and measurement approaches for tracking green and brown jobs vary and the terms are sometimes used rhetorically (OECD, 2024[6]), these can still serve as a basis for evidence-base policy development. Current indicators can help highlight where workforce adjustments are taking place and identify gaps, such as sectors that may require greater policy attention or support.

Data to assess the green transition of education remain limited. One measurable aspect is the variation in compulsory instruction time for natural sciences (including physics, chemistry, biology, environmental sciences, and ecology) at primary and secondary levels across OECD countries (OECD, 2024[7]). Scientific literacy developed through these subjects is critical for understanding and engaging with human-environment interactions and for fostering evidence-based decision-making (Nusche, Fuster Rabella and Lauterbach, 2024[8]; White et al., 2023[9]). Research indicates a positive link between strong performance in science and higher environmental awareness, unlike achievement in other domains (OECD, 2023[10]). While instruction time in natural sciences captures only part of the competencies required for the green transition, it offers a quantifiable starting point. Broader measurement of green education would benefit from more targeted indicators assessing students’ knowledge, attitudes, and engagement with environmental issues.

Advancing the green transition requires new and transformative ideas across industries and societal systems, with scientific research being a foundational driver of this process. Science is also fundamental for advancing our understanding of Earth systems and the interdependence between humans and nature. Scientific knowledge guides evidence-based policies and informs shared goals. The direction and intensity of scientific production are shaped by policy priorities, public engagement, and funding allocations (OECD, 2023[11]). Consequently, the level of scientific activity in disciplines like environmental sciences, has undoubtable bearing on a country’s progress towards a green economy. Bibliometric indicators enable the tracking of scientific activity in environmentally related fields, providing a key measure of a country’s innovative capacity and its potential for sustainable living standards.

To more fully understand labour market transitions, these indicators should be viewed in conjunction with the structure of the economy, demographic and migration patterns, and broader environmental, trade and industrial policy setting.

Main trends and recent developments

OECD countries have seen a gradual rise in green jobs and a slow decline in GHG-intensive jobs

Between 2011 and 2022, the share of workers in green-task jobs increased across OECD countries, indicating gradual progress in adapting labour markets to the green transition. However, the average OECD share remains modest (under 18% of total jobs), and well below top performers such as Luxembourg, Lithuania and Norway, where green-task jobs exceed 25%. Increases were observed in nearly all countries, though at varied magnitude. Progress was strongest in countries with innovation-driven economies and high service-sector employment. OECD analysis at the subnational level confirms this trend and highlight sector-specific challenges. Green-task jobs are often concentrated in urban and high-skilled regions (OECD, 2023[12]) and are more likely to be held by high-educated men. Evidence also suggests that low-skill green occupations offer lower pay and poorer working conditions than other low-skill occupations (OECD, 2024[6]), reflecting labour market imbalances that hinder a just transition.

In contrast, jobs in GHG-intensive industries declined in OECD countries over the past decade. In 2022, GHG-intensive jobs accounted for 20% of total employment in the OECD on average, down from over 21% in 2011. The decline is marginal but consistent, reflecting overall decarbonisation trends. Nevertheless, some countries remain heavily reliant on GHG-intensive employment, particularly in Slovak Republic and Czechia, where such jobs exceed 30% of total employment, compared to under 10% in Luxembourg. At the country level, the pace of decline varies, and reductions in GHG-intensive jobs are not always accompanied by a corresponding increase in green-task jobs. This suggests the need for broader structural changes and stronger enabling policies to facilitate a just transition.

Primary and secondary school students receive little and unequal class time in natural sciences

Compulsory instruction time in natural sciences varies significantly across countries, both in duration and as a share of total curriculum time. A persistent cross-country gap in instruction time is evident across OECD countries. For example, primary level instruction time ranges from more than 6.7 full-day equivalents per year (161 hours) in Costa Rica to less than 0.6 days (14 hours) in Hungary. Costa Rica far exceeds other countries, both in absolute time and in share of total instruction time (14%). At the other end, Germany, Bulgaria, and Hungary allocate less than 4% of total instruction time to natural sciences. Since 2011, notable increases in natural sciences instruction have occurred in Greece and Korea. In contrast, Türkiye, Spain, Japan, and Hungary all saw declines exceeding 13 hours.

At lower-secondary level, countries generally allocate more time to natural sciences instruction. In 2023, Estonia, Korea, and Denmark allocated more than 6.5 full-day equivalents (roughly 160 hours or more), while Iceland, Luxembourg, and Bulgaria allocated close to three days. The share of natural sciences in total instruction time averages between 9% and 15% but varies widely – from over 20% in Estonia and Korea to below 10% in Luxembourg, Norway, and Iceland. Since 2011, Korea and Portugal reported meaningful increases, while Finland Poland, Hungary, and Iceland saw declines.

Most countries have intensified their research in environmental sciences

Across OECD and Key Partner countries, the share of scientific output in environmentally related fields has increased steadily since 2010. Growth in environmental science publications has outpaced total research output, particularly in Key Partner economies. However, OECD countries remain less specialised in these fields than the global average, and trail Key Partners in publication shares. Within OECD countries, there is marked heterogeneity. Countries such as Latvia, Costa Rica, Lithuania, and Iceland record high shares of publications in Environmental Sciences. In contrast, Israel, Japan, and the United States show relatively low shares, well below both OECD and global averages.

In absolute terms, environmentally related scientific output has increased across both OECD and Key Partner countries since 2009. However, since 2021, growth has diverged: publication volumes have declined in OECD countries, reflecting stagnating or declining research activity in the United States and the European Union, while continuing to rise among Key Partner countries, largely driven by the People’s Republic of China (hereafter “China”) and, to a lesser extent, India. Overall, this shift reflects a broader trend of increasing environmentally related research output from outside OECD countries. Nonetheless, at the per capita level, smaller OECD countries such as Norway, Iceland, and New Zealand stand out for the highest research intensity in environmental fields. In contrast, several others such as Colombia, Mexico or Türkiye remain below the OECD average both in volume and in per capita output rates. Research on Renewable Energy, Sustainability, and the Environment has expanded its share within the energy research portfolio. Iceland, Norway, and New Zealand are the most research-intensive in this area, with over 30% of energy-related publications focused on sustainability topics.

Indicators

One third of 15-year-old students in OECD countries have achieved environmental sustainability competence

Environment and climate literacy underpins long-term support for the green transition and strengthens acceptance of environmental policy. At COP29, the OECD announced that climate literacy will become a permanent component of the OECD Programme for International Student Assessment (PISA) framework starting in 2029. Climate literacy fosters evidence-based decision-making, critical thinking, and active engagement for both individual and collective action (OECD, 2024[13]).

The 2018 cycle of PISA included an assessment of student’s foundational levels in four environmental sustainability competence areas:

• Environmental awareness: understanding environmental issues and climate change, underpinned by basic science knowledge.

• Caring: personal importance placed on environmental protection.

• Science self-efficacy: perceived ability to complete tasks related to climate change.

• Environmental behaviour: engagement in energy-saving and collective action to protect the environment.

Results varied widely across countries. In 2018, less than half of 15-year-old students in most countries demonstrated foundational competence in all four areas. The highest share was observed in Korea (50%), while rates in the lowest-performing countries, such as the Slovak Republic and Romania, fell below 20%. These differences may reflect variation in national curricula, science education quality, and educational systems’ emphasis on environmental awareness.

Comparability and interpretation

The indicators presented provide only a partial view of green employment and should be interpreted with caution. Green-task jobs are identified by linking occupations to environmental objectives using task-level data from the United States. Occupations are classified as green when at least 10% of their tasks are environmentally related. While this approach provides a useful starting point, it does not capture differences in the intensity, innovativeness or evolution of green tasks over time. It also may not fully reflect variation across countries, changes in job content or indirect roles (e.g., security guard at a solar PV company). Applying international crosswalks between occupation classifications may also introduce inconsistencies. This work is part of ongoing efforts at the OECD to refine measurement approaches and better reflect the complexity of green jobs content across labour markets (e.g., (OECD, 2024[6])). Finally, note that the label “green” for these indicators follows source data terminology, but should be understood as a relative, context-dependent concept rather than an absolute measure of environmental benefit (see Measuring the green transition for discussion).

GHG-intensive employment is defined as the number of workers in high-emitting industries, regardless of their specific role. This includes administrative and support functions, as well as green-oriented positions within these industries, but excludes environmentally harmful work in low-emitting industries. While useful, this measure abstracts from important factors such as differences in firm-level practices, environmental standards, and regulatory compliance. For example, a country with strict regulation may operate high-emission sectors more sustainably than others but this nuance is not yet captured. Additionally, the indicator does not currently consider other environmental pressures beyond GHGs, such as air and water pollution or biodiversity loss. This approach remains under development and is being improved as part of the OECD’s broader work to enhance the environmental relevance and comparability of employment indicators.

For education, instruction time dedicated to the natural sciences offers a tangible but narrow proxy. It captures intended instruction time as established in public regulations, and not the actual number of hours that students receive nor the learning outside of formal classroom settings. Instruction time on natural sciences omits broader dimensions such as sustainability education, systems thinking, problem-solving skills, and human-environment relations (OECD, 2023[10]; Jeong, Sherman and Tippins, 2021[14]; Zoller, 2012[15]). Broader societal efforts to promote green education, such as investments in teacher training and curriculum development, remain unaccounted. Further, the indicators are restricted to primary and lower-secondary education levels, leaving out higher education’s role in green skills development (OECD, 2024[16]). Cross-country comparability is limited by differences in curriculum structures, content, and pedagogical approaches. More targeted indicators, such as measures of students’ knowledge, attitudes, and engagement with environmental issues, are needed to capture green education more broadly. The forthcoming inclusion of climate literacy assessments in PISA (OECD, 2025[17]) will provide more refined and comparable measures of education systems’ preparedness to the green transition.

Bibliometric indicators link scientific publication output to authors’ country affiliations and field classifications. While publication counts do not reflect research quality or impact, tracking the relative specialisation in environmental fields signals the generation of knowledge important in support of the green transition. Coverage limitations include underrepresentation of interdisciplinary work, non-English journals, and socio-economic research relevant for the green transition. Similarly, the classification system does not easily distinguish between scientific research contributing to green solutions and that supporting environmentally harmful activities. To improve interpretation, bibliometric indicators should be complemented with information on academia and research funding structures.

While foundational education, scientific output, and employment patterns are critical, green skills within the workforce remain a major measurement gap. Many competencies needed for the green transition are not yet systematically measured. Anticipating these skill needs can reduce labour market mismatches and mitigate the social cost of the transition (OECD, 2023[10]). Despite their recognised importance, no internationally harmonised dataset on green skills currently offers representative and comparable measures at the country level.

Markets of goods and services

Key messages

• Environmentally related innovation surged over two decades, but global momentum has slowed since 2010 and declined further after 2019, with growth in renewable energy patenting slowing sharply since 2015. Innovation activity remains highly concentrated in a few countries and domains, with climate change mitigation energy-related technologies leading, and adaptation, carbon capture, and ocean-related technologies underdeveloped.

• Green trademark registrations have risen steadily, with OECD Key Partner countries driving global growth. Their expansion signals wider diffusion of environmentally related technologies and an increasing market orientation towards environmental goods and services.

• Trade in environmentally related goods has grown across OECD and Key Partner countries, led by renewable energy technologies and equipment. Stronger environmental policy support and trade facilitation can enhance both market access and the diffusion of cleaner technologies.

• Green start-ups make up a small share of entrepreneurial activity across the OECD, with limited growth over time. However, their access to venture capital has expanded significantly, indicating rising investor interest in markets of environmentally related goods and services. Despite progress, financing remains uneven, and more inclusive entrepreneurial ecosystems are needed to accelerate market transformation.

Policy context

Innovation plays an important role in the transition to a greener economy, delivering both environmental benefits and competitive market advantages. The development and diffusion of environmentally related technologies are critical to the cost-effective attainment of global environmental goals. By providing protection and exclusivity, patents encourage investment in research and technology development. They also disclose valuable technical information, enabling more inventions and directing research towards emerging fields. However, exclusive patent rights may increase costs, and limit access for customers who cannot pay the price premium. Patent design, such as duration and breadth of rights (i.e., level of differentiation needed to avoid infringement), can influence the balance between protection and accessibility. Patent data on green technologies are a measurable proxy for innovation dynamics in the green transition, providing early indicators on innovation trends and market opportunities.

Green trademarks play an important role in consumer markets by signalling the environmental characteristics of products, services, and firms. They serve as indicators of the commercialisation of environment- and climate-related innovation by large-scale corporations and new firms. Indicators on green trademarks reveal which countries have positioned themselves to seize market opportunities linked to more environmentally sustainable consumption and production. By analysing green trademark registrations, policymakers can assess technological advancements and long-term strategies to shape market competitiveness.

Achieving rapid diffusion and adoption of green technologies, critical for delivering global public goods in climate and environment domains, relies on well-functioning markets and effective trade. In this context, indicators on trade in environmentally related goods (EGs) help track the greening of markets and allow assessing how policies influence domestic and international demand. Evidence suggests that higher environmental policy stringency correlates with increased trade of EGs because, for instance, firms and households may need to adopt new equipment and techniques to meet environmental standards. Stricter policies can also induce economies of scale and avoid transportation costs, which can lead to an increase in the size of the domestic market and higher net exports of EGs (Garsous, 2019_[25]). Understanding how trade contributes to diffusing cleaner technology and improving market access for greener products and services, allows for more informed economic and environmental policymaking.

While indicators on trademarks and international trade provide valuable insights, they often reflect the activities of large incumbent firms, dominant products, or national specialisations. In contrast, indicators on green start-ups offer a complementary perspective, as start-ups are frequently the source of radical innovations that challenge established practices. Start-ups can introduce breakthrough technologies, novel business models, and organisational change – particularly in emission- and resource-intensive sectors – helping to reshape production and consumption patterns and accelerate the transition to a greener economy. As demand grows for solutions such as smart grid technology, eco-design and repairability of products, and digital services for the sharing economy, indicators on start-ups show how entrepreneurial activity and venture capital are mobilised to deliver environmentally related goods and services.

Main trends and recent developments

Environmental innovation surged over two decades, but global momentum is fading.

Global patenting activity in environmentally related technologies has surged over two decades, but global momentum is fading. ICT applications for mitigation and ocean-related innovations led the rapid growth, but with a recent decline observed across all domains. In terms of technological domains, climate change mitigation (CCM) technologies related to energy account for the largest share of environmentally related patenting. These are followed by technologies focused on environmental management. Innovation in climate change adaptation, sustainable ocean economy, climate-related ICT, and capture, storage, sequestration and disposal of greenhouse gases, represent smaller shares.

OECD countries are losing ground in environmentally related technological innovation.

In two decades, the relative specialization on green inventions slowed in two thirds of Member countries. When adjusted for population, environmentally related inventions remain low in most countries. Denmark and Korea lead in per capita terms, but overall patenting activity remains limited, with few countries exceeding 10% of environmentally related inventions in their total patent output.

After two decades of dominance, renewable energy innovation is now nearing the pace of fossil fuel technology developments.

Renewable energy generation remains the most active subdomain within CCM technologies. However, growth has slowed since 2020 and declined further in 2022, bringing patenting trends close to those of fossil fuel energy supply technologies. Fossil fuel inventions have remained relatively stable since 2010. In 2022, for the first time in over a decade, patenting activity in renewables and fossil fuels converged. At the country level, innovation in renewable energy technologies is highly concentrated. Denmark, Korea, and Norway record the highest numbers of renewable energy patents per capita, while fossil fuel-related inventions are primarily concentrated in Europe, led by Norway, Luxembourg, and the Netherlands.

Green trademark activity is rising globally, but levels vary widely

Since the early 2000s, green trademarks have gained significant traction globally, reflecting growing demand for environmental branding and market protection for environment- and climate-related products and activities. OECD Key Partner countries have led this expansion, outpacing the OECD average since 2003. By 2022, green trademarks accounted for over 6% of all trademarks, signalling improved diffusion of environmental and climate-related technologies across the global economy. Despite global progress, since 2015, OECD countries have exhibited slower growth compared to the world average. At the national level, Slovenia and Norway lead all countries, with green trademarks filings representing 40.1% and 14.9% respectively in the 2020-22 period. Green trademark activity has increased significantly across most countries over the past two decades, though the magnitude and pace of growth vary widely. The largest gains in the intensity of green trademark filing were recorded in Norway, Lithuania, Hungary, and Austria, where the share of green trademarks has increased more than eight percentage points since the early 2000s. Among Key Partner countries, China is the main driver of growth, supported by strong advancements in environment- and climate-relevant technologies.

OECD countries sustain strong trade of EG as Key Partner imports fall

Trade in environmentally related goods (EGs) has grown steadily over the past two decades. Between 2002 and 2008, both OECD and Key Partners countries increased EGs exports and imports as a share of GDP, with Key Partners peaking at over 2.5% for exports. Following the 2007-08 financial crisis, OECD countries have maintained relatively stable trade levels, reaching 2.4% of GDP for exports and 2.6% for imports by 2022. By contrast, Key Partner imports steadily declined, resulting in a trade surplus in EGs, suggesting rising domestic production capacity and increased competitiveness in green technologies.

Renewable energy technologies dominate EG trade

In 2022, renewable energy equipment made up the largest share of EG trade globally, accounting for over one-third of total EG exports and imports. Clean and resource-efficient products and wastewater- and water-related technologies also ranked among the top traded categories. At the country level, Hungary, Japan, the Slovak Republic, and Germany were top exporters of EGs, each exceeding 15% of total goods exports in 2022. For imports, Norway, Sweden, and Hungary led with EGs imports representing about 15% of goods imported. Among Key Partner countries, China had the highest shares of EG exports, driving the Key Partner average. Other countries in the group showed more limited participation, with smaller and more uneven shares of EG trade.

Funding for green start-ups is increasing faster than firm creation

Across the OECD, green start-ups have consistently represented around 4-7% of all start-ups since the early 2010s, with limited change in the share over time. However, their access to venture capital (VC) has improved. In 2021-22, green start-ups received approximately 8.5% of total VC financing, up sharply from less than 1% in 2000-01, and rebounding from earlier peaks seen in 2008 likely influenced by stimulus packages during the global financial crisis which would have also crowded-in private VC capital (Dechezleprêtre and Kelly, forthcoming[18]).

At the country-level, small economies lead in the relative share of green start-ups: Costa Rica, Slovenia, Norway, and Latvia recorded the highest proportions between 2020-22, all exceeding 18% of total start-ups. In absolute terms, larger countries such as the United States (122 start-ups in 2020-22), Germany (36), the United Kingdom (32), and account for the highest numbers. On the funding side, green start-ups in Greece, Norway, and Estonia attracted the largest VC shares in 2020-2022, each exceeding 20% of total national VCs. While several OECD countries show upward trends, access to VC financing for green start-ups remains uneven and limited in many economies, constraining their role in accelerating the green transition.

Indicators

Emissions are decoupled from value added and employment in the European Union

Integrating environmental data (e.g., on air emissions, material use, ecosystem extent and condition) based on the System of Environmental-Economic Accounting (SEEA) and economic data (e.g., value added, employment, or capital formation) based on the System of National Accounts (SNA), is crucial for understanding the emission and resource intensities of economic activities. Sector-level indicators derived from this integration offer valuable insights into the greening of markets for goods and services and enables policymakers to identify the sectors generating the highest environmental pressures relative to their economic contributions. They reveal where progress towards decoupling is being made, help guiding the design of effective environmental policy, and support the coordination of industrial and environmental strategies. In the absence of optimal economy-wide policies, focusing mitigation efforts on the most polluting industries can still mitigate environmental and social impacts of pollution.

In the European Union, environmental and economic accounts reveal a disconnect between sectors responsible for air emissions and those contributing economic output, capital formation, and employment. The four most carbon-intensive industries (i.e., manufacturing, electricity, transportation, agriculture, retail and construction) account for 85% of all CO₂ emissions. However, these same sectors generate just 25% of value added, 28% of capital formation and 24% of jobs. Methane emissions are heavily concentrated in the agriculture and waste management sectors, while fine particulates (PM_2.5), nitrogen oxides (NO_x), and ozone precursors are mainly emitted by the transportation and agriculture sectors. Fluorinated gases (NF₃ and SF₆) are heavily concentrated in the water and waste management sector.

Comparability and interpretation

Patent indicators offer a quantitative, and internationally comparable measure of technological output. They can be disaggregated with a highly detailed classification system – an important advantage when analysing environmentally related technologies. However, not all innovations are patented, and patent counts do not indicate their relative economic or environmental impact. Analytical techniques can help address some of these limitations (e.g. patent family size, relative technological advantage), but carefully interpretation is essential (OECD, 2009[19]).

Green trademarks capture firms’ intent to commercialise environmentally relevant goods and services, reflecting downstream innovation activities. They complement earlier-stage innovation indicators, like patents, by identifying actors positioning themselves to compete in key markets. However, trademarks do not measure actual market impact, technology adoption rates, or environmental outcomes. Moreover, not all trademarks correspond to impactful or scalable technologies. Current green trademark tagging focuses on climate change mitigation and adaptation technologies (e.g. electric cars, renewables, batteries, and hydrogen) which do not reflect the full range of technologies relevant for the environment. Improving the indicator would require more detailed and standardised classification frameworks of environmental activities. Finally, note that the label “green” for these indicators follows source data terminology, but should be understood as a relative, context-dependent concept rather than an absolute measure of environmental benefit (see Measuring the green transition for discussion).

There is currently no internationally agreed list of products to be considered “environmental”. In this section, environmentally related goods are identified based on the OECD Combined List of Environmental Goods (CLEG) by (Moïsé and Tresa, 2025[20]). The label “environmentally related” for these indicators deviates from source data terminology to highlight potential differences with the scope and lists of environmental goods and services aligned with the SEEA.The OECD CLEG enables consistent comparisons across countries and over time but faces some classification challenges. Harmonized System (HS) codes often combine products that have environmental and non-environmental purposes, making the precise identification of environmental relevance difficult. Moreover, many goods have multiple uses, which does not allow them to be classified as environmental along their whole life cycle. Further efforts are needed to extend the measurement of trade in goods to turnover, value added, and employment in line with the SEEA.

Data on green start-ups are sourced from the New Startup database (see Data sources), which sources data on green start-ups from venture capital-oriented private databases, capturing firms with external funding and media visibility needs. As a result, start-ups without external funding or located outside major innovation hubs may be underrepresented. The use of shares of green in total start-ups, partly mitigates comparability issues, as the bias likely affects green and non-green firms similarly. Classification of green start-ups relies on keyword-based tagging, subject to misclassification due to vague firm descriptions or evolving definitions of green activities (see Glossary). Cross-country comparability may also be influenced by differences in venture capital ecosystem maturity and the database coverage. Finally, note that the label “green” for these indicators follows source data terminology, but should be understood as a relative, context-dependent concept rather than an absolute measure of environmental benefit (see Measuring the green transition for discussion).

Current indicators capture selected aspects of markets and sectors. Limited information is available on non-technological market innovation, such as changes in business models, work patterns, and managerial and organisational innovations more broadly. Yet, these types of innovation play a key role in enabling the uptake and effective use of new technologies and are instrumental for the green transition. Additionally, an important data gap is the quantification of green services. Many economies have shifted towards service-driven structures, making the absence of robust service-sector indicators a key limitation. Harmonised definitions, detailed classifications, and integrated data frameworks are needed to improve the scope, comparability, and policy relevance of green transition indicators for markets of goods and services.

Finance and investment

Key messages

• Global investment in low-carbon energy reached nearly USD 2 trillion in 2024, equivalent to 6.8% of total investment, outpacing fossil fuel investment for the fourth consecutive year. Low-carbon energy investment is rising across countries, particularly in renewables, energy efficiency, grids and storage. However, some countries still allocate more than half of energy investment to fossil fuels, signalling the need for targeted divestment support.

• Green investment can be financed through a range of financial instruments and asset classes. Available estimates indicate that finance contributing to climate change mitigation goals remains low.

• Green bond issuance continues to expand globally, reaching close to USD 600 billion in 2023, with European countries leading in issuance relative to GDP. Despite the importance and variety of green financial instruments, green bonds remain the only instrument with sufficiently detailed and comparable data. To scale up the use of green financial instruments, stronger alignment of taxonomies, greater transparency, and enhanced financial infrastructure are needed.

• Environmentally related ODA has increased steadily, now representing around 50% of total bilateral sector-allocable aid. However, ODA represents a small part of cross-border financial flows. Besides public finance flows, further efforts are needed to assess the alignment of private finance flows with environmental objectives.

Policy context

Finance and investment flows are major drivers of the green transition. Finance refers to the mobilisation and allocation of capital, while investment involves the actual deployment of this capital into physical or intangible assets (OECD, 2024[21]). Capital is essential to scale up clean technologies, upgrade infrastructure, and foster innovation. Finance and investment indicators inform policymakers about progress in reallocating resources towards low-carbon, biodiversity-friendly, resource-efficient, and resilient economic activities. Both domestic and international public policies play a key role in shaping finance, managing risks, and creating incentives to channel finance flows and stimulate public and private investment in green activities.

Real economy investment in energy reflects spending on the physical assets and technology critical for decarbonisation, energy security, and affordability. Investment flows indicators inform on distribution of capital across supply chains, including fuels and electricity generation, grids, storage, critical minerals, and R&D. Tracking these flows helps identify priority areas for private and public investment in clean energy. Public co-financing, long-run policy predictability, and risk mitigation are crucial to mobilise private capital and accelerate the deployment of low-emission energy solutions.

The green transition is required across the financial system as whole, from real-economy investment to financial assets, institutions, and jurisdictions. Yet, detailed information on all layers of finance is scarce. Green-labelled bonds are currently the only instrument with comparable cross-country data. Since the 2007-08 financial crisis, green bonds have become a dedicated instrument for financing environmentally friendly projects. These bonds differ from conventional bonds through their explicit commitment to green assets or activities. They support the green transition by attracting new investors, reducing financing costs, enabling long-term financing by addressing maturity mismatches, and enhancing issuers’ credibility. Green bonds also facilitate the integration of environmental objectives into financial markets and help channelling capital from responsible long-term investors into sectors undergoing deep transformation to reduce their environmental footprint (OECD, 2017[22]).

Official development assistance (ODA) differs conceptually from other indicators, as it captures cross-border concessional public finance flows, particularly for developing countries. Yet, ODA provides the most detailed and comparable information currently available on international public green finance. ODA tracks concessional finance targeting environmentally related sectors such as water supply or renewable energy, and environmental objectives through the Rio markers covering climate change mitigation, adaptation, biodiversity, and desertification (see Glossary). ODA also supports the monitoring of progress towards the USD 100 billion climate finance objective (OECD, 2025[23]). More details are provided in the Climate change chapter of Environment at a Glance Indicators (OECD, 2025[24]).

Main trends and recent developments

Private sector drives surge in low-carbon energy investment, surpassing fossil fuels spending

In 2024, global investment in low-carbon energy technologies and infrastructure reached nearly USD 2 trillion, outpacing fossil fuel investment for the fourth consecutive year. The shift is primarily driven by private investment, which now represents about two-thirds of total energy investment globally. This underscores a gradual but steady transition in the private sector towards greener energy systems.

Energy investment shifting towards clean sectors and technologies across regions

IEA estimates for 2025, show that OECD and Key Partner countries have nearly doubled their low-carbon energy investment in a decade, reaching close to USD 1 trillion and USD 0.7 trillion respectively. This growth is largely due to rising investment in energy efficiency technologies and infrastructure, as well as power grid and storage systems. The increase has been primarily driven by advances in energy-efficient buildings and the growing adoption of electric vehicles (IEA, 2025[25]). Over the same period, OECD countries are also doubling their investment in grid infrastructure and storage systems, with North American countries accounting for more than half of this increase. The expansion of global investments in grid storage was supported by increased deployment of renewable energy technologies, and new policies implemented across Europe, the United States, China, and Latin America, to balance renewable energy supply and demand (IEA, 2025[25]).

Renewable energy technology and infrastructure investment is growing in most countries

Solar and wind energy investment continues to increase around the world. In countries such as South Africa, Costa Rica, Chile, and Japan, renewables are expected to exceed 40% of total energy investment by 2025. South Africa leads globally, allocating 75% of its energy investment to renewables. In Japan, the share remained unchanged compared 2015, as a divestment in solar PV was offset by higher investment in wind energy. In absolute terms, investment in solar PV in OECD countries has nearly doubled over the past decade, yet projected 2025 levels remain significantly lower than in China. These trends of global investment in renewables reflect the strong policy support and falling technology costs in energy generation technologies and grid infrastructure (IEA, 2025[25]).

Fossil fuel investment is declining but remains much higher than renewable energy

Investment in fossil fuel technology and infrastructure has declined in both OECD and Key Partner countries. In many economies, the share of fossil fuel in total energy investment has dropped below 10%. European countries are among those with the lowest shares. Despite the decline, fossil fuels still attract significant capital in some regions. Argentina, Mexico, Canada, Indonesia, and North African countries continue to dedicate more than half of their total energy investment to fossil fuel technology and infrastructure.

The European countries and the corporate sector dominate the issuance of green bonds

The global issuance of green bonds has grown substantially over the past decade, reaching close to USD 600 billion in 2023, more than triple the 2017 level. The corporate sector remains the dominant issuer, accounting for most of the global volume. Issuance by national governments (sovereign), development banks and government-backed entities has also expanded but to a lesser extent. When scaled to economic output, European countries lead global issuance. In 2024, the decade-long cumulative volume of green bonds issued in Luxembourg was over 19% in GDP-equivalent, followed by Sweden, Netherlands and Norway, all above 10%. The OECD average stood at approximately 3%. Beyond the European Economic Area, Chile and New Zealand are the only countries recording cumulative green bond issuance well above the OECD average expressed in GDP-equivalent. Among Key Partner countries, China leads in issuance but remains at 2%, while other Key Partners show very low issuance over the decade, indicating untapped potential in attracting environmentally responsible long-term investors.

Official Development Assistance for environmental objectives reaches record levels.

Official Development Assistance (ODA) targeting environmental objectives has increased significantly over the past two decades. In 2023, environmentally related ODA from OECD-DAC donors reached about 50% of total bilateral sector-allocable ODA, up from 37% in 2010 and around 25% in early 2000s. Climate change mitigation remains the largest thematic focus, accounting for USD 41 billion, followed by climate change adaptation (USD 31 billion). By contrast, biodiversity- and desertification-related objectives received only a combined total of USD 13 billion from OECD-DAC countries. Looking at ODA commitments by country, in 2023, the Netherlands, New Zealand, Korea, Iceland and Japan contributed more than two-thirds of bilateral sector-allocable ODA to environmentally related activities, the highest shares across all donor countries. Korea and Japan allocated more than half of their ODA to climate change mitigation activities. Netherlands and New Zealand also stand out for high shares directed to climate change adaptation (about 70%), while Norway, New Zealand and Iceland show the highest shares for biodiversity-related ODA, though barely reaching 25% of total ODA in 2023.

Increasing ODA directed at water, environment, and renewable energy sectors.

ODA flows to environmentally related sectors accounted for over 10% of the total ODA in OECD-DAC countries in 2023, with funding directed to areas such as: water and sanitation, the environment sector, and renewable energy. At the country-level, there is substantial variation. Israel, New Zealand, and Iceland led in overall support to these sectors, with Lithuania, New Zealand, Iceland and Germany leading in renewable energy, each, dedicating over 6% of their total sector-allocable ODA. However, despite the growing environmental focus of ODA, few countries meet the international target to allocate 0.7% of Gross National Income (GNI) to ODA. Only Norway, Luxembourg, Sweden, and Denmark, met the target both in 2000 and 2024, highlighting a persistent gap between environmental spending and broader development finance commitments.

Indicators

Accelerating the alignment of financial flows towards Paris Agreement goals

Article 2.1c of the Paris Agreement calls for finance flows to be consistent with a pathway towards low-carbon and climate-resilient development. This involves scaling up finance for activities contributing to climate goals and transition activities and redirecting finance away from activities undermining climate mitigation and resilience goals.

Assessing progress requires tracking all financial flows and stocks across all layers of the financial system: real-economy investment, financial asset classes, financial institutions, and financial jurisdictions. Although data remain partial and fragmented, available evidence from the OECD Review on Aligning Finance with Climate Goals highlights a low degree of alignment with climate mitigation goals:

• Real-economy investment: In 2022, low-carbon energy outpaced fossil fuel investment. However, it remains only a small share when compared with total gross fixed capital formation.

• Financial asset classes: The share of assets supporting low-carbon activities remains low – below 5% across all categories. For asset classes with available data, fossil fuel financing is higher than low-carbon.

• Financial institutions: In 2022, bank-facilitated financing for fossil fuels represented 0.55% of total assets, compared to 0.38% for low-carbon energy.

• Financial jurisdictions: Data remain scarce. Methodological differences in scope, definitions, and information disclosure policies, limit comparability and transparency across countries.

Low-carbon finance falls short across the financial system

Note: There are no available estimates on the share of fossil fuel activities in total private corporate equity or total sovereign bonds issuance. The first panel shows real-economy energy investment expressed as a share gross fixed capital formation.

Source: OECD (2024), OECD Review on Aligning Finance with Climate Goals: Assessing Progress to Net Zero and Preventing Greenwashing, Green Finance and Investment, OECD Publishing, Paris,

Comparability and interpretation

Energy investment data varies across the energy sector due to differences in the types of expenditure included. These data are also sensitive to market conditions, technology uptake and consumer behaviour. Beyond real-economy investment in energy, comparability is further constrained by inconsistent reporting frameworks and limited data availability. As accounting methods vary, investment trends should be interpreted with caution as they are not always directly comparable.

There is currently no internationally agreed methodology for comprehensively tracking green and brown finance flows and stocks across financial instruments and asset classes. This is illustrated in the classification of green bonds. While the Climate Bonds Initiative data provides some degree of harmonisation, it also presents some limitations. Inclusion thresholds are difficult to set, and they necessarily exclude some instruments that are partially green. Sector-specific taxonomies vary widely across jurisdictions and available data captures issuance volume but lacks information on investor composition, secondary markets or environmental impacts. Furthermore, differences in financial market development affect comparability. While green or climate bonds are the most prominent green finance instruments, others such as blue bonds, target-linked bonds, green private equity, and green real estate investment trusts (on energy-efficient buildings or sustainable real estate development) are less consistently tracked or reported. Finally, note that the label “green” for these indicators follows source data terminology, but should be understood as a relative, context-dependent concept rather than an absolute measure of environmental benefit (see Measuring the green transition for discussion).

Official Development Assistance (ODA) donors use the Rio markers to identify environmental objectives, as well as classify ODA targeting economic sectors. Rio markers reflect stated intent rather than actual environmental outcomes. Donor self-reporting may also introduce variability in marker application. Moreover, the reporting system for ODA, by definition, does not systematically capture private co-financing or non-concessional flows, limiting its scope for tracking total international green finance.

Existing indicators provide only a very partial coverage of green financial flows. Significant gaps remain in tracking private national and cross-border finance and investment. A “green” FDI-based indicator could help fill this gap. However, transparency challenges in the private sector and, consequently, of open data on finance and investment flows, continue to limit the measurement and comparability of green finance. Comprehensive reporting standards and an internationally harmonised taxonomy are needed to bridge the data gap.

Multifactor productivity

Key messages

• Environmentally adjusted multifactor productivity (EAMFP) is the main source of income growth in many countries. It highlights the central role of innovation, resource efficiency, and effective environmental management in supporting environmentally sustainable long-term growth.

• Natural capital contributes unevenly to income growth, many countries showing limited reliance on domestic resource extraction. This indicates structural shifts toward technology- and productivity-driven growth; but also exposes gaps in the valuation and integration of natural capital. Strengthening natural capital accounting is essential to capture the full economic value of ecosystems and environmental inputs.

• Non-renewable natural capital, particularly fossil fuels, remains a key income source in several countries, while the contribution of renewable resources is limited or declining in most. Some countries show offsetting trends between renewable and non‑renewable inputs’ reliance. However, these are limited, and structural shifts toward renewable natural capital as a source of income need to remain within ecological boundaries to be a sustainable source of growth.

• Most OECD countries show higher pollution-adjusted GDP growth (or total income growth accounting for emissions reductions), highlighting gains from cleaner and more resource-efficient production. In contrast, rising emissions in several Key Partner economies are reducing income growth, signaling the need for stronger efforts to achieve absolute decoupling.

Policy context

Environmentally Adjusted Multifactor Productivity (EAMFP) growth provides a comprehensive view of a country's potential for long-run economic growth. The EAMFP framework identifies sources of income growth and highlights trade-offs between economic growth and natural resource depletion and pollution. Unlike traditional productivity metrics, which typically exclude the degradation of natural capital, EAMFP offers a more accurate picture of economic performance in the context of the green transition.

Higher EAMFP growth signals that an economy is achieving income growth with lower economic and environmental costs, reflecting an improved use of resources. Conversely, lower values may indicate economic inefficiencies, overreliance on resource extraction, or unsustainable practices that threaten long-term growth. By identifying the drivers of economic performance, including the role of natural capital, the EAMFP framework enables countries to monitor progress, identify areas for improvement, and implement policies that advance the green transition of the economy.

The contribution of natural capital to pollution-adjusted GDP growth reflects how countries use natural resources to support economic development. This contribution depends on the proportion of natural capital rents and the extent of resource use. Positive contributions typically reflect increased reliance on natural resources, often at odds with sustainable practices, while negative contributions indicate a shift toward alternative growth factors. The valuation of natural capital within growth accounting frameworks helps capture the economic role of natural resources and ecosystem services. The divergence in contributions from non-renewable and renewable resources underscores a critical challenge for the green transition: the overreliance on commodified, finite resources. For countries heavily dependent on non-renewable inputs, this signals vulnerabilities and the urgent need to diversify income sources through more sustainable practices.

Accounting for pollution as part of macroeconomic output reveals that pollution abatement can enhance growth by reducing environmental externalities. When countries invest in pollution reduction, particularly in tandem with technological and institutional reforms, they not only improve environmental outcomes but also strengthen the capacity of the economy to generate income sustainably over the long term. These findings reinforce the case for integrating pollution abatement measurement and policies into productivity-enhancing strategies and national accounts.

Policies that enforce property rights for natural capital, price or regulate environmental externalities, and support green innovation are key drivers of environmentally adjusted productivity growth. By aligning fiscal and regulatory tools with environmental goals (such as carbon pricing, pollution standards, and support of greener activities), governments can foster more sustainable production systems, promote resource efficiency, and improve long-term economic resilience.

Indicators derived from the EAMFP measurement framework underscore the importance of incorporating non-renewable and renewable natural resources, ecosystem services, and pollution abatement into macroeconomic analysis and policymaking. The EAMFP framework builds on previous work (Brandt, Schreyer and Zipperer, 2014[26]), (Brandt, Schreyer and Zipperer, 2013[27]), (Cárdenas Rodríguez, Haščič and Souchier, 2018[28]) and (Cárdenas Rodríguez et al., 2023[29]) and isclosely linked with other indicators in this chapter (e.g., on technology, innovation and research, policies and regulations) and related topics across this webbook (e.g., indicators on natural resource stocks).

Main trends and recent developments

Pollution abatement lifts income growth in most countries

Pollution-adjusted GDP growth, which accounts for the cost of reducing pollution, provides a more accurate measure of income growth. A positive adjustment indicates lower emissions (e.g., due to economic restructuring, investment in cleaner technologies or effective climate policy), while a negative adjustment reflects rising pollution. Most OECD countries show a positive adjustment, with pollution-adjusted GDP growth exceeding standard GDP growth. Between 1996 and 2018, pollution abatement added an average of 0.11 percentage points (pp) annually to income growth in OECD countries. The adjustment is strongest in Belgium, the United Kingdom, and the Slovak Republic, where pollution abatement contributed around 0.38 pp per year in each country. In contrast, rising emissions in Key Partner economies reduced income growth by an average 0.5 pp. The most negative adjustments were observed in Türkiye, Indonesia, China, and India.

Environmentally adjusted multifactor productivity is the main source of income growth.

Decomposition of pollution-adjusted GDP, into the contributions of inputs, reveals that environmentally adjusted multifactor productivity (EAMFP) is the main source of income growth across most countries. The strongest contributions of EAMFP are observed in China, India, Latvia and Lithuania, where EAMFP was the source of more than 2.9 pp of income growth annually, indicating structural improvements and efficient use of inputs. In contrast, countries such as Brazil, Indonesia, Italy and Greece show weak EAMFP, with income growth largely driven by input accumulation. Low EAMFP growth is an indicator of pollution-intensive production, inefficient resource use, and lagging technological, institutional, and organisational change, compromising long-term growth potential.

Produced capital, or gross fixed capital formation, is a key driver of income growth in several OECD countries such as Türkiye and Costa Rica, and in Key Partner countries including China and India. Capital formation generally contributes more than labour or natural capital in China, Chile, and Romania. Labour plays a smaller role overall, with notable exceptions such as Luxembourg and Israel. A key distinction between OECD and Key Partner economies lies in the role of natural capital in generating income. In Key Partners, natural capital contributes on average 0.06 pp to income growth annually, significantly more than the OECD average of just 0.0003 pp. However, natural capital in Australia, Chile, China, and Brazil, plays a stronger role, adding over 0.11 pp annually.

Growth is still heavily dependent on non-renewable natural resources

Disaggregating natural capital into renewable and non-renewable components, reveals a high reliance on exhaustible resources. Overall, non-renewable natural capital contributes four times more to income growth than renewable natural capital. In Australia, Chile and Peru income growth relies heavily on exhaustible resources. In some countries (e.g., Peru, Mexico, Argentina and Colombia), contributions of non-renewables and renewables partially offset each other. Disaggregated data show that fossil fuels are a major source of growth in Colombia, China and Brazil, while mineral resources drive significant income growth in Australia, Peru and Chile. In contrast, countries like Norway and Denmark show negative contributions from non‑renewables, specifically fossil fuels, indicating a continued decline in extraction and greater decoupling of income growth from exhaustible resources.

Renewable natural capital contributions to growth are limited.

Renewable natural capital makes a marginal or negative contribution to income growth in most countries. Only Estonia, Latvia, Indonesia and Argentina show clear positive gains from renewables. Data disaggregated by type of input provides further insights. Land use expansion, once a major growth factor, now plays a limited role. Only Indonesia, Argentina, and Estonia show contributions above 0.1 pp over the period. In fact, income from land resources (pastureland, cropland, and production-forest land) declined in most countries, with important drops in Costa Rica and Finland - specifically in production-forest land and pastureland (Cárdenas Rodríguez et al., 2023[29]). Non-cultivated biological resources are a positive source of income only in Estonia, Latvia and New Zealand, mainly via non-cultivated timber extraction. In contrast, biological resources show negative income growth in Peru and Iceland, due to strong declines in wild fisheries. Ecosystem services (coastal protection from mangroves, watershed protection from forests, non-wood forest products) contribute minimally to income growth. Among countries, China and Chile show the highest positive contributions, while Brazil and Argentina the most negative, adding or reducing around 0.004 pp of income growth per year.

The low contribution of renewable natural capital to income growth, compared to non‑renewables, could reflect structural and measurement limitations. Non-renewable resources are typically commodified, with well-established markets capturing their private value. In contrast, renewable resources, such as ecosystem services or non-cultivated biological assets, are often non‑commodified, undervalued, or underreported. Many are subject to open access and overexploitation, limiting their long-term growth potential. Incomplete data on informal, subsistence or illegal extraction (e.g., fish, timber), further contributes to their underrepresentation in income growth indicators.

Heavy reliance on non-renewable natural capital poses major long-term risks. Countries depending on extractive growth face vulnerabilities from declining stocks and demand shocks. In some cases, offsetting trends (e.g., growth from renewables counterbalancing declines in fossil fuels), suggest gradual shifts in the composition of growth. However, these trends remain limited. Long-term income from renewables must be based on levels of extraction that preserve ecological stocks. Ensuring that resource flows remain within sustainable limits is essential. Diversifying sources of income growth sources and improving the valuation of natural capital are critical to better reflect environmental risks in macro-economic indicators. Environmentally adjusted metrics and targets are key tools to align economic performance with sustainable long-term living standards.

Indicators

Renewable energy remains a largely untapped source of income growth

The OECD’s environmentally adjusted multifactor productivity (EAMFP) framework captures how renewable energy contributes to pollution-adjusted GDP growth (Cárdenas Rodríguez, Haščič and Souchier, 2018[28]). While results remain exploratory due to limited data on costs, prices, and output volumes, the framework offers insight into the long-term potential of renewables to support income generation.

Between 1996 and 2018, hydro, onshore wind, and solar PV made a small but consistently positive contribution to pollution-adjusted GDP growth across OECD countries. In contrast, fossil fuel inputs showed more volatile and often had negative contributions. In the EAMFP framework, fossil fuels contribute to income primarily through their resource rent. However, because the use of fossil fuels leads to GHG and they are an exhaustible resource, their extraction and use result in a downward adjustment to output growth, and their long-term contribution to growth is structurally limited.

Despite its potential, renewable energy remains largely unexploited as a source of income growth. Yet, decreasing technology costs are making renewables increasingly cost‑competitive in most countries. To accelerate the green energy transition and reap rents from renewables, governments need to reduce barriers in the form of permitting delays, limited grid access, and restrictions on cross-border technology transfer. At the same time, energy pricing must internalize environmental externalities to better reflect the true cost of fossil fuels. Recognising renewable energy as a productive input allows a more informed energy policy and supports long-term environmentally sustainable income growth.

Comparability and interpretation

The relationship between growth accounting and environmental sustainability needs to be interpreted with caution. These indicators do not measure resource depletion or environmental quality per se and should not be viewed as indicators of environmental sustainability, that is, the capacity of the environment to continue providing natural resources and ecosystem services. In growth accounting, inputs and outputs are evaluated from the producers’ perspective, using market prices or private opportunity costs. This approach makes no account of non-market environmental damages and other social costs of pollution. Indicators from the EAMFP framework are therefore not measures of social welfare but tools to improve the conventional productivity measurement by better assessing the efficiency of input use in generating income. To capture the non-market aspects of environmental degradation, EAMFP indicators should be used in combination with a broader set of indicators covering social and the environmental outcomes.

In this edition, the coverage of the environment remains partial. It is limited to 12 types of air emissions and 22 natural capital inputs (see Glossary). Due to data limitations, many natural resources (e.g. soil, biodiversity) and ecosystem services (e.g. crop pollination, air and water purification, avalanche and landslide prevention, landscape amenities, etc.) are not yet considered. Pending better data availability, future work will aim to broaden coverage to include more natural resources and ecosystem services. In addition, recent crises, including the Covid-19 pandemic, energy price shocks, geopolitical tensions and inflation, have altered long-term growth trajectories, creating lasting economic disruptions. Current EAMFP indicators capture pre- and early-crisis trends; some findings may need reassessment as new data emerge and structural impacts become clearer.

Growth accounting measures only year-to-year changes (“growth”), not the volume (“level”) of inputs or outputs in national income. Consequently, these indicators should not be interpreted as contributions to the level of GDP. For example, a zero contribution of natural capital in a given year does not mean that no resources were extracted, rather, that the level of extraction remained unchanged from the previous year. Similarly, a zero adjustment for pollution abatement means that emissions remained constant year-on-year. These indicators provide an aggregated picture of the economy over a year. As a country-level economic measure, it can hide important sectoral or firm-level differences. These annual indicators are also sensitive to the business cycle and may be more volatile during economic recessions. Analysing long-term trends, as presented here, helps to mitigate these concerns.

Policies and regulations

Key messages

• The share of environmentally related tax revenue (ERTR), both as a percentage of total tax revenue and GDP-equivalent, continues to decline across most OECD countries. This limits the effectiveness of fiscal policy in addressing environmental externalities. Pollution and resource taxation remains underutilised, despite their potential to support circular economy and biodiversity objectives. Environmental taxes can be a cost-effective instrument for achieving environmental goals and, when well designed, can enhance economic efficiency by shifting the tax burden away from labour and capital.

• Feed-in tariffs played a key role in the early deployment of solar and wind energy technologies but have largely been phased out as these technologies matured. Their replacement with other market-based instruments reflects improved cost‑competitiveness and highlights the importance of adaptive policy design in advancing the green transition.

• Environmental taxes and subsidies remain unevenly distributed across sectors, often lacking internal alignment. Better coherence of economic instruments within sectors can support green technology adoption, reduce policy distortions, and strengthen structural shifts towards less environmentally harmful economic activities.

• Public investment in environmentally related R&D in OECD countries has remained stable in the last two decades, with few countries making progress. Continued and well-targeted public R&D investment is needed to lower costs and scale up deployment of emerging green technologies.

• Governments have increasingly redirected public energy RD&D away from fossil fuels towards renewables, with notable growth in solar, wind, and geothermal support. This shift is key to bridging clean energy technology gaps and enabling large-scale transformation of the energy sector.

• The use of regulatory instruments such as bans or phase outs on GHG-intensive activities has grown, especially in transport and energy supply, but adoption remains uneven. Broader, long-term regulation and better policy tracking are critical to accelerate the green transition.

• Mandatory energy-efficiency labelling for appliances and vehicles has expanded significantly, helping to reduce information asymmetries and promote greener consumption. However, there is scope to further strengthen compliance, harmonise standards, and assess their overall effectiveness.

Policy context

Environmental policies and regulations are essential to advancing sustainable development. Governments shape environmental and economic outcomes through a mix of market-based (so-called economic) instruments such as taxes and subsidies, and non-market-based instruments like direct government investment, regulatory instruments, and information instruments. Together, these policy instruments shape prices and quantities across the economy, aligning production and consumption with environmental goals.

Pricing mechanisms, including environmentally related taxes and permit trading systems, aim to internalise the environmental cost of pollution and encourage investment in cleaner processes and technologies. Revenue from environmentally related taxes can support fiscal reform by shifting the tax burden away from labour and capital towards environmentally harmful activities, potentially improving both environmental effectiveness and economic efficiency. Potential negative distributional concerns of these policies can be addressed through measures targeted at the most vulnerable firms and population groups.

Subsidies and tax incentives (e.g., tax credits, tax exemptions) play a complementary role in the green transition, accelerating innovation and uptake of less harmful technologies. Subsidies can drive structural shifts in the economy, by making greener alternatives more competitive. Feed-in tariffs (FiTs), in particular, were instrumental in the early deployment of renewable energy, providing price stability and reducing investment risk. These instruments stimulated rapid innovation and economies of scale, driving down costs and enabling solar and wind to become increasingly cost-competitive without continued subsidy support (e.g., (Johnstone, 2010[30])). As markets mature, many countries are phasing-out FiTs to manage fiscal costs and address new challenges such as grid integration. However, green subsidies may generate unintended effects, for example, increased private car ownership from vehicle purchase incentives, which may counteract broader environmental goals. The evolving renewable energy policy mix reflects the need of the green transition to balance fiscal sustainability, technological progress, and long-term public policy goals.

Government investment, particularly in research and development (R&D), is critical for supporting activities that generate long-term benefits for society, especially in areas where markets alone fall short. In the transition to a greener economy, public R&D spending drives innovation, technological advancement, and the diffusion of new solutions to address environmental degradation. Globally, the market for green technologies, such as renewable energy and low-carbon technologies, is growing. However, some of these technologies are still at an early stage and greater government support is needed to further reduce their costs, improve their effectiveness, and enable widespread deployment. Government spending on environment- and green energy-related R&D lays the groundwork for tackling pollution at its source and securing environmental and economic benefits for future generations.

Regulatory instruments remain essential, especially where market signals alone are insufficient or environmental damages are high. Standards, bans and phase-outs, set binding requirements that ensure an environmental outcome. They provide clear, long-term signals to markets and investors on public policy objectives, aligning industrial strategies with environmental sustainability goals. Their effectiveness depends on sound design, long-term predictability (low regulatory uncertainty) to spur the necessary investment in abatement technologies and robust enforcement capacity.

Information-based instruments complement public investment, regulatory approaches and economic instruments by guiding decision-making without imposing direct constraints or changing relative prices. Instruments such as mandatory energy labelling reduce information asymmetries and help consumers and businesses make less environmentally harmful choices. By lowering search costs and increasing transparency, these instruments can influence market behaviour and stimulate demand for greener products and services.

A coherent policy mix, disincentivising environmentally harmful activities while supporting sustainable alternatives, can enhance policy predictability and the effectiveness of policy reforms. There is no one-size-fits-all approach to greening the economy. Countries’ economic context, administrative capacity, and political economy concerns, shape the effectiveness of environmental policies and regulations. In some contexts, phasing out environmentally harmful subsidies may require parallel investment in green infrastructure or social protection. Environmentally positive subsidies and taxes on externalities can jointly support transitions in resource- and pollution-intensive sectors, by enabling green technology adoption, sectoral change, and mitigating competitiveness and economic risks. Policy packages that combine regulation, taxation, and targeted support, especially when tailored to national circumstances, tend to deliver better outcomes (Stechemesser et al., 2024[31]) (OECD, 2024[32]).

To assess the feasibility and fairness of green policy reforms, indicators presented here should be interpreted alongside socio-economic indicators (e.g., employment, income distribution, and competitiveness) available in the OECD Green Growth Indicators database on OECD Data Explorer (OECD, n.d.[5]), and with broader economic policy indicators (e.g., product market regulation, trade policy) which provide context on the alignment of public policy with the green transition.

This section complements policy and government action indicators presented in other chapters and sections of Environment at a Glance Indicators. Those indicators, often tailored to specific environmental domains, are not repeated in this section to avoid redundancy. They include:

• Climate change: Climate action and policies, taxes relevant for climate change mitigation, fossil fuel support measures, net effective carbon rates, and Official Development Assistance for climate change mitigation and adaptation.

• Air quality: Taxes relevant for air pollution abatement.

• Freshwater resources: Taxes relevant for freshwater resource management (forthcoming).

• Circular economy, waste and materials: Taxes relevant for circular economy and resource efficiency (forthcoming).

• Biological resources and biodiversity: Designation of protected areas, taxes relevant for biodiversity.

• Sustainable ocean economy: Marine protected areas, economic instruments related to ocean sustainability, and policies and practices against illegal, unreported and unregulated fishing.

• Green transition (other sections): Public R&D for the environment, public RD&D for energy, mandatory instruction time on natural sciences, public investment in energy infrastructure, sovereign green bonds, and environmentally related Official Development Assistance.

Main trends and recent developments

Downward trend in environmentally related tax revenue across OECD countries since 2018

Environmentally related tax revenue (ERTR) continues to decline in relative terms across OECD countries. As a share of GDP-equivalent ERTR has decreased since the early 2000s. In 2023, it represented 1.3% of GDP-equivalent, while the share slightly increased from the preceding year, it remains well below 2000 levels of 1.8%. This trend suggests that economic growth and inflation have outpaced environmental taxation, rather than structural reforms. This diminishing role of fiscal policies for addressing environmental externalities can undermine the cost‑effectiveness of achieving environmental goals.

The share of environmentally related taxes in total tax revenue is also decreasing

The share of ERTR in total tax revenue has also declined. In 2023, ERTR accounted for about 4.3% of total tax revenue on average across the OECD, down from 5.9% in 2000. Romania, Bulgaria, Costa Rica, and Türkiye have the highest shares, while Argentina, the United States, Brazil, and Colombia show the lowest. Unlike labour or sales taxes, environmental taxes are generally defined in physical units (e.g. EUR per litre of fuel) and require regular adjustment or indexation to keep pace with inflation and economic growth. Energy taxes remain the main source of ERTR in OECD countries. Transport taxes are particularly relevant in countries such as Türkiye, Costa Rica and Brazil. Pollution and resource taxes make up a small share, indicating underuse of more targeted environmental taxation to support pollution control, waste management, and biodiversity protection.

The shift of tax burden from labour to environmental degradation remains limited

Only a few countries have increased ERTR faster than labour tax revenue over the past two decades. Brazil, Estonia, Türkiye, Poland, and Israel stand out for positive shifts in their tax mix. In most countries, labour taxes have continued to grow faster than environmental taxation, limiting the role of fiscal reform in simultaneously addressing environmental and economic objectives. Some countries have introduced new environmentally related taxes as part of fiscal consolidation efforts, e.g. taxes on nuclear fuel and air travel, carbon taxes or vehicle tax rates linked to CO₂ emissions and to local air pollution.

Most tax revenue comes from household activities, transport, and electricity production

SEEA-based sectoral data show that ERTR by economic and household activities is primarily raised from households, while manufacturing and electricity generation tend to dominate the revenue raised from production sectors. However, distribution varies substantially across countries. For example, Romania and Poland raise a high share of ERTR from manufacturing, while Luxembourg relies heavily on transportation taxes. Services and wholesale/retail activities also contribute significantly to total ERTR in Bulgaria and Lithuania respectively. The distribution of environmental taxes by economic and household activities helps to identify potential areas for more tailored reform.

Feed-in tariffs for renewable electricity have declined sharply across countries.

Feed-in tariffs (FiTs) for solar PV and wind power peaked in 2010 across OECD and Key Partner countries. Solar PV FiTs remained higher than those for wind for most of the previous two decades. By 2020, most countries had phased out FiTs. Nonetheless, a few countries maintained relatively high FiTs such as Argentina and Bulgaria for solar PV, and Switzerland, Argentina, and Japan for wind power. Governments have increasingly replaced FiTs with other market-based instruments, such as competitive auctions, contracts for difference, or renewable portfolio standards with tradable certificates.

Uneven distribution of environmental taxes and subsidies across economic activities

Environmental taxes and subsidies can support the green transition across the economy. Governments can discourage environmentally harmful processes by levying environmental taxes, while promoting sustainable alternatives (i.e., clean technology adoption, input substitution) through environmentally (positive) subsidies and transfers. Data from 17 European countries suggests that these instruments are concentrated only in a few sectors, albeit unevenly distributed. Economic activities such as electricity, gas, and steam, as well as the agriculture, forestry, and fishing sectors receive the largest share of environmental subsidies but are subject to little environmental taxes. In contrast, household consumption activities and economic sectors such as transportation, bear a disproportionately high share of environmental taxes while receiving minimal subsidy support. This asymmetry highlights an opportunity for better alignment and coherence between taxation and subsidy policies to reinforce within-sector shifts towards greener activities while maintaining competitiveness and mitigating potential distributional impacts. Harmful subsidies need to be accounted for to provide a complete picture of policy incentives, along with other instruments such as regulations and direct public investment.

Little progress in boosting public R&D for environmental objectives.

Since the early 2000s, the share of environmentally related R&D in total government R&D budgets across OECD countries has remained relatively stable, reaching on average of 1.8% in 2022-24. However, some countries are making progress. Colombia recorded the highest share, with 15.6% of its total government R&D budget dedicated to environmental objectives. Latvia (8.6%) and Finland (8.1%) followed. Notable increases over the last decade were also observed in Portugal and Australia in addition to Latvia and Finland.

Governments are funding less fossil RD&D but also less innovation in renewables

Public spending on energy research, development, and demonstration (RD&D) has shifted significantly over the past two decades, with slowdown during the last decade Many countries have reallocated their RD&D spending away from fossil fuels and towards renewable energy sources, such as hydro, geothermal, solar, and wind. In OECD countries, the share of renewable energy RD&D rose from 12.0% in 2000-02 to 14.8% in 2021-23. This shift could be explained by a number of factors, including the need to bridge technology deployment gaps in the clean energy sector, the growing recognition of renewable energy as a promising investment opportunity, and public support (IEA, 2025[25]).

Wide gaps in public investment in renewable energy RD&D

The composition of energy-related RD&D varies across OECD and Key Partner countries. In 2021-23, New Zealand allocated nearly 80% of its energy RD&D budget to renewable energy technologies. Ireland, Chile, Denmark, the Netherlands, and Switzerland also recorded high shares. Since 2000-02, New Zealand, Ireland, and Norway have significantly increased their public support for renewables. Conversely, fossil fuel RD&D has decreased in most OECD countries. Notable reductions since 2000-02 were observed in Norway, New Zealand, Australia, Canada, and France.

Regulatory action on fossil fuels has gained momentum but coverage remains uneven.

Governments are increasingly adopting bans or phase-outs on fossil fuel extraction, use, and consumption of derived products. They also target GHG-inducing activities such as pollution-intensive waste practices and international public finance on fossil fuel products and infrastructure. Adoption of bans and phase-outs accelerated after 2010, but progress remains limited. Across 47 countries with data, about a third are decommissioning coal-fired power plants and only four have legislated bans or phase-outs on fossil fuel extraction. Long-term supply-side policies are crucial, as they provide clearer investment signals, improve policy predictability, and support aligning industrial strategies with environmental goals.

As of 2024, bans or phase-outs on ICE passenger vehicles are most common, adopted by 28 of 47 countries, followed by coal power plant phase-outs and landfill bans, enacted in 18 and 16 countries, respectively. In contrast, no country has legislated an end to coal export credits, and only one is phasing-out gas heating in buildings. Over the past decade, most progress has occurred in the adoption of bans or phase outs for ICE vehicles, coal power, and international public financing of fossil fuel infrastructure. Nonetheless, major gaps persist, with many high-emission activities remaining unregulated. These findings highlight the growing role of regulatory (non-market-based) approaches, and the need for consistent, harmonised tracking of such policies to assess the green transition of government action.

Energy efficiency labelling expands across products and countries.

Most countries have introduced energy-efficiency labelling for consumer products, and passenger vehicles. As of 2024, all 47 countries with available data have mandatory labels in place for refrigerators, freezers and air conditioners. Most schemes were adopted before 2010, with the exception of labels for air conditioners adopted by three fourths of countries after 2010. Nevertheless, coverage can be improved, mandatory energy-efficiency labelling is not in force in seven countries for passenger vehicles and two countries for lighting. These developments reflect a growing recognition of information instruments as a key government action, helping reduce information asymmetries and guiding households towards more resource-efficient consumption.

Indicators

Green provisions in Regional Trade Agreements

Regional Trade Agreements (RTAs) have increasingly incorporated environmentally related provisions. These provisions do not impose market-based mechanisms or regulations but promote environmental cooperation, legal commitments and attention to issues such as biodiversity, air pollution, and climate change. They foster coherence between trade and environmental goals.

While evidence on their environmental effectiveness remains limited, some studies (George and Yamaguchi, 2018[33]) suggest that environmental provisions in RTAs can support environmental goals through strengthened legal frameworks, enforcement cooperation mechanisms, and institutional capacity. Common tools include implementation and enforcement mechanisms, public participation, and dispute settlement procedures.

Where provisions are followed by concrete actions, such as joint programmes or improved transparency, RTAs can contribute to better environmental assessment practices, natural resource management, and environmental protection. However, the extent and quality of implementation remain uneven across agreements.

Trade agreements increasingly include environmental provisions

Cumulative number of RTAs with environmental provisions, 1990-2021

Source: OECD based on TREND database (Bellmann and Bulatnikova, 2022[34]).

Climate policy uncertainty

Policy uncertainty is a major barrier for long-term investment decisions. Uncertainty in climate change policies causes market unpredictability, which can lead firms to delay investments in low-carbon technologies and infrastructure. This adds to the barriers faced by private firms to access finance and invest in green or low-carbon solutions.

The Climate Policy Uncertainty indicator may be used to gauge the evolution of domestic climate policy, which can help assess the impact of policy uncertainty on investment and innovation. Using newspaper coverage frequency, the indicator measures the spikes associated with political events and climate policy changes. It focuses on 12 OECD countries: Australia, Canada, Chile, France, Germany, Ireland, Italy, Mexico, New Zealand, Spain, the United Kingdom, and the United States from 1990 to 2018. The figure below shows country-specific peaks corresponding to events around major climate policies and initiatives, such as:

• Repeal of the ETS in Australia in 2011-13.

• Debate in 2016-18 around Canada’s Federal Carbon tax policies.

• Germany spike between 2011-2014 is a result of major reforms of ‘Energiewende’

• In 2016-2017, COP21 Paris Climate Summit drove France’s peak

• In 2017, withdrawal from the Paris agreement by the United States

• New Zealand suspension of the ETS led to large spike in 2008.

Tracking climate policy uncertainty across major OECD economies

Note: The index represents the average of standardized series across newspapers per country by year. Each series is normalized to have a mean of 100 over the 1990-2018 period. Higher values reflect greater media-reported uncertainty about domestic climate policy.

Source: OECD, “Climate Policy Uncertainty”, OECD Economic Department (Database); Berestycki, C. et al. (2022), “Measuring and assessing the effects of climate policy uncertainty”, OECD Economics Department Working Papers, No. 1724, OECD Publishing, Paris,..

Comparability and interpretation

The level of environmentally related tax revenue should not be used to assess the environmental effectiveness of the tax system. Revenue is an outcome that depends on tax design, tax rates, tax base elasticity, and existing pollution or consumption levels. Some environmentally related taxes may generate little or no revenue while still inducing significant behavioural changes. Fees, charges, and royalties related to resource management are not included, limiting coverage of broader fiscal instruments. Cross-country comparability requires careful interpretation in light of differences in economic structures and approaches to environmental policy (instrument choice). Expressing environmentally related tax revenue as a share of GDP allows for cross-country comparison of the scale of such taxes relative to the size of the economy. However, this indicator should be interpreted strictly in that context, as tax revenue is not a component of GDP.

Subsidies supporting the green transition are diverse and evolving. Instruments such as feed-in tariffs have been widely used to promote renewable energy deployment. Yet, in recent years, many countries have shifted towards other market-based mechanisms. As countries diversify their policy mix, it becomes increasingly important to track policy support for environmental objectives systematically and comprehensively. In this context, progress is needed in improving national reporting practices of public environmental protection expenditures, including green budgeting. The lack of transparent and harmonised reporting limits comparability and the assessment of policy effectiveness. Ultimately, subsidies on less environmentally harmful alternatives are a second-best approach, direct pricing of harmful activities (e.g., carbon taxes, natural land conversion taxes) remains the most cost-effective and first-best approach.

The SEEA framework standardises the classification and reporting of environmentally related tax revenue and subsidy accounts, ensuring alignment with the System of National Accounts. This enhances comparability across countries and over time. Nonetheless, challenges remain due to differences in national policy design and important gaps in the compilation of these accounts. Ambiguities in defining environmental intent and impact further complicate international benchmarking.

R&D budget is an input indicator that reflects a government’s effort to invest in knowledge. It reflects intent, rather than outcome; high R&D spending does not guarantee superior innovation performance. While OECD data are harmonised and comparable across countries for broad functions of government, such as environmental protection, health or education, national surveys coverage, sampling and estimation techniques may vary at a more detailed level. Harmonised data on private-sector R&D expenditure present significant gaps and are not well-suited for cross-country comparisons.

Command-and-control or regulatory instruments can be effective tools to achieve specific environmental outcomes, unlike market-based tools such as taxes which influence behaviour through prices but do not guarantee quantity limits. However, cross-country comparability is affected by differences in scope, legal forms, enforcement capacity, and implementation timelines. Variations in sectoral coverage, type of standards, and national context, also affect the interpretation of these indicators. For example, the absence of a regulation may reflect structural shifts rather than policy gaps. Normalised and comprehensive metrics on policy coverage, stringency, and enforcement, can enhance comparability of indicators on regulatory instruments and support more informed assessments of overall environmental policy effectiveness.

Information instruments, such as mandatory labelling schemes for energy-efficiency appliances and passenger vehicles, aim to influence consumer behaviour and encourage the adoption of energy efficient technology. While data on existence and coverage of such schemes are available, limited information exists on compliance, enforcement, or actual market impacts. Definitions of label classes and thresholds vary across jurisdictions, hindering comparability.

Current indicators cover selected fiscal, regulatory, and information instruments but do not capture the full breath of environmental policy frameworks. Important public policy approaches, such as environmentally harmful subsidies, government procurement, targets and voluntary agreements, remain beyond the scope of this chapter. Comprehensive cross-country comparison is constrained by definitional gaps or reporting inconsistencies. Expanded and harmonised policy measurement frameworks, such as the Climate Actions and Policies Measurement Framework (CAPMF) (Nachtigall et al., 2022[35]) and the Environmental Policy Stringency (EPS) (Kruse et al., 2022[36]) which focus mostly on climate change mitigation and air pollution, are a step forward to fully assess the policy landscape steering the green transition.

Measurement agenda

While notable advances have been made in measuring the green transition, significant statistical gaps remain, limiting the ability of governments to design and evaluate effective policies. At the macro level, a priority is to accelerate the implementation of the SEEA Central Framework and the SEEA Ecosystem Accounting. These frameworks provide a harmonised basis for measuring the use and conditions of natural resources and ecosystems. Currently, however, they are fully implemented in only a few countries, often limited to specific accounts such as physical energy flows or air emission accounts.

Broader adoption of the SEEA would improve national capacities to assess environmental pressures, resource efficiency and ecosystem degradation. It would also facilitate the integration of environmental-economic indicators such as adjusted productivity, natural capital accounts, and environmental footprints into policymaking. This integration is essential for tracking structural change and the decoupling of resource use and pollution from economic growth, particularly in economies where natural capital and pollution-intensive activities play a central role.

Measuring progress in the green transition is further constrained by the limited availability of suitable outcome metrics complementary to GDP. Policymakers require better data to monitor social, environmental and economic sustainability in an integrated manner (Stiglitz, Fitoussi and Durand, 2018[37]). The upcoming revision of the SNA in 2025 is expected to provide a more robust statistical foundation for embedding sustainability indicators into core economic frameworks (United Nations, forthcoming[38]).

Priorities for indicator development

Strengthening the evidence base for green transition policies requires filling persistent data gaps and aligning measurement efforts with policy needs. Priorities include:

• Developing ecological limits indicators to capture ecosystem pressures and limits, such as carbon budgets, nutrient balances, pollination rates, biodiversity loss, and biocapacity.

• Expanding indicators on household behaviour and consumption, including price signals, social norms and behavioural drivers of green choices.

• Building better metrics on green skills, through international comparability of training programmes and re-skilling initiatives for labour market transitions.

• Improving indicators on green and brown trade in services, especially in sectors critical to the transition, such as energy and transport.

• Developing harmonised economic activity classifications to better distinguish environmentally sustainable production and consumption within conventional industry categories including for organic agriculture, sustainable forestry, renewable energy in manufacturing, eco-tourism.

• Enhancing data on green and brown private finance, including capital flows, green bonds and investment in low-carbon infrastructure.

• Capturing representative data on private sector R&D and business innovation to understand green innovation trends and technology diffusion.

• Tracking policy coherence and enforcement across sectors, to assess whether government action aligns with stated environmental sustainability goals.

• Improving classification of green and brown components of public finance, including climate-related public expenditure, environmentally harmful subsidies, and green budgeting practices.

Advancing these priorities will provide the robust, harmonised, and foundational data needed to monitor structural changes, inform policy design and evaluate outcomes. Strengthening the indicator base is essential to ensure that the green transition is measurable, actionable and aligned with long-term environmental sustainability goals.

Glossary

Consult the Environment at a Glance Indicators page to find the full publication, a reader’s guide, country notes, regional analysis, and direct access to the data.

Employment and education

Green-task jobs are defined as occupations in which at least 10% of tasks are classified as “green”. Tasks are the functional components of occupations, performed at varying frequencies. Green tasks contribute to environmental sustainability or pollution reduction. The share of green-task jobs is calculated by linking this task classification with employment data, such as labour surveys or administrative records.

The green task taxonomy is based on the Occupational Information Network (O*NET) database, which identifies 1369 green tasks across 923 occupations in the United States. The identification of green tasks follows the work of (Vona, Marin and Consoli, 2018[39]) (Vona et al., 2018[40]). This “bottom-up”, task-based, approach supports the analysis of labour market impacts of the green transition and is a widely used approach for studying labour market structural transformations. Cross-country estimates use mappings (crosswalks) to align national occupation classifications with this reference. The 10% threshold provides a robust binary classification consistent across classification levels (from 6-digit to 3-digit). The robustness of the threshold, the concepts, and the measurement methodology are discussed in (OECD, 2025[41]; OECD, 2023[12]).

GHG-intensive jobs are defined as all jobs within industries that rank in the top 20% of greenhouse gas (GHG) emissions per unit of value added. This includes manufacturing, electricity, transportation and storage, and mining and quarrying, which are identified as GHG-intensive in at least half of the countries covered by the OECD.

This indicator uses a “top-down”, industry-based approach, in contrast to the task-based method used for green-task jobs. GHG intensity is calculated using national emissions and economic output data. Employment in GHG-intensive sectors is then estimated using labour force or national accounts data. For further methodological details, see (OECD, 2024[6]; OECD, 2025[41]).

Compulsory instruction time on natural sciences is the mandatory teaching time, that must be provided in public schools, dedicated to science-related subjects such as physics, chemistry, biology, environmental sciences, and ecology. This excludes elective or optional courses and varies across countries depending on national curricula. Primary education typically covers ages 5/7 to 10/12 and lower secondary general education ages 10/13 to 14/16. The average instruction time per year accounts for the duration of education levels by country. It is expressed in full-day (24 hours) equivalent, and as a share in total compulsory instruction time per year.

The indicators cover education levels as defined by the International Standard Classification of Education (ISCED 2011): primary education, typically for ages 5/7 to 10/12, corresponds to ISCED 1; and lower secondary general education, typically for ages 10/13 to 14/16 ISCED 24. In some countries, the scope may include additional subjects (OECD, 2018[42]). For example, Korea includes ICT, technology, and vocational content under natural sciences. For more information, please consult (OECD, 2024[7]). Across OECD countries where instruction time per subject is specified, the six subject areas with the largest shares in the compulsory curriculum at primary education level are 1) reading, writing, and literature; 2) mathematics; 3) arts; 4) physical education and health; 5) natural sciences; and 6) social sciences. The curriculum at lower-secondary education level is mainly composed of 1) reading, writing, and literature; 2) second and other languages; 3) mathematics; 4) natural sciences; 5) social sciences; and 6) physical education and health – other subject areas in compulsory curricula include, for example, ethics or religion.

Bibliometric data capture scientific research output by linking publications to the country affiliations of their authors and to journal classifications. Research output is measured as fractional publication counts by scientific field and by country, i.e., each publication is weighted by the number of co-authors, their country affiliations, and the publication classifications according to the All Science Journal Classification (ASJC) taxonomy.

The OECD Bibliometric indicators database (see Data sources) used, is based on Scopus Custom Data (Elsevier) and the SCImago Journal Rankings. Indicators are presented as index and per million inhabitants. The field Environmental Sciences corresponds to ASJC code 23, while Renewable Energy, Sustainability and the Environment corresponds to ASJC code 2105, within the broader Energy field ASJC code 21. For more information on concepts and methodology, please consult (Dechezleprêtre and Kelly, forthcoming[18]).

Markets of goods and services

Environmentally related patents measure inventive activity using patent microdata classified across a wide range of environmentally related technological domains. Patent counts include only higher-value inventions (i.e., with applications in at least two jurisdictions). Patent indicators are calculated as fractional counts of patent applications by country of residence of the inventor(s), priority date (i.e., first application date), and technology domain. The indicator environmentally related inventions includes environmental management and climate change mitigation technology domains (i.e., technologies related to adaptation or the ocean economy are excluded).

Environmentally related patents are identified using a number of technology classes pertaining to environmental management (air and water pollution, waste management, soil remediation and monitoring), climate change mitigation (buildings, transports, goods, information and communication technologies, greenhouse gases, renewable energy, wastewater and waste), climate change adaptation and sustainable ocean economy (renewable energy generation, pollution abatement, transport, fishing and aquaculture, desalination and adaptation in coastal areas). Patents can be classified in more than one technological domain; therefore, the sum of environmental domains is higher than the total of environmentally related inventions. For more details on the patent class cartography for environmentally related technology please consult Haščič and Migotto (2015[43]).

Fossil fuel energy supply patents are identified using a cartography of technologies contributing to the exploration for and extraction of fossil fuels, as well as their transformation and the delivery of fossil fuel products to users. Fossil fuel energy resources include oil (petroleum oils and liquefied petroleum gas), natural gas, coal and other solid fossil fuels.

Fossil fuel energy supply inventions are divided into upstream, processing and downstream, and transmission and distribution sectors. Upstream technologies, consist of inventions that are directly related to the exploration and extraction of fossil fuel resources from conventional and unconventional sources. Processing and downstream technologies include the transformation of fossil fuels into products such as motor fuels, aviation fuel, synthetic gas, hydrogen and others. Transmission and distribution technologies include the distribution of fossil fuels to end-users via pipes, ships, fuelling stations and including fuel storage. Technologies designed to improve devices for the combustion or other energy-related uses of fossil fuels, including reducing their greenhouse gas emissions, are not included due to significant overlaps with energy efficiency technologies among the low-carbon end-use energy technologies. For details on the cartography of fossil fuel energy supply technologies developed by the EPO and the IEA please consult (IEA, 2021[44]) (IEA, 2021[45]).

The relative advantage in environmentally related technologies is an index of the specialisation in environmental innovation of a given country relative to the world average. It reflects whether a country is more or less focused on environmentally related technologies than the world as a whole. This index is calculated as the ratio of 1) the share of environmentally related inventions on all inventions in all technologies at home and 2) the share of environmentally related inventions on all inventions in all technologies in the world. An index equal to 1 means that a country innovates as much in 'green' technologies as the world; an index above 1 indicates a relative technological advantage (RTA), or specialisation, in environmentally related technologies compared to the world value.

Green trademarks refer to trademark applications for products or services related to climate change mitigation or adaptation. These include technologies such as electric vehicles, batteries, renewable energy, and hydrogen. The indicator captures climate change-related trademark filings at the European Union Intellectual Property Office (EUIPO), the Japan Patent Office (JPO), and the United States Patent and Trademark Office (USPTO). The indicator reflects the average across the three offices, and it is expressed as a share of total filings by country.

Trademarks are filed in accordance with the International Classification of Goods and Services, also known as the Nice Classification (see https://www.wipo.int/classifications/nice/en). Climate-related applications are identified using natural language processing (NLP) applied to the trademark text. The classification draws on the Y02 patent taxonomy, selected keywords from academic literature (Garcia-Valero, Faria and Wajsman, 2021[46]), the Elsevier SDG vocabulary (Jayabalasingham, 2019[47]), and the topic modelling of climate change-related start-up activity. Accuracy depends on the quality of textual descriptions and the relevance of the keyword set.

Environmentally related goods (EGs) refer to products that serve environmental protection, resource efficiency, or climate mitigation purposes. Product categories include renewable energy equipment, air pollution control, waste and wastewater management, and resource-efficient technologies. EGs are identified using the OECD Combined List of Environmental Goods (CLEG), while trade volumes are measured using the BACI International Trade database. Note that the label “environmentally related” for these indicators deviates from the label “environmental” in source data terminology to highlight potential differences with the scope and lists of environmental goods and services in line with the SEEA.

The OECD Combined List of Environmental Goods (CLEG) maps 345 Harmonized System (HS) 6-digit codes to relevant environmental categories, based on Sauvage (2014[48]), and Moïsé and Tresa (2025[20]). The CLEG consolidates environmentally related products from the OECD’s plurilateral environmental goods and services (PEGS) list, the WTO Friends Group list, and the APEC list. For more information on the BACI International Trade database see (Gaulier, G. and Zignago, S, 2010[49]). For full HS code coverage and classification criteria see Moïsé and Tresa (2025[20]). EGs categories are:

• Air pollution control

• Cleaner/resource-efficient technologies and products

• Environmentally preferable products (end-use or disposal characteristics)

• Heat and energy management

• Monitoring and assessment equipment

• Natural resource protection

• Noise and vibration abatement

• Renewable energy plants

• Waste management and recycling

• Soil and water remediation

• Wastewater management and potable water treatment

Green start-ups are early-stage companies developing products or services aligned with environmental sustainability. Green start-ups are grouped into eight sectors: energy storage, energy efficiency, low-carbon vehicles and transportation infrastructure, clean energy, sustainable food and agriculture processes (e.g., plant-based protein), pollution abatement processes (e.g., bioremediation), circular economy (e.g., waste reduction, recycling, repurpose, reuse), and an “Other” category for unclassified green firms.

Data on green start-ups is sourced from The New Startup Database by the Science, Technology and Innovation Directorate at the OECD. The Identification of green start-ups follows a two-step method. First, green-labelled firms are selected from the two private data providers: Crunchbase and Dealroom. Second, a string-searching algorithm applies a curated green vocabulary to firms’ descriptions to identify additional start-ups. Manual checks decrease false positives and excludes potential green washing (i.e., misleading impression on the environmental sustainability of products or services). The green vocabulary draws on the Y02 patent taxonomy, selected keywords from academic literature (Garcia-Valero, Faria and Wajsman, 2021[46]), the Elsevier SDG vocabulary (Jayabalasingham, 2019[47]), and the topic modelling of climate change-related start-up descriptions selected in the first step. For more information consult (Dechezleprêtre and Kelly, forthcoming[18]).

Air emission accounts in line with the System of Environmental-Economic Accounting Central Framework (SEEA-CF) record the release of air pollutants and greenhouse gases into the atmosphere resulting from production activities and household consumption. These accounts apply the same classifications and boundaries as the System of National Accounts, enabling integration with economic data.

Value added, capital formation, and labour input are standard measures used to quantify economic activity in the System of National Accounts (SNA). These indicators are fully harmonised and form the basis for calculating GDP and productivity metrics. Gross Value Added (GVA) measures sectoral output net of intermediate inputs and is presented at basic prices excluding taxes and subsidies. Gross fixed capital formation (GFCF) tracks investment in fixed assets with a productive life over one year. Fixed assets include buildings, machinery, transport equipment, and intellectual property products such as software and research and development. Labour input captures employment in terms of jobs, hours worked, or persons employed, covering both employees and the self-employed.

Finance and investment

Low-carbon energy investment refers to the investments in technologies aimed at reducing carbon emissions. These include various renewable power generation, electricity grids, battery storage, low carbon fuels, energy efficiency end-use technologies, and other low-carbon energy sources:

• Solar PV: utility-scale, commercial, residential, and off-grid solar photovoltaic systems.

• Wind energy: onshore and offshore wind power projects.

• Nuclear power: new plants (greenfield), refurbishments, and upgrades for long-term operations.

• Other low-carbon energy sources: H2-based fuels (hydrogen and ammonia) used in energy generation, solar CSP, tidal, geothermal, hydro and pumped hydro storage, bioenergy and renewable waste.

• Low-carbon fuels liquid biofuels and biogas, hydrogen and hydrogen-based fuels for transport

• Electricity grids infrastructure development for transmission, distribution, and public EV charging stations.

• Battery storage, including the deployment of utility-scale and building-integrated battery storage solutions.

• Energy efficiency & other end-use: energy-efficient building materials, appliances, Heating, Ventilation, and Air Conditioning (HVAC) systems, smart meters, heat pumps, home energy management systems, Electric Vehicles (EVs), and EV charging infrastructure.

The estimates of investments for renewable power generation, electricity grids, and battery storage correspond to the annual capital spending on the development, deployment, and maintenance of clean energy technologies and infrastructure. This includes the spending carried out for new plants or the replacement of aging assets. Investment in biofuels is based on capacity expansion plans of production facilities and plant costs; and excludes additional spending on agricultural feed. Investments for energy efficiency is the incremental spending to acquire equipment that consumes less energy than would otherwise have been used to provide the services had the consumer did not purchase a more efficient option. For more information consult the IEA (2025[25]).

Fossil fuel investment refers to the investments in fossil fuel-based power generation, including those with carbon capture and storage (CCUS) projects, oil and gas supply (upstream, midstream pipelines and refining), and coal supply (mining and transportation).

The investment estimates refer to annual capital spending for new power plants or replacement of old assets. Investments outlays are spread evenly from the year of the final investment decision until the year the asset becomes operational. Investments in fuel supply is the total capital spending incurred in the given year. The spending for upstream oil and gas are based on the 90 major oil companies, which represent over three-quarters of global oil and gas production; oil refining sector is based on new and upgrading projects; oil and gas shipping (midstream sectors) are based on demand, supply, and trade in oil and gas products. For more information consult the IEA (2025[25])

Source of energy investment is defined as the type of economic entity that invests in an asset regardless of the original finance provider. There are three types of entities: public (including stated-owned companies or state-owned assets), corporates and households. Corporates and households are considered private sources for data aggregations in IEA energy investment statistics.

Green bonds are fixed-income financial instruments with proceeds earmarked for projects and assets that deliver environmental benefits. These bonds are labelled as such by the issuer for easier identification by investors.

These bonds can be issued by national governments (sovereign bonds), multi-national banks or corporations, or supranational entities (i.e. a financing agency backed by multiple governments). The issuing entity guarantees to repay the bond over a set period with either a fixed or variable rate of return. The Climate Bonds Standard and Certification Scheme is a voluntary self-labelling scheme for investments consistent with climate change mitigation goals. The certification is based on the Climate Bonds Taxonomy, which provides guidance on the criteria to identify assets, activities, and projects consistent with the goals of the Paris Agreement. For more information, see www.climatebonds.net/.

Environmentally related ODA refers to bilateral ODA annual commitments marked with a principal or significant environmental objective. These objectives are identified using at least one of the four “Rio markers” or the “Environment maker”. Environmentally related ODA covers DAC donors and is calculated by aggregating at the individual project-level data to avoid double-counting. It is expressed as a share of total bilateral sector-allocable ODA. Data are collected on a calendar year basis.

The four Rio markers and the Environment marker allow screening aid projects with cross-sectoral environmental objectives. They include:

• Climate change mitigation-related aid aims to reduce GHG emissions or enhance of sequestration.

• Climate change adaptation-related aid supports resilience to climate change impacts.

• Biodiversity-related aid promotes conservation and sustainable use of biodiversity.

• Desertification-related aid addresses land degradation and drought impacts.

• Environment marker, introduced before the Rio markers, includes activities intended to improve the physical or biological environment, and activities that integrate environmental concerns with a range of development objectives through institution building or capacity development.

Environmentally related sector-allocable ODA refers to bilateral aid commitments in sectors with direct environmental relevance, such as general environmental protection, water supply and sanitation, and renewable energy. It is expressed as a share of total bilateral sector-allocable ODA, which excludes general budget support, debt-related aid, humanitarian aid and internal transactions in the donor country. Sectors of environmental relevance include:

• Environment sector covers environmental policy and administrative management, biosphere protection, biodiversity, site preservation, flood prevention/control, environmental education, training and environmental research.

• The water supply and sanitation sector covers water sector policy and administrative management, water resources conservation, water supply and sanitation, basic drinking water supply and basic sanitation, river basin' development, waste management, education and training in water supply and sanitation.

• Renewable energy sector covers energy generation from hydro-electric power plants, geothermal energy, solar energy, wind power, ocean power and biomass.

The denominators used in the calculations for the ODA section is total bilateral sector-allocable aid. The sector allocations follow DAC purpose codes 11110 to 43082 (see purpose code list): Sector-level identification complements the marker-based approach and supports analysis by thematic area. Confidentiality restrictions apply to private flows at market terms (e.g. bank lending, direct investment), so reporting focuses on public aid flows.

Net ODA consists of disbursements of grants and loans made on concessional terms (net of repayments of principal) by DAC and non-DAC donors, expressed as percentage of gross national income (GNI). It measures financial flows to countries on the DAC list of ODA recipients, aimed at promoting economic development and welfare.

To qualify as ODA, loans must include a grant element of at least 25% (using a 10% discount rate). The longstanding UN target for ODA is 0.7% of donor GNI. Net ODA excludes private flows at market terms and focuses on official, public financing.For more information on ODA definitions and classifications please consult the methodologies for development finance reporting (OECD, 2024[50]).

Multifactor productivity

Environmentally adjusted multifactor productivity (EAMFP) growth measures the change in productivity at the country level, accounting for both desirable and undesirable outputs, as well as the use of factor inputs (including natural capital). EAMFP growth is expressed in percentage points of output growth and is derived using a growth accounting approach.

EAMFP captures the residual growth in the joint production of GDP, GHG emissions and air pollution, that cannot be explained by changes in the use of labour, produced capital, and natural capital inputs. Therefore, for a given growth of input use, EAMFP increases when GDP increases or when pollution decreases. EAMFP growth is a proxy measure of efficiency improvements in the use of inputs, technological change, institutional or organisational improvements.

The adjustment of output growth for pollution abatement is expressed in percentage points. It measures the extent in which GDP growth is scaled up or down to reflect pollution abatement efforts. This shadow cost of pollution abatement reflects the valued of resources diverted to pollution abatement, or the ‘excess' growth generated at the expense of environmental quality. Air emissions included in the adjustment account for five greenhouse gases (CO₂, CH₄, N₂O, NF₃, SF₆) and seven air pollutants (PM₁₀, CO, NMVOC, SO_X, NO_X, NH₃, BC).

The contribution of natural capital to output growth measures the extent in which a country's growth in output is due to natural resource use, including non-renewable and renewable natural resources, and ecosystem services. The contribution of natural capital is expressed in percentage points of output growth and is derived using a growth accounting approach.

Natural capital inputs in the EAMFP measurement framework include non-renewable resources (hard coal, brown coal, crude oil, natural gas and minerals such as bauxite, copper, gold, iron ore, lead, nickel, phosphate, silver, tin, zinc), renewable resources (cropland, pastureland, and forestland), two biological resources (non-cultivated timber and marine capture fisheries), and ecosystem services (watershed protection by forests, non-wood forest products, coastal protection by mangroves).

Renewable energy sources include hydroelectricity, wind energy and solar photovoltaics. The contribution of renewable energy is expressed in percentage points of output growth and is derived using a growth accounting approach. The contribution of renewable energy to output growth is shown separately because the valuation methodology is exploratory, and data are not available for all countries.

The inclusion of renewable energy entails an adjustment for all other metrics in the growth accounting framework. Therefore, results in box "Renewable energy remains a largely untapped source of income growth" are not directly comparable against results presented in the body of the section. For more details on concepts and the growth accounting methodology please consult (Cárdenas Rodríguez et al., 2023[29])

Policies and regulations

Environmentally Related Tax Revenue (ERTR) refers to government revenue from taxes levied on environmentally harmful tax-bases. These include taxes on energy products for transport and stationary purposes (e.g., petrol, diesel, electricity), transport (e.g., vehicle registration, vehicle import or sales taxes, road use), pollution (e.g., landfill, emissions to air or water, waste disposal, packaging, landfill), and natural resources (e.g., fishing, hunting, land conversion). Revenues from the auctioning of emission allowances (e.g. from the EU Emissions Trading System) are also included under energy taxes.

ERTR is shown as a share of total tax revenue and compared to GDP. Only payments classified as taxes under national accounts are included. The classification of environmentally related taxes follows OECD and Eurostat guidelines (see OECD (2023[51])), ensuring international comparability. The structure of the tax base (e.g., share of energy vs. transport taxes) is provided to support interpretation of fiscal instruments for environmental policy.

Environmentally related tax revenue accounts (ERTR accounts) record government revenue from environmentally related taxes disaggregated by economic activities and resident household activities. ERTR accounts are aligned with the System of Environmental-Economic Accounting Central Framework (SEEA-CF).

ERTR is allocated to resident economic and household activities, according to International Standard Industry Classification (ISIC) Rev. 4. Revenue not possible to disaggregate by specific ISIC classes, according to the tax-payer principle, is classified as ‘unallocated’. ERTR includes payments classified as taxes under national accounts, as well as the revenue raised by the auctioning of permits for tradable permit schemes. For more details consult OECD (2023[51]) and Eurostat (2024[52]) methodological guidelines.

Labour tax revenue refers to tax income collected by all levels of government (i.e., supranational, federal/central, state/regional, local) from labour-related sources. It includes taxes on income, profits and capital gains of individuals; social security contributions from employees, employers, self-employed or non-employed; and taxes on payroll and workforce. This follows the OECD classification of Taxes and ensures comparability across countries (OECD, 2023[53]). Labour tax revenue is used to assess the structure of the tax systems relative to environmentally related tax bases.

Feed-in tariffs (FITs) are policy instruments that promote renewable electricity by guaranteeing fixed prices to producers through long-term contracts. They apply to specified technologies and require grid operators to purchase electricity at regulated rates (OECD, 2018[54]).

FITs values reflect mandated payments set by laws, regulations, or power purchase agreements (PPAs). Reported values are annual country averages by technology subsector and are presented in nominal USD per kWh. FITs support investment certainty, induce innovation, and facilitate renewable energy deployment.

Environmental subsidies and similar transfers (ESST) accounts record government support measures aimed at environmental protection and resource management. These include subsidies, tax rebates, tax exemptions, credits, and deferrals that reduce the cost of environmentally friendly products and activities.

Compiled in line with the SEEA Central Framework and Eurostat guidelines, these accounts provide a breakdown by economic activity and beneficiary sector. They support the analysis of the scale and distribution of public financial support for environmental objectives. ESST accounts are compared with ERTR accounts to illustrate the complementarity of fiscal measures across sectors.

Government Budget Allocations for Research and Development (GBARD): Measure the funds that governments allocate to R&D to meet various socio-economic objectives, using the Nomenclature for the Analysis and Comparison of Scientific Programmes and Budgets (NABS 2007) classification.

Environmentally related government R&D budget: captures the share of public budget allocated to R&D activities aimed at environmental objectives. These are measured Government Budget Allocations for R&D (GBARD), representing government planned spending, rather than actual spending. In the NABS 2007 classification, the socio-economic objective “environment” includes R&D on pollution control and the development of tools to monitor, prevent, or eliminate pollution.

Renewable energy public RD&D budget: captures the public budget directed at research, development and demonstration (RD&D) of renewable energy technologies. These measures represent government planned spending, rather than actual spending. Renewable energy technologies include hydro, geothermal, solar (thermal and PV), wind and tide/wave/ocean energy, as well as combustible renewables (solid biomass, liquid biomass, biogas). It also covers RD&D directed at measuring, monitoring and verifying renewable energy technologies.

Fossil fuel public RD&D budget: Public budget directed at research, development and demonstration (RD&D) related to fossil fuels, including oil, gas and coal and excluding RD&D related to CO2 capture and storage (CCS) technologies. These measures represent government planned spending, rather than actual spending.

Public RD&D budget for energy covers research, development, and demonstration (RD&D) related to all forms (renewable and non-renewable) of energy production, storage, transportation, distribution, and rational use. RD&D budgets are reported from the funder's perspective as budgets. These measures represent government planned spending, rather than actual spending.

Bans and phase-outs send a clear policy signal and provide certainty for investors, firms and households to switch to less harmful alternatives. Bans refer to the prohibition of the purchase of or investment in new assets whereas phase-outs refer to the prohibition of using already existing asset. Bans and phaseouts are considered as adopted if they are enshrined in law or have been achieved, i.e., announcements unaccompanied by corresponding legislation are excluded. Bans and phase-outs include:

• Fossil fuel extraction: Ban or phase-out of (new) extraction fields for coal, oil or natural gas.

• Coal-fired power plants: Ban or phase-out of (the construction of new) unabated coal-fired power plants

• Oil heating in buildings ban or phase-out

• Gas heating in buildings ban or phase-out

• Passenger cars with ICE ban or phase-out of (new) passenger cars with internal combustion engines

• Landfill disposal ban of landfill disposal of combustible, organic, recyclable, or untreated waste.

• Export credits for coal: phase-out of governments’ export credits for new unabated coal power plants

• Public funding for fossil fuel infrastructure abroad phase-out of government finance for fossil fuel infrastructure investment abroad

Mandatory energy labels are information instruments that require products to display standardized information on energy consumption and efficiency. The indicator captures the existence of mandatory labelling schemes for five product categories: refrigerators, freezers, lighting, air conditioners, and passenger vehicles.

These labels inform consumers at the point of purchase and typically include efficiency ratings and other performance characteristics. The presence of such schemes reflects national efforts to promote energy efficiency in both the residential and transport sector.

Data sources

Employment and education

• OECD (2024), Twin Transition Tracker: Assessing Regional Resilience (Dashboard), https://www.oecd.org/en/data/dashboards/twin-transition-tracker-assessing-regional-resilience.html (accessed on 6 June 2025).

• OECD (2023), Education at a Glance 2023: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/e13bef63-en.

• OECD (2013), Education at a Glance 2013: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/eag-2013-en

• OECD (2025), Bibliometric indicators , Science, technology and innovation (database), http://data-explorer.oecd.org/s/2dx (accessed on July 2025).

Markets of goods and services

• OECD (2025), Patents – technology development, OECD Environment Statistics (database), https://doi.org/10.1787/data-00760-en (accessed on  July 2025).

• OECD (2025), “Patents - indicators”, OECD Environment Statistics (database), https://doi.org/10.1787/e478bcd5-en (accessed on  July 2025).

• OECD (2024), STI Micro-data Lab: Intellectual Property Database, http://oe.cd/ipstats (accessed on  November 2024).

• OECD (forthcoming), New Startup, OECD Science, Technology and Innovation (Database).

• Eurostat (2024), Air emissions accounts by NACE Rev. 2 activity (database), https://doi.org/10.2908/ENV_AC_AINAH_R2 (accessed on  July 2025).

• Eurostat (2024), Gross value added and income by detailed industry (NACE Rev.2 ) (database), https://doi.org/10.2908/NAMA_10_A64 (accessed on  July 2025).

• Eurostat (2024), Capital formation by industry (NACE Rev.2) and detailed asset type (database), https://doi.org/10.2908/NAMA_10_A64_P5 (accessed on  July 2025).

• Eurostat (2024), National accounts employment data by industry (up to NACE A*64) (database), https://doi.org/10.2908/NAMA_10_A64_E (accessed on  July 2025).

Finance and investment

• IEA (2025), World Energy Investment (datafile), IEA, Paris, https://www.iea.org/data-and-statistics/data-product/world-energy-investment-2025-datafile (accessed on July 2025).

• CBI (2025), Interactive data platform, Climate Bond Initiative, https://www.climatebonds.net/data-insights/market-data/insights-from-climate-bonds-data (accessed on  July 2025).

• OECD (2025), Credit Reporting System, International Development Statistics (database), https://data-explorer.oecd.org/s/2lt (accessed on  April 2025)

• OECD (2025), Total Flows by donor (ODA+OOF+Private), OECD International Development Statistics (database), http://data-explorer.oecd.org/s/25e (accessed on  April 2025).

Multifactor productivity

• OECD (2024), “Environmentally Adjusted Multifactor Productivity”, OECD Environment Statistics (database), https://doi.org/10.1787/55a11744-en (accessed on July 2025).

Policies and regulations

• OECD (2025), “Environmental policy: Environmentally related tax revenue”, OECD Environment Statistics (database), https://doi.org/10.1787/df563d69-en (accessed on  August 2025).

• OECD (2024), “Environmentally related tax revenue accounts”, OECD Environment Statistics (database), https://data-explorer.oecd.org/s/2lu (accessed on  January 2025).

• Eurostat (2024), Environmental taxes by economic activity (NACE Rev. 2) (database), http://data.europa.eu/88u/dataset/nsnav8yoesmt84kovybphq (accessed on July 2025).

• OECD (2025), “Revenue statistics”, OECD Tax statistics (database), https://data-explorer.oecd.org/s/2lv (accessed on  July 2025).

• Eurostat (2024), “Environmental subsidies and similar transfers from general government to corporations, by environmental activity, ESA category of transfer and NACE Rev. 2 activity of recipient” (database), http://data.europa.eu/88u/dataset/2cmlikxnnvx7ap0shp2oqq (accessed on  July 2025).

• OECD (2025), Government budget allocations for R&D, OECD Research and Development Statistics (database), https://data-explorer.oecd.org/s/2ls (accessed on July 2025)

• IEA (2025), Energy Technology RD&D Budgets (database), https://www.iea.org/data-and-statistics/data-product/energy-technology-rd-and-d-budget-database-2, (accessed on July 2025).

• OECD (forthcoming), “Climate Actions and Policies Measurement Framework (CAPMF)”, OECD Environment Statistics (database), https://data-explorer.oecd.org/s/2lx .

References and further reading

[65] Amoroso, S. (2021), World Corporate Top RandD investors: Paving the way for climate neutrality, Publications Office of the European Union, https://doi.org/10.2760/49552.

[61] Armitage, C., M. Lorenz and S. Mikki (2020), “Mapping scholarly publications related to the Sustainable Development Goals: Do independent bibliometric approaches get the same results?”, Quantitative Science Studies, Vol. 1/3, pp. 1092-1108, https://doi.org/10.1162/qss_a_00071.

[34] Bellmann, C. and A. Bulatnikova (2022), “Incorporating environmental provisions in regional trade agreements in chapters and articles dealing with trade in services”, OECD Trade and Environment Working Papers, No. 2022/01, OECD Publishing, Paris, https://doi.org/10.1787/6e976798-en.

[56] Berestycki, C. et al. (2022), “Measuring and assessing the effects of climate policy uncertainty”, OECD Economics Department Working Papers, No. 1724, OECD Publishing, Paris, https://doi.org/10.1787/34483d83-en.

[67] Borgonovi, F. et al. (2022), “The environmental sustainability competence toolbox: From leaving a better planet for our children to leaving better children for our planet”, OECD Social, Employment and Migration Working Papers, No. 275, OECD Publishing, Paris, https://doi.org/10.1787/27991ec0-en.

[26] Brandt, N., P. Schreyer and V. Zipperer (2014), “Productivity Measurement with Natural Capital and Bad Outputs”, OECD Economics Department Working Papers, No. 1154, OECD Publishing, Paris, https://doi.org/10.1787/5jz0wh5t0ztd-en.

[27] Brandt, N., P. Schreyer and V. Zipperer (2013), “Productivity Measurement with Natural Capital”, OECD Economics Department Working Papers, No. 1092, OECD Publishing, Paris, https://doi.org/10.1787/5k3xnhsz0vtg-en.

[28] 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 Publishing, Paris, https://doi.org/10.1787/fdd40cbd-en.

[29] Cárdenas Rodríguez, M. et al. (2023), “Environmentally adjusted multifactor productivity: Accounting for renewable natural resources and ecosystem services”, OECD Green Growth Papers, No. 2023/01, OECD Publishing, Paris, https://doi.org/10.1787/9096211d-en.

[68] CBI (2025), Sustainable Debt Global State of the Market 2024, Climate Bonds Initiative, https://www.climatebonds.net/files/documents/publications/Climate-Bonds-Initiative_Global-State-of-the-Market-Report_May-2025_2025-06-18-123430_mejk.pdf.

[18] Dechezleprêtre, A. and P. Kelly (forthcoming), The New Green Economy: Venture Capital, Innovation and Business Success in Cleantech Startup, STI Working paper.

[52] Eurostat (2024), Environmental taxes – A Statistical Guide, Luxembourg: Publications Office of the European Union, https://doi.org/10.2785/730717.

[46] Garcia-Valero, F., M. Faria and N. Wajsman (2021), Green EU trade marks – Analysis of goods and services specifications, 1996-2020, European Union Intellectual Property Office (EUIPO), https://doi.org/10.2814/900650.

[66] Garsous, G. (2019), “Trends in policy indicators on trade and environment”, OECD Trade and Environment Working Papers, No. 2019/01, OECD Publishing, Paris, https://doi.org/10.1787/b8d2bcac-en.

[49] Gaulier, G. and Zignago, S (2010), BACI: International Trade Database at the Product-Level, CEPII Working Paper, N°2010-23. BibTex, https://www.cepii.fr/CEPII/en/bdd_modele/bdd_modele_item.asp?id=37.

[33] George, C. and S. Yamaguchi (2018), “Assessing Implementation of Environmental Provisions in Regional Trade Agreements”, OECD Trade and Environment Working Papers, No. 2018/01, OECD Publishing, Paris, https://doi.org/10.1787/91aacfea-en.

[43] Haščič, I. and M. Migotto (2015), “Measuring environmental innovation using patent data”, OECD Environment Working Papers, No. 89, OECD Publishing, Paris, https://doi.org/10.1787/5js009kf48xw-en.

[25] IEA (2025), World Energy Investment 2025, IEA Publishing, https://iea.blob.core.windows.net/assets/1c136349-1c31-4201-9ed7-1a7d532e4306/WorldEnergyInvestment2025.pdf.

[44] IEA (2021), Methodology for identifying fossil fuel supply related technologies in patent data, EPO and OECD/IEA, http://epo.org/trends-energy.

[45] IEA (2021), Patents and the Energy Transition, IEA, Paris, https://www.iea.org/reports/patents-and-the-energy-transition.

[47] Jayabalasingham, B. (2019), Identifying research supporting the United Nations Sustainable Development Goals, https://doi.org/10.17632/87txkw7khs.1.

[14] Jeong, S., B. Sherman and D. Tippins (2021), “The Anthropocene as we know it: posthumanism, science education and scientific literacy as a path to sustainability”, Cultural Studies of Science Education, Vol. 16/3, pp. 805-820, https://doi.org/10.1007/s11422-021-10029-9.

[30] Johnstone, N. (2010), “Renewable Energy Policies and Technological Innovation: Evidence Based on Patent Counts.”, Environ Resource Econ, Vol. 45, pp. 133–155, https://doi.org/10.1007/s10640-009-9309-1.

[36] Kruse, T. et al. (2022), “Measuring environmental policy stringency in OECD countries: An update of the OECD composite EPS indicator”, OECD Economics Department Working Papers, No. 1703, OECD Publishing, Paris, https://doi.org/10.1787/90ab82e8-en.

[20] Moïsé, E. and E. Tresa (2025), “Non-Tariff Measures Affecting Environmental Goods Trade”, OECD Trade Policy Papers, No. 294, OECD Publishing, Paris, https://doi.org/10.1787/ca790510-en.

[35] Nachtigall, D. et al. (2022), “The climate actions and policies measurement framework: A structured and harmonised climate policy database to monitor countries’ mitigation action”, OECD Environment Working Papers, No. 203, OECD Publishing, Paris, https://doi.org/10.1787/2caa60ce-en.

[8] Nusche, D., M. Fuster Rabella and S. Lauterbach (2024), “Rethinking education in the context of climate change: Leverage points for transformative change”, OECD Education Working Papers, No. 307, OECD Publishing, Paris, https://doi.org/10.1787/f14c8a81-en.

[59] OECD (2025), Bibliometric indicators , Science, technology and innovation (database), http://data-explorer.oecd.org/s/2dx (accessed on  July 2025).

[23] OECD (2025), Climate FInance and the USD 100 billion goal, https://www.oecd.org/en/topics/sub-issues/climate-finance-and-the-usd-100-billion-goal.html.

[24] OECD (2025), Environment at a Glance Indicators, OECD Publishing, Paris, https://doi.org/10.1787/ac4b8b89-en.

[69] OECD (2025), “Measuring Science and Innovation for Sustainable Growth – Chapter 2 (forthcoming)”.

[58] OECD (2025), OECD Data Explorer, https://data-explorer.oecd.org.

[17] OECD (2025), PISA 2025 Science Framework, https://pisa-framework.oecd.org/science-2025/.

[41] OECD (2025), Twin Transition Tracker: Assessing regional resilience: Technical Background Paper, https://www.oecd.org/content/dam/oecd/en/data/dashboards/twin-transitions-tracker/OECD-TTT-Technical-Background-paper.pdf/_jcr_content/renditions/original./OECD-TTT-Technical-Background-paper.pdf.

[16] OECD (2024), “Cultivating the next generation of green and digital innovators: The role of higher education”, OECD Education Policy Perspectives, No. 95, OECD Publishing, Paris, https://doi.org/10.1787/bb6e432e-en.

[50] OECD (2024), Development finance statistics: Resources for reporting, http://www.oecd.org/en/data/insights/data-explainers/2024/10/resources-for-reporting-development-finance-statistics.html.

[7] OECD (2024), Education at a Glance 2024: OECD Indicators, OECD Publishing, Paris, https://doi.org/10.1787/c00cad36-en.

[13] OECD (2024), “Empowering young people through climate literacy”, https://www.oecd.org/content/dam/oecd/en/publications/reports/2024/11/ClimateBrochure_Nov2024_FIN.pdf/_jcr_content/renditions/original./ClimateBrochure_Nov2024_FIN.pdf.

[6] OECD (2024), OECD Employment Outlook 2024: The Net-Zero Transition and the Labour Market, OECD Publishing, Paris, https://doi.org/10.1787/ac8b3538-en.

[21] OECD (2024), OECD Review on Aligning Finance with Climate Goals: Assessing Progress to Net Zero and Preventing Greenwashing, Green Finance and Investment, OECD Publishing, Paris, https://doi.org/10.1787/b9b7ce49-en.

[64] OECD (2024), Pricing Greenhouse Gas Emissions 2024: Gearing Up to Bring Emissions Down, OECD Series on Carbon Pricing and Energy Taxation, OECD Publishing, Paris, https://doi.org/10.1787/b44c74e6-en.

[32] OECD (2024), “The IFCMA’s Climate Policy Database: Policy instruments typology and data structure”, Inclusive Forum on Carbon Mitigation Approaches Papers, No. 5, OECD Publishing, Paris, https://doi.org/10.1787/68529f35-en.

[60] OECD (2023), Education Policy Outlook 2023: Empowering All Learners to Go Green, OECD Publishing, Paris, https://doi.org/10.1787/f5063653-en.

[12] OECD (2023), Job Creation and Local Economic Development 2023: Bridging the Great Green Divide, OECD Publishing, Paris, https://doi.org/10.1787/21db61c1-en.

[51] OECD (2023), Methodological Guidelines for Environmentally Related Tax Revenue Accounts, OECD Publishing, Paris, https://doi.org/10.1787/d752d120-en.

[11] OECD (2023), OECD Science, Technology and Innovation Outlook 2023: Enabling Transitions in Times of Disruption, OECD Publishing, Paris, https://doi.org/10.1787/0b55736e-en.

[10] OECD (2023), OECD Skills Outlook 2023: Skills for a Resilient Green and Digital Transition, OECD Publishing, Paris, https://doi.org/10.1787/27452f29-en.

[63] OECD (2023), Policy Perspectives - OECD work on Regional Trade Agreements and the environment, https://issuu.com/oecd.publishing/docs/rtas_pp_web-2.

[53] OECD (2023), Revenue Statistics: Interpretatice Guide, https://www.oecd.org/content/dam/oecd/en/topics/policy-sub-issues/global-tax-revenues/oecd-classification-taxes-interpretative-guide.pdf.

[42] OECD (2018), OECD Handbook for Internationally Comparative Education Statistics 2018: Concepts, Standards, Definitions and Classifications, OECD Publishing, Paris, https://doi.org/10.1787/9789264304444-en.

[54] OECD (2018), Renewable energy feed-in tariffs Methodological note for data compilation.

[4] OECD (2017), Green Growth Indicators 2017, OECD Green Growth Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264268586-en.

[22] OECD (2017), Mobilising Bond Markets for a Low-Carbon Transition, OECD Publishing, Paris, https://doi.org/10.1787/9789264272323-en.

[70] OECD (2015), Frascati Manual 2015: Guidelines for Collecting and Reporting Data on Research and Experimental Development, The Measurement of Scientific, Technological and Innovation Activities, OECD Publishing, https://doi.org/10.1787/9789264239012.

[3] OECD (2014), Green Growth Indicators 2014, OECD Green Growth Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264202030-en.

[57] OECD (2011), Towards Green Growth, OECD Green Growth Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264111318-en.

[2] OECD (2011), Towards Green Growth: Monitoring Progress: OECD Indicators, OECD Green Growth Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264111356-en.

[19] OECD (2009), OECD Patent Statistics Manual, OECD Publishing, Paris, https://doi.org/10.1787/9789264056442-en.

[5] OECD (n.d.), “Green growth indicators”, OECD Environment (database), https://doi.org/10.1787/data-00665-en.

[1] Rockström, J. (2009), “Planetary Boundaries: Exploring the Safe Operating Space for Humanity”, http://www.ecologyandsociety.org/vol14/iss2/art32/.

[71] Röttgers, D. (2018), Power struggle: Decarbonising the electricity sector, OECD Environment Working Papers, No. 139, OECD Publishing, https://doi.org/10.1787/900f4c72-en.

[48] Sauvage, J. (2014), “The Stringency of Environmental Regulations and Trade in Environmental Goods”, OECD Trade and Environment Working Papers, No. 2014/3, OECD Publishing, Paris, https://doi.org/10.1787/5jxrjn7xsnmq-en.

[55] SEEA Central Framework (2012), System of environmental-economic accounting: a central framework.

[31] Stechemesser, A. et al. (2024), “Climate policies that achieved major emission reductions: Global evidence from two decades”, Science, Vol. 385/6711, pp. 884-892, https://doi.org/10.1126/science.adl6547.

[37] Stiglitz, J., J. Fitoussi and M. Durand (eds.) (2018), For Good Measure: Advancing Research on Well-being Metrics Beyond GDP, OECD Publishing, Paris, https://doi.org/10.1787/9789264307278-en.

[38] United Nations, E. (ed.) (forthcoming), System of National Accounts 2025, United Nations, https://unstats.un.org/unsd/nationalaccount/.

[39] Vona, F., G. Marin and D. Consoli (2018), “Measures, drivers and effects of green employment: evidence from US local labor markets, 2006–2014”, Journal of Economic Geography, Vol. 19/5, pp. 1021-1048, https://doi.org/10.1093/jeg/lby038.

[40] Vona, F. et al. (2018), “Environmental Regulation and Green Skills: An Empirical Exploration”, Journal of the Association of Environmental and Resource Economists, Vol. 5/4, pp. 713-753, https://doi.org/10.1086/698859.

[9] White, P. et al. (2023), “Agency in the Anthropocene: Supporting document to the PISA 2025 Science Framework”, OECD Education Working Papers, No. 297, OECD Publishing, Paris, https://doi.org/10.1787/8d3b6cfa-en.

[62] Yamaguchi, S. (2020), “Greening regional trade agreements on investment”, OECD Trade and Environment Working Papers, No. 2020/03, OECD Publishing, Paris, https://doi.org/10.1787/4452a09d-en.

[15] Zoller, U. (2012), “Science Education for Global Sustainability: What Is Necessary for Teaching, Learning, and Assessment Strategies?”, Journal of Chemical Education, Vol. 89/3, pp. 297-300, https://doi.org/10.1021/ed300047v.