This chapter starts with a view on the direct drivers of climate change, biodiversity loss and pollution, their underlying socio-economic trends, and interactions under current policies. The chapter then unpacks these drivers and provides quantifications of past and future trends of key drivers from a supply- and demand-side perspective, revealing the critical role of efficiency and related resource use. Finally, it provides a unified outlook on the evolution of key indicators that drive the triple planetary crisis in the absence of more stringent policies.
Environmental Outlook on the Triple Planetary Crisis
Stakes, Evolution and Policy Linkages
Abstract
This chapter explores the common underlying drivers of the triple planetary crisis and their evolution over time under current policies. It builds on the general analytical framework of driver-pressure-state-impact-response (DPSIR) described in Chapter 1 and applies it to assess how the pressures on the environment can be explained by changes in their drivers.
The drivers of the triple planetary crisis are multifaceted. Some drivers are direct: for example, driving a car with an internal combustion engine emits greenhouse gases (GHGs) and contributes to air pollution. Other drivers are indirect: population growth increases demand for food and other commodities, and the production of these goods leads to increased pressures on the environment. To unpack these drivers, it is helpful to distinguish between broad socioeconomic trends and specific drivers. Socioeconomic trends – such as population growth – are the primary drivers for changes in economic activity, which in turn affect environmental pressures (Figure 2.1).1 These links are further influenced by several crosscutting aspects, such as resource use. Figure 2.1 presents the links explored in this chapter, as represented in the Environmental Outlook modelling toolbox.2
As is also considered in certain strands of the literature (e.g. Riahi et al (2017[1])), and as detailed in Chapter 1, the projections provided in this chapter exclude feedback effects of environmental degradation. These projections form the reference point against which assessments of impacts can be made. The modelling in this report provides projections of how drivers affect pressures, and subsequently states, impacts and responses. Assessing feedbacks from environmental degradation is important for cost-benefit analysis of environmental policy, but the indirect effects of feedbacks on environmental pressures and thus states is much more limited.
Under current environmental policies, and in the absence of feedback effects of environmental degradation on these trends and drivers, the projected global evolution of these main socioeconomic trends can be summarised as:
Populations continue to grow in many countries, albeit at a slower global pace than in the past.
Per capita income levels converge partially, although major differences between regions persist.
Labour productivity growth differs by sector and continues to drive economic growth across all regions.
Services become an increasing part of the global Gross Domestic Product (GDP).
Current environmental – and related – policies continue affecting the links between economic activity and environmental pressures, i.e. environmental pressures do not scale proportionally with economic activity.
The key data sources used to make these projections are summarised in Annex 2.A.
Note: The projections presented in this chapter are under current policies and thus exclude feedback effects of environmental degradation. Therefore, feedbacks from environmental degradation from pressures are not shown in this infographic.
Source: Authors’ own elaboration.
The economic drivers of environmental pressures can be grouped into three aspects: (i) production and supply, (ii) composition of demand (including trade), and (iii) efficiency (or productivity). Quantifying these aspects entails determining the scale of economic activities, the structure of these activities and the intensity with which they lead to environmental pressures (this is usually done through decomposition of changes in pressures, see e.g. (Kaya and Yokobori, 1997[2]; Grossman and Krueger, 1991[3]). Efficiency is largely driven by technological change and is sometimes referred to as a technique effect.
While all economic activity results in some degree of environmental pressure, two key sectors have significant impacts across climate change, biodiversity loss and pollution: (i) energy and (ii) agriculture and land use. Other economic activities also contribute to the pressures, including, for example, polluting mining activities, production and use of plastics and other chemicals, and industrial processes that emit GHGs beyond fossil fuel combustion. Therefore, the following sections explore the production, composition and efficiency aspects at the sectoral level. It should be noted that underlying economic and institutional factors, including market failures, influence the evolution of environmental pressures, although these are difficult to quantify and project, as discussed in Box 2.1.
While analysing the socioeconomic trends and specific drivers described above is critical to identifying the causes of and solutions to the triple planetary crisis, considering underlying economic and institutional factors behind these drivers is also essential.
Economic systems can over-exploit limited resources and fail to supply adequate public goods in cases of market failures. Negative externalities such as the damages caused by particulate matter (PM) emissions and positive externalities such as the ecosystem services provided by natural habitats need to be accounted for. If negative externalities are not correctly priced, the associated economic activities do not reflect the full cost of their use leading to emissions of pollutants (and other environmental pressures) and therefore environment degradation at a cost for society.
Some polluting activities benefit from public subsidies to achieve a particular socio-economic outcome, like protecting jobs. Environmentally harmful subsidies, like certain types of support to livestock or reduced taxation of fossil fuels, can worsen environmental pressures (Lankoski, Nales and Valin, 2025[4]; OECD, 2022[5]; Iovino, 2023[6]). Better pricing of carbon emissions could contribute to reduce carbon emissions at a level that is consistent with limiting global warming to well below 2°C by 2050.
Other market failures are due to the public or common good nature of planetary systems. A stable climate is a global public good, everyone has access to it and the use by one individual does not affect its availability for others. Free riding can undermine global climate change mitigation efforts as the implementation of carbon mitigation policies in a group of countries benefits other countries as well. As mitigating climate change comes at a cost, free riding incentivises countries to let other countries incur this cost while still benefitting from mitigation. International agreements can help countries to work together towards common goals and combat free riding.
Biodiversity is closely linked to the global commons, including oceans and tropical rainforests, that provide global benefits but are de facto open access resources and very hard to manage sustainably. Some are international by definition (e.g. high seas, polar regions, pristine boreal area), while others are under national jurisdiction where national rather than global objectives determine the way they are exploited. The unsustainable overexploitation of fish stocks in international waters illustrates the potential complexities to manage shared resources and the need for international co-operation.
Another important market failure comes from the fact that policies that mitigate climate change, conserve biodiversity and reduce pollution often deliver benefits and costs over long periods of times. For example, if a tonne of carbon dioxide (CO2) is emitted now into the atmosphere, it will stay in the atmosphere and impacts the global climate for many years (IPCC, 2021[7]). Current generations may favour immediate rewards and heavily discount future payoffs beneficial for future generations. This leads to current policies that are not ambitious enough and postpone a problem that is getting worse over time.
This chapter presents projections to 2050 for the drivers of climate change, biodiversity loss and pollution. Section 2.2 unpacks the different drivers, focusing especially on energy, agriculture and land use and other key sectors including plastics and other chemicals. It does this in three steps, in line with the middle part of Figure 2.1: (i) the evolution of total production (reflecting the overall size of these economic activities), (ii) the composition of demand (reflecting changing preferences and structural change), and (iii) the role of efficiency (including technological change) that decouples economic activity from environmental pressures. Section 2.3 then links these discussions on drivers to the cross-cutting aspects of resource use (material resources and water). In order to identify the importance of the different drivers for changes in environmental pressures, Section 2.4 brings the various elements together to provide a unified outlook on biodiversity loss, climate change and pollution.
While the main drivers of the triple planetary crisis have been studied individually (see for instance reports from the IPCC (2023[8]) or the IPBES (2024[9])), the purpose of this section is to look at cross-cutting perspectives common to or linking the aspects of the triple planetary crisis. Sectoral emphases allow to disentangle the effects of the overall size of economic activity (total production and supply), the composition of demand (including structural change), and technology factors, i.e. efficiency.
Several key dynamics, often sectoral, are chosen to highlight the interlinkages among each aspect of the triple planetary crisis. These include a focus on (i) agriculture and land use, including urbanisation, (ii) energy, including the reliance on fossil fuels and electrification and (iii) other key sectors including chemicals and plastics. These sectoral emphases make it possible to disentangle the effects of the overall size of economic activity (total production and supply), the composition of demand (including structural change), and technology factors, i.e. efficiency. The following subsections are structured accordingly, going through these three effects (production and supply, composition of demand and the pivotal role of efficiency) for each of these three sectoral focuses. The choice to focus the analysis on a broad sectoral perspective rather than on a global lifecycle assessment is explained in Box 2.2.
The evolution of the triple planetary crisis is fundamentally driven by a number of socioeconomic trends, including population and income growth, technological developments and current policies. Together, these socioeconomic trends have consequences for the evolution of environmental pressures that are the result of multiple, opposing trends, even under a business-as-usual scenario. These trends – and their underlying factors – are described in more detail in Annex 2.A. The world population is expected to continue to grow and reach 8.5 billion by 2030 and 9.6 billion by 2050. Global economic activity, measured through GDP, is projected to grow substantially faster than population: on average by 2.7% per year, versus 0.7% for population. Thus, global GDP (at constant prices) is projected to grow from USD 126 trillion (tln) in 2020 to USD 283 tln in 2050. Productivity increases lead to changes in the structure of GDP, as measured by the contribution of value added generated in the different sectors to economy-wide value added generated. For least developed countries, agriculture represents a large share, while production in emerging economies is characterised by industrialisation, including a significant growth in construction to support infrastructure development. Sectoral shares are more stable in developed economies, with high levels of services and relatively low levels of agricultural production. As countries reach different stages in their developments, the structure of the economy evolves accordingly, taking into account the countries’ specificities.
Lifecycle assessments (LCAs) are methods used to quantify the net impacts of products throughout their lifecycles (from materials extraction to production and then transport, use and end of life). LCAs enable direct comparisons of environmental impacts per “functional unit” of products within the context of the triple planetary crisis (Hellweg et al., 2023[10]).
For instance, LCA analysis allows to compare the environmental impacts of grocery bag alternatives. Considering a functional unit, such as grocery items carried per bag, one can assess which grocery bag type is least damaging in terms of climate change, biodiversity loss and pollution. Paper bags emit five times more GHGs than single-use plastic bags (Meng, Brandão and Cullen, 2024[11]). While this means that paper bags are more harmful for climate change, improper disposal of plastic bags contributes significantly to global pollution (OECD, 2022[12]). The number of times the bags can be reused also alter their environmental impacts. Adding more complexity to the analysis, other alternatives such as biobased plastic bags require more land to be produced, contributing to biodiversity loss through deforestation (OECD, 2022[12]). Similarly, Tacket et al. (2025[13]) estimate that 68% of the considered alternatives to polyethylene packaging on the European market emit more GHG emissions (while 26% emit less).
Although LCAs offer a critical toolkit to identify synergies and trade-offs in product and material choices, they are more useful at a granular level. At the global level, the LCA of a given product is limited by data availability and current scientific knowledge. With countless product varieties, gathering data on each stage of the product lifecycle may not be feasible at the global scale. In the grocery bag example for instance, there is limited scientific knowledge on the impacts of plastic ingestion and entanglement by affected species (Woods et al., 2021[14]). Also, the same product can be produced using different processes. For instance, transport emissions associated to wood used in the production of paper bags will vary depending on the distance between the production plant and the wood source. As a result, estimates of transport emissions for the same product could vary drastically within and across countries. This observation can be generalised for any lifecycle stage and impact evaluated.
While previous OECD reports such as the Global Materials Resources Outlook to 2060 (OECD, 2019[15]) and the Global Plastics Outlook (OECD, 2022[12]) adopted global LCAs to evaluate selected material and plastic polymer lifecycles, a full LCA is beyond the scope of this report due to lack of data to make a robust assessment. Nevertheless, conceptually, the modelling tools used in this chapter can be interpreted as top-down LCAs as they offer evidence-based and multi-dimensional insights into the environmental consequences of different stages of production lifecycles. Environmental impacts from materials extraction, production, transport and end of life are modelled at sector and region levels within the modelling toolbox developed for this Outlook.
2.2.1. Production and supply of goods and services
Agriculture, forestry and other land uses
Agriculture occupies a complex and important role for climate change, biodiversity loss and pollution. While it remains essential for human systems, providing food to a growing population and offering jobs and livelihoods to over a billion people (FAO, 2023[16]), agriculture also generates sizeable negative externalities. It is the largest source of methane (CH4) and nitrous oxide (N2O) emissions, the main contributor of nutrient pollution in water bodies, and the primary driver of land use changes, which in turn drives terrestrial biodiversity loss. Agriculture development is closely linked to land use change, cropland expansion in particular. Beyond land, agriculture also consumes substantial resources, including water and energy, further straining ecosystems and natural resources. But the sector also holds considerable mitigation potential, such as through reduction of deforestation, improved livestock practices and carbon sequestration options (IPCC, 2023[8]), which remain untapped under current policies. Moreover, agriculture is highly dependent on healthy ecosystems and natural processes, making it particularly vulnerable to environmental degradation.
Current policies won’t unlock the potential positive contribution of agriculture, forestry and other land uses
Agricultural activities are one of the main sources of global CH4 emissions (43% of global emissions in 2020), mainly from enteric fermentation from livestock and rice cultivation, and the main source of N2O emissions (76% of global emissions in 2020), due to fertiliser use and manure management (Figure 2.2). More generally, Agriculture, Forestry and Other Land Uses (AFOLU) are responsible for one-fifth of all GHG emissions, mainly CO2, CH4 and N2O. At the same time, the mitigation potential from agricultural activities is estimated to be large. Reduction in non-CO2 emissions from agriculture, particularly CH4, is key but costly (OECD, 2025[17]). The potential for carbon sequestration is larger and mainly arises from carbon management in croplands and grasslands, as well as agroforestry. The IPCC (2023[8]) finds that AFOLU activities have the largest potential to reduce emissions or to sequester carbon, estimated at 8-14 gigatonnes of carbon dioxide equivalent (Gt CO2e) per year (for a carbon price below USD 100 per tonne CO2e). Through the deployment of nature-based solutions with safeguards (e.g. bioenergy plantations and afforestation) over a third of the required climate change mitigation required to keep climate warming below 2°C can be achieved (IPBES, 2019[18]). Other negative impacts include pollution from pesticide use, water pollution, air pollution and water resource depletion (Lankoski, Nales and Valin, 2025[4]).
Current policies are expected to be unable to unlock the potential for emissions reductions and sequestration from AFOLU activities.3 Emissions are projected to continue increasing, particularly CH4 and N2O in the middle-income region, due to a high growth in ruminant production (OECD/FAO, 2024[19]). However, with current policies, the overall emission intensity is projected to decrease, with agricultural production growing faster than emissions, through the improvement of management practices and technology. The emission intensity is expected to decline more significantly in production systems that are currently more emissions intensive, compared to systems and regions where substantial efforts have already been made.
Global emissions in Gt CO2e
Note: Emissions from AFOLU are net emissions. Scales differ by panel.
Source: Environmental Outlook modelling toolbox.
The use of synthetic fertilisers – in addition to improved farm management practices – to boost crop production can help limit agricultural land expansion and the associated environmental impacts.4 However, fertiliser use also contributes significantly to the negative externalities generated by agriculture (see also Chapter 6). Beyond emitting GHGs, fertilisers are also the main source of nutrient pollution in water bodies (see also Chapter 3). When more nutrients are applied on the land than can be taken up by plants (nutrient uptake), this generates a nutrient surplus that often leaks into the surrounding environment (for example through runoff into water bodies), generating nutrient pollution.
Under current policies, both nitrogen and phosphorus surpluses are projected to increase by 2050. In 2020, global nitrogen and phosphorus surpluses on all agricultural land, i.e. including both cropland and grassland, amounted to 129 and 11 million tonnes, respectively, with over 78% of nitrogen and 92% of phosphorus surplus reported in the middle- and lower-income regions (Figure 2.3).5 Fertiliser intensity (fertiliser use per hectare) was lower in these regions compared to the higher-income region, as the total agricultural land is also significantly higher than in the higher-income region (around 1.5 bln ha in lower-income, 2.2 bln ha in middle-income and 1.0 bln ha in higher-income region). Nonetheless, nutrient surpluses remained high. This reflects differences in agricultural area and in nutrient use efficiency, which are affected by factors such as agronomic practices, soil conditions and access to agricultural technologies. Agricultural areas in the higher-income region are roughly stable and increasing in the middle- and lower-income regions (to 1.6 and 2.4 bln ha in 2050 respectively). These regions also face a pressing need to increase yields to meet future food demands, which adds complexity. Under current policies, nitrogen surpluses are expected to increase by 2050, particularly in the lower- and middle-income regions (38% for nitrogen compared to 2020), while phosphorus surpluses are projected to increase by 59% in the lower-income region.6 In parallel, in many high-income countries, nutrient surpluses have decreased over the past decades, and despite a slowdown, there is no expectation that the trend would reverse (OECD, 2025[20]). The overall future trend of rising fertiliser use underscores the need for further policies promoting efficient nutrient use to boost crop production while protecting the environment (as discussed in Chapter 6). Other limiting factors should be addressed to optimise crop production, such as efficient water use (irrigation) and soil conditions (e.g. micronutrients and organic matter).
Along with other actions such as reducing food loss and waste as well adopting lower-impact consumption patterns such as moving towards plant-based diets (FAO, 2020[21]), intensification (i.e. increasing agricultural production per hectare) remains key in reducing the conversion of natural land into agricultural land, while continuing to produce food for a growing population. Although limiting land expansion is key for agriculture, intensification of production should not come at the expense of other sustainability objectives, which highlights the importance of the concept of sustainable productivity growth (OECD, 2024[22]).
Million tonnes of nutrients on agricultural land (bars, left axes) and million hectares of agriculture land (dots, right axes)
Note: Scales for nutrient surpluses (on the left axes) differ for nitrogen and phosphorus. Agricultural land includes cropland and grassland.
Source: Environmental Outlook modelling toolbox.
Land use change, driven by agricultural land expansion, is the main driver of biodiversity loss
Land use and land use change are the main drivers of biodiversity loss, as well as both a large source and sink of GHG emissions, and underpin food production. Winkler at al. (2021[23]) estimate that land use change has affected almost a third of the global land area in just six decades (1960-2019). These changes result from geographically diverging processes, with afforestation and cropland abandonment in the Global North and deforestation and agricultural expansion in the South.
Precise measurement of land use changes is challenging, and measures of land covers are not unanimous (Li et al., 2018[24]). Overall, the trends are clear, but not their order of magnitude: between 1992 and 2015, databases confirm that global forested area significantly decreased, while cropland expanded. Li et al. (2018[24]) report a decrease in global forested areas by around 600 million hectares (Mha), close to estimates by Hurtt et al. (2011[25]), while other sources find a decrease which is double or more, larger than 1200 Mha (Houghton and Nassikas, 2017[26]; Hansen et al., 2013[27]). Thus, current deforestation rates could range from 1.6% to 3.1% compared to an estimated at over 3800 Mha in 2020. More recently, FAO (2024[28]) find an 80 Mha cropland expansion whilst forest land area decreases by 100 Mha between 2001 and 2022. Over time, FAO (2024[29]) observes that deforestation rates have declined from 15.8 Mha per year in 1990-2000 to 10.2 million hectares per year in 2015-2020. These aggregate numbers are net impacts. For instance, between 2000 and 2012, Brazil’s well-documented reduction in deforestation was offset by increasing forest loss in Indonesia, Malaysia, Paraguay, Bolivia, Zambia, Angola, and elsewhere, according to Hansen et al. (2013[27]).
Striking a balance between intensification and extensification, linking to the debate between land sparing (minimising demand for farmland by increasing yields) and land sharing (favouring biodiversity on cultivated land but decreasing yields), is central to limiting the degradation of nature.7 Under current policies, with no increase in protected areas, agriculture is projected to remain the main driver of land expansion between 2020 and 2050, accounting for 87% of the increase in land use (Figure 2.4) through cropland and pasture expansion. However, the increase in agricultural land (meaning in cropland and pasture) remains small in relative terms (+5%), indicating an overall intensification of agricultural production, which reduces pressure on natural land but brings its own environmental impacts locally.
Regional differences in land use change, measured in Mha, are important (Figure 2.4). From 2005 to 2020, the higher-income region generally reduced agricultural land (i.e., cropland and pasture), while in other regions, agricultural land expanded at the expense of forest and other natural land. Between 2020 and 2050, agricultural land – mainly cropland – is projected to increase in the lower-income region (+124 Mha, representing +9% compared to 2020), replacing forests and, primarily, other natural land. Sub-Saharan Africa is projected to see the largest expansion, with an increase of 117 Mha, which is driven by an increase in demand for food and feed due to population growth. Yields are expected to increase as well, but not enough to prevent agricultural expansion into forests and other natural land. In the middle-income region, agricultural land is also projected to grow by 109 Mha (+5%), with pasture expansion playing a significant role, particularly in Central and South America, driven by a continued increase in international demand for both crops and animal products from this region. In the higher-income region, the availability of natural land suitable for agricultural expansion is limited, and growth in food production continues to rely on intensification.
Change in land cover in Mha
Source: Environmental Outlook modelling toolbox.
While agricultural land use is the dominant driver of land use change, other factors such as urbanisation also contribute to land use change. Urbanisation has a lower impact on land use in terms of area (less than 1% of global land use), but it could constitute an opportunity to reduce environmental impacts, provided it is sustainably managed, as discussed in Box 2.3.
The share of urban population has increased significantly over the last decades in all regions of the world, reaching around 80% in Northern and Latin America, between 75% and 70% in Europe and Oceania, 50% in Asia, and 43% in Africa (UN, 2022[30]).
Overall, urbanisation tends to decrease energy consumption and emissions in more mature and highly developed areas (IPCC, 2022[31]), however it has complex environmental impacts. By attracting people to cities, rural areas recover, and biodiversity is enhanced in previously pressured areas (IPBES, 2019[18]). Agglomeration economies also promote resource efficiency and energy savings (UNEP, 2019[32]). However, as cities expand, they may replace natural land with infrastructure, reducing carbon sinks, biodiversity and increasing impervious surfaces, which exacerbate heatwaves and flooding (Lemonsu et al., 2015[33]; Zhang et al., 2018[34]). As cropland is lost to urbanisation, productive land can be replaced with intensification and expansion on less productive land (Bren d’Amour et al., 2016[35]).
Urban expansion can lead to natural habitat conversion and degradation (van Vliet, 2019[36]; McDonald et al., 2020[37]) but also to natural habitat fragmentation (Haddad et al., 2015[38]; Liu, He and Wu, 2016[39]), which is a major driver of biodiversity loss. Urban expansion often occurs through urban sprawl and low-density development patterns that have a significant environmental footprint per capita (OECD, 2018[40]). Urban land grew at almost double the speed of the global population between 1975 and 2015 and is projected to triple in 2030 relative to the 2000 cover (UNEP, 2019[32]). This will result in loss of natural habitats in key biodiversity hotspots, although land cover of human settlements remains small compared to agricultural land (Seto, Güneralp and Hutyra, 2012[41]).
Urban expansion has large impacts on biodiversity loss at both global and local levels. Globally, it is estimated that urban expansion has caused about a 50% loss of local within-site species richness and a 38% loss of total abundance of species in intensively used urbanised areas (Newbold et al., 2015[42]). In local areas, urban expansion has resulted in more than 80% of natural habitat loss (Ke et al., 2018[43]).
In terms of climate change global urban expansion led to a net loss in carbon sinks estimated at 22.4 Mt CO2e per year between 2000 and 2010 (Liu et al., 2019[44]). This loss is equivalent to 0.7% of the 3.4 Gt CO2e absorbed by land ecosystems every year globally (IPCC, 2021[7]). Low density suburbs contribute more GHG emission per person than cities. Urban areas that are more compact, walkable, and co-located can reduce future urban energy use by 20–25% in 2050, while providing a corresponding mitigation potential of 23–26% (IPCC, 2022[31]).
As for the impacts of climate change, urbanisation processes generate vulnerability and exposure which combine with climate change hazards to drive urban risk and impacts (IPCC, 2022[45]). Urban flooding risks increase with urban expansion and land use changes, which enlarge impermeable surface areas, impacting floodwater drainage and causing sewer overflows. Urban development in coastal cities exposes them to high risk of sea level rise impacts (IPCC, 2022[45]).
In terms of air pollution, a 1% increase in cities’ density is associated with a 0.3% reduction in per capita emissions of CO2 and PM2.5 (Castells-Quintana, Dienesch and Krause, 2021[46]). Large metropolitan regions with high urban sprawl experience 60% more high ground-level ozone days than more spatially compact metropolitan regions (Stone, 2008[47]). Fragmentary urban patterns with many urban patches are positively linked with rising PM2.5, PM10, ground-level ozone and NO2 concentrations (McCarty and Kaza, 2015[48]; Cárdenas Rodríguez, Dupont-Courtade and Oueslati, 2015[49]).
Energy
Energy supply is projected to continue to be dependent on fossil fuels
Energy production and use contribute significantly to both GHG emissions and air pollutant emissions. Indeed, a large share of GHG emissions originate from energy use, which includes the combustion of fossil fuels by end-users and for electricity generation. Combustion of fossil fuels is also the source of most air pollutants, with the exception of ammonia (NH3) and non-methane volatile organic compounds (NMVOC). In 2020, combustion accounted for 8% of total NH3 emissions and 33% of NMVOC emissions, shares that are decreasing over time. For other air pollutants, the contribution of combustion ranges from 60% for carbon monoxide to 94% for sulphur dioxide (SO2) in 2020. In all cases, this share decreases over time, indicating that emissions from industrial processes are being abated less than those from combustion.
Current policies embedded in the baseline scenario result in a continuation of past trends. Overall, total primary energy supply (TPES) is projected to grow, driven largely by increased affluence and improved access to energy services, especially in the middle- and lower-income regions. While the deployment of renewable energy is expected to accelerate, it will not be sufficient to offset the continued reliance on fossil fuels. More specifically, TPES is projected to reach 755 exajoules (EJ) by 2050, a 36% increase compared to 2020 (Figure 2.5). By mid-century, slightly more than one-quarter of this supply will come from renewable sources, up from 14% in 2020 (and 12% in 2005), reflecting a 2.6-fold increase in renewable energy supply over 30 years. However, despite this substantial growth, fossil fuels are still projected to remain the dominant energy source under current policies, reaching 541 EJ in 2050, a 16% increase relative to 2020.
The aggregate evolution of TPES hides contrasting regional patterns. While TPES in the higher-income region is projected to decrease by 10% between 2020 and 2050, TPES is projected to more than double in the lower-income region and increase by 31% in the middle-income region in the same time period. The reduction in TPES in the higher-income region is driven by absolute reductions in fossil fuel supply with coal, oil and gas supply decreasing by 61%, 45% and 22% respectively. Much of the growth of TPES in the middle-income region is driven by a shift towards gas whose share increases from 25% to 31% and solar whose share rises from less than 1% to 9%. In parallel, the middle-income region is projected to decrease its reliance on coal, whose share decreases from 38% to 23%. In the lower-income region, TPES growth is driven by an absolute and relative increase in coal whose share in total TPES increases from 30% in 2020 to 42% in 2050 an absolute increase of 241%.
The reliance on fossil fuels is projected to decrease in the higher-income region but remain in the middle-income region and even grow in the lower-income region. The TPES evolutions leads to a projected decrease in the share of fossil fuels in TPES from 84% in 2020 to 56% in 2050 (corresponding to an absolute decrease from 162 EJ to 98 EJ) in the higher-income region, from 91% to 76% in the middle-income region (while supply still increases from 238 EJ to 262 EJ). In the lower-income region, the share of fossil fuels in TPES is projected to increase from 69% to 78%, corresponding to an increase in fossil fuel supply from 66 EJ to 180 EJ.
Energy supply in EJ/yr
Source: Environmental Outlook modelling toolbox.
The energy system continues to electrify, with a continued dependence on fossil fuels and the emergence of renewables
Global electricity production is projected to rise substantially by mid-century, faster than the growth of primary energy demand, indicating an increased electrification of energy services. While a growing share of this production will be met through wind and solar power, fossil fuels, particularly gas and coal, are expected to continue playing a significant role under current policies. Power generation is projected to more than double between 2020 and 2050, globally increasing from 96 EJ to 220 EJ in 2050. The majority of these 124 extra EJ will be generated in the middle-income region (59 EJ) and the lower-income region (53 EJ). Further details on regional electrification rates by sector are provided in Section 2.2.2. In 2020, solar and wind together made up less than 13% of the power mix in the higher-income region, and less than 8% in other regions. The contribution of solar and wind is expected to increase markedly, reaching 37% of global electricity generation in 2050 (60% in the higher-income region, 34% in the middle-income region and 24% in the lower-income region). Nevertheless, gas and coal remain key components of the power system under current policies: the use of gas and coal to produce power will continue to expand in absolute terms, albeit at a slower pace than lower-carbon sources, resulting in a reduced relative contribution to the overall power mix.
The increase in emissions of both air pollutants and GHGs from energy use is moderated by greater electrification and the expansion of solar and wind. At the same time, the rapid development of renewable energy technologies can affect resource use and potentially have impacts on biodiversity, as discussed in Chapter 6. Carefully designed policies will be crucial to minimise these negative trade-offs and ensure that the energy transition support all dimensions of the triple planetary crisis.
Other sectors, including plastics and other chemicals
Beyond agriculture and energy, other economic sectors also contribute significantly to environmental pressures. For example, the environmental impacts of mining are well documented (Macklin et al., 2023[50]), and certain industrial processes, such as cement production, directly emit GHGs. The rapid growth of digital services has driven a surge in the number of data centres, which causes rising energy use and adds pressure on land use. Additionally, electronic waste is becoming an increasingly urgent issue: in 2022 alone, 62 million tonnes of electrical and electronic waste were generated globally (Baldé et al., 2024[51]). The following paragraphs provide details on the future trends of selected key sectors that play a substantial role in fuelling the triple planetary crisis. These include the chemicals sector, including plastics, followed by a broader perspective on materials use, and concluding with key developments in waste generation and management.8
The chemicals sector encompasses a broad range of products, including plastics and pharmaceuticals, that are used in a wide variety of applications, both as inputs in production (e.g. fertilisers and pesticides in agriculture, plastics in car manufacturing) and as consumer products (e.g. cleaning products, paints). The sector represents a major area of economic activity. Some substances, often labelled as chemicals of concern, have significant adverse effects on human health, ecosystems and pollution levels (Landrigan et al., 2018[52]). The cumulative impact of continued exposure to multiple chemicals is increasing over time, raising growing concern about chemical pollution (OECD, 2023[53]). In recent years, widespread environmental contamination and human exposure to “forever chemicals”, i.e. PFAS (per- and polyfluoroalkyl substances), has drawn heightened attention (Goldenman et al., 2019[54]). Beyond PFAS, many hazardous chemicals and pharmaceutical pollutants, as well as artificial fertilisers and plastics, are released in large quantities and contribute significantly to environmental degradation (UNEP, 2019[55]). In this context, the consumption, trade, production and use of these substances can be considered as drivers of climate change, biodiversity loss and pollution.
Chemical production is a major industry. Including pharmaceuticals, the chemical industry was the second-largest manufacturing sector globally in 2017, with total sales reaching USD 5.7 trillion (tln). Asia is currently both the largest producer and consumer of chemicals (UNEP, 2019[55]). By 2030, global chemical sales are projected to double, with particularly rapid expansion expected in middle-income countries. The production of a wide range of chemicals, including fertilisers, pesticides, pharmaceuticals, perfluorinated chemicals, flame retardants and nanomaterials, is increasing across many regions, with production of petrochemicals (including plastics) growing even faster.
Industrial use of plastics is expected to continue increasing to 2050, leading to a consistent stream of plastic waste that countries will need to manage. The low cost and versatility of plastics are expected to drive demand up, with little change in plastic intensity. As plastic waste closely follows plastics use (albeit with a delay), this growth raises concerns about leakage to the environment, particularly where waste is not managed in a manner that is protective of human health and the environment (OECD, 2022[56]). With a significant share of global plastics use projected to take place in regions with underdeveloped waste management systems by 2050, plastic leakage can increase if plastic waste fails to be collected, sorted and treated in an environmentally sound manner, resulting in additional plastic pollution.
More broadly, resource extraction, including the mining of metals and mining and quarrying of non-metallic minerals, is highly energy-intensive and thus causes high GHG emissions as well as significant air, soil and water pollution (Macklin et al., 2023[50]; Coelho, Teixeira and Gonçalves, 2011[57]; Dietler et al., 2021[58]). Downstream material refining and processing further contributes to emissions. In total, more than half of all GHG emissions can be directly or indirectly attributed to “materials management”, although the share attributed to the extraction of non-fossil fuels remains a minority (OECD, 2019[15]). Recycling materials is generally less energy intensive than extraction (OECD, 2019[15]). For instance, producing the most commonly used metals by recycling metallic waste and scrap needs 60% to 97% less energy than producing them from mined material.
2.2.2. Composition of demand for goods and services and structural change of the economy
Agriculture and land use: the impact of the nutrition transition on agricultural demand and environmental sustainability
The nutrition transition refers to the link between rising income and changing diets. It typically involves an increase in overall calorie consumption, a shift away from starchy staples, and a significant rise in calories from animal products, fats, and sweeteners. This shift has a considerable impact on future agricultural demand.
The growing demand for animal-based calories, driven by income growth (particularly from low- and middle-income regions), plays a key role in increasing the environmental impact of the agricultural sector. Global meat consumption increased from 231 Mt in 2010 to 300 Mt in 2024 as demonstrated in Table 2.1 (OECD/FAO, 2025[59]). This represents an average annual growth rate of 2%. Between 2010 and 2024, average annual growth in meat consumption was particularly high in Asia and Africa which recorded increases of 2.7%. Gouel and Guimbard (2018[60]) estimate that, under current policies and historical trends, by 2050, demand for animal-based calories will double compared to 2010, while food demand for starchy staples is projected to rise by just 19%.
Quantities of meat in million tonnes (Mt)
Note: The regions in this table differ from the rest of the chapter as they are based on data availability.
Source: Authors’ own calculation based on data from OECD/FAO (2025[59]). Tonnes of carcass weight equivalent converted to edible weight equivalent using conversion factors 0.67 for beef and veal, 0.73 for pig meat and 0.66 for sheep meat.
This dietary shift towards animal-based calories drives the environmental impact of agriculture. Current production practices for animal products in many places can have detrimental effects on biodiversity, increase pollution and contribute negatively to climate change. While the environmental impacts of animal production practices can vary 50-fold depending on producers, even the impacts of the lowest-impact animal farming products often exceed those of vegetable substitutes (Poore and Nemecek, 2018[61]). Livestock activity can affect hydrological inflows (such as infiltration and soil moisture) and outflows (such as runoff, sediment) although the situation varies across many dimensions including climate, herbivore types and soil textures (Eldridge, Ding and Travers, 2022[62]). Additionally, meat production is a major source of GHG emissions, notably from ruminants due to enteric fermentation and manure. Livestock systems with cattle, buffaloes, sheep, goats, pigs and chickens collectively contributed to 6.2 Gt CO2e emissions in 2015 approximately 12% of all anthropogenic GHG emissions (FAO, 2023[63]) and half of all GHG emissions from AFOLU (IPCC, 2022[31]). Manure runoff also contributes to nutrient pollution, while enteric fermentation leads to ground-level ozone formation (Benton et al., 2021[64]). Furthermore, livestock production is closely linked to land-use changes and biodiversity loss, as increased demand for feed crops expands agricultural land, and newly deforested areas are in some regions primarily driven by cattle grazing needs (Campbell et al., 2017[65]; IPBES, 2019[18]). Moreover, the livestock sector increases the pressure on water scarcity by using an equivalent of 11 900 km³ of freshwater annually or approximately 10% of the estimated annual global water flows. The livestock sector impacts the whole water cycle as 2 290 km³ of green water and 370 km³ of blue water was attributed to feed production on cropland in 2010 (FAO, 2019[66]).
Energy: the role of electrification in facilitating the transition towards clean energy
Electricity contributes more to the increased final energy use than other energy sources, resulting in larger electrification in 2050 compared to 2020 in all key sectors. In commercial buildings, the share of electricity in total energy demand is already larger than 50% in 2020 in six of the nine regions and continues to increase over time (Figure 2.6). Sub-Saharan Africa, which has the lowest share in 2020 around 25%, shows a rapid catch-up over time. Residential energy follows a similar pattern, with increased electrification over time and a catch-up of Sub-Saharan Africa. Industry, although less electrified than commercial or residential buildings, continues to electrify its energy use over time at a steady pace. Through electrification, industry can decrease process emissions, such as with the replacement of blast furnace by electric arc furnace, which decrease the use of fossil energy (coal) as well as CO2 process emissions.
Electrification rate in key sectors: transport, buildings, and industry, % of final energy demand
Source: Environmental Outlook modelling toolbox.
In 2020, the share of electricity in transport final energy demand was less than 7% in all regions (Figure 2.6). This share is projected to increase over time, although at different speeds. In Europe, transport is projected to have the fastest electrification (39% of its energy demand electrified by 2050). East Asia, Eurasia, Central & South America and North America are projected to have slower uptakes than Europe by 2050, ranging between 21% for Eurasia to 27% for East Asia. Other regions are projected to have electrified less than 20% of total energy demand by 2050 with the Middle East & North Africa projected to have electrified less than 2% of their total energy demand.
Other sectors, including plastics and other chemicals: recycling can alleviate the transition to lower impacts of materials
Further to the GHG impacts of industrial sectors, some industrial activities lead to other environmental impacts. Many of these impacts can be linked to materials extraction and transformation (OECD, 2019[15]). Recycling is a solution to reduce the potential impacts of materials, and therefore of industrial sectors. Recycling reflects an increase of the efficiency of raw materials use, as the same materials are not discarded but can be used again (often after some form of re-processing), thereby increasing production volume per unit of materials input over time.
However, recycling is not equally possible for different materials. Secondary materials (the result of processing recyclable waste into raw materials that can be used again) currently make up a modest part of total materials use, while having lower environmental impacts across a range of LCA indicators. Many metals have substantial recycling rates, and scrap metals are used as secondary material. The share of secondary lead has surged in recent years to above 50%, while secondary steel has gradually declined to below 30%. Secondary shares for aluminium, zinc and copper are even lower. Recycling – and hence secondary materials use – is rare for non-metallic minerals; concrete for example is often used as low-value road filler. The use of both primary and secondary materials is projected to rise with current policies. Given the projected technology trends and unchanged policies, the supply of secondary materials is insufficient (or too expensive) to meet the demands of a growing economy.
Increased recycling can allow for a decrease in the (primary) resources used for production. For instance, OECD (2022[56]) projects that the growth of virgin plastics will be partly dampened even with current policies, as the share of recycled plastics doubles to 12%. This shift towards more circular economies can allow materials to be used more efficiently and decrease the exploitation of natural resources, for instance metals mining or fossil fuels inputs for plastics, although substitution away from specific materials towards other inputs always requires careful consideration of the net environmental impacts (as projected by LCA). Combined with policies to curb demand for materials, the resulting reduction in virgin materials can lessen the pressures on the triple planetary crisis.
2.2.3. Efficiency improvements in production of goods and services
The intensification of agricultural practices can reduce land use change
The decomposition of the drivers of changes in cropland illustrates the key role played by intensification and the need for sustainable intensification of agricultural production. Based on the methodology proposed in Huber at al. (2014[67]), Figure 2.7 decomposes the change in cropland expansion into different drivers. On the demand side, the increase in population, combined with the increase in per capita consumption drive a scale effect (see Section 2.2.1); in parallel, the change in diets modifies the types of agricultural goods that are consumed, driving a composition effect (see Section 2.2.2). On the production side, the increase in yields, i.e. a decrease in the area intensity of cropping, counteracts the impact of these two effects, limiting cropland expansion. This reflects the efficiency effect: more food can be produced on the same land. While scale effects accounted for 63% of cropland expansion between 2005 and 2020, diet intensity is projected to account for over 50% of global cropland expansion between 2020 and 2050 demonstrating the increasing importance of the composition of diets for cropland change. At the global level, while agricultural demand increases, yields also rise, but not sufficiently to fully fulfil the larger demand, resulting in cropland expansion. Nonetheless, the projected cropland expansion is only about a quarter of what it would be without the yield improvement, reflecting the importance of efficiency effects.
Global cropland expansion (in Mha) and contribution of identified factors of change.
Note: The variables are calculated as follows. Area intensity of crop production is the ratio of cropland area (in million ha) to crop production (in million tonnes dry matter), reflecting the inverse of crop yields and cropping intensity through fallow or multi-cropping (ha/ tonne dry matter). Diet intensity describes the intensity of the diet, namely the ratio of crop use for food and feed (in million tonnes dry matter) to consumption (in million kcal) and describes how much crops are used per unit of consumption (tonne dry matter / kcal). Per capita consumption is measured as the ratio between food demand over population per year (in kcal/capita). Population is reported in billion people.
Source: Environmental Outlook modelling toolbox.
Energy efficiency mitigates the growth of energy demand
The continuous increase in kilometers travelled by road, air and maritime transportation leads to a significant increase in energy demand. Efficiency improvements (detailed in Annex Figure 2.A.7) can only partially compensate this. All modes included, the increase in transport activity only results in a 23% global increase in energy demand for transport (compared to 60% increase in freight transport and 122% in passenger transport), showing that efficiency improvements partially decouple the increase in the economic service provided from the underlying polluting economic activity, thereby avoiding significant emissions of e.g. GHGs and air pollutants.
Similarly, residential buildings show significant efficiency improvements but no reduction in energy demand. The floor space for residential buildings increases significantly, though the associated energy demand is almost completely offset by increased efficiency in the energy use for heating (by switching to heat pumps for example). Floor space for residential buildings is projected to increase by 53% between 2020 and 2050, while energy demand for residential buildings only increases by 3% over the same period, showing large improvements in energy efficiency over time.
Industrial value added more than doubles between 2020 and 2050, increasing in all regions. Indeed, global industrial value added increases by 131% between 2020 and 2050. Over the same period, energy demand by industry only increases by 31%, again demonstrating increased efficiency in energy use.
Other sectors: efficiency can allow to reduce environmental impacts
Improved efficiency translates into a reduction in the use of inputs, leading to decreased environmental impacts (for a given production level). In addition to recycling (discussed in Section 2.2.2), improved technology can increase the efficiency of use of raw (e.g. metals and non-metallic minerals) and synthetic materials (e.g. plastics and chemicals) in their production processes.
This trend of efficiency improvements is visible for plastics: the global plastic intensity (the amount of plastics used that accompanies the generation of a dollar of GDP) is projected to fall by 12% by 2050. The reduction in plastic intensity allows to dampen the growth of plastics use and the implied environmental impacts, especially stemming from waste mismanagement. This decrease in plastics use intensity to the 2050 horizon is observed in most sectors and regions, highlighting the key role of improvements of technologies (OECD, 2022[12]). Exceptions include the food products and construction sectors in developed economies which increasingly use plastics, reflecting a shift towards commodities that use more plastics rather than a decline in efficiency, i.e. the shift in the composition of sectoral economic activity more than compensates for the decline in plastic demand in each individual sector. Economy-wide servitisation, i.e. increased use of services in manufacturing, drives a large part of plastics use increases, reflecting the central role of packaging plastics, which make up 40% of total plastics use.
In contrast, chemical intensity, that is the ratio of the value of chemicals used in the production to the gross output value, is projected to remain relatively stable between 2020 and 2050 as shown in Figure 2.8.9 In other words, current policies do not lead to a significant improvement in the efficiency in which chemicals are used in production, and thus the associated environmental impacts remain unabated. The projected changes in sectoral chemicals intensity are negligible compared to the variation in chemical intensity for the same sector across regions. This points to the potential need for further policy action to reduce chemical pollution.
Chemical intensity expressed in %
Note: Chemical intensity is the ratio of the value of chemicals used during the production process to the gross output value, computed for each sector in each region. These values are measured in 2020 in the blue boxes and 2050 in the yellow boxes, and expressed in constant 2017 USD bln. The boxplots illustrate the distribution of chemical intensities across regions, for each sector. The top and bottom values of the boxes correspond to the first and third quartile of the distribution. The upper and lower whiskers extend from the hinge to the largest (resp. lowest) value no further than 1.5 times the inter-quartile range. The bar within the plot corresponds to the median value. Sectors with a median intensity lower than 1% in 2020 or 2050 are not represented in the figure. For instance, the chemicals sector displays a median chemical intensity of 11.8%, with a distribution ranging from 5.5% to 18.5% in 2020.
Source: Environmental Outlook modelling toolbox.
Material and water resources represent the physical foundations of human life and thus are at the core of the triple planetary crisis. The way material and water resources are extracted, used and disposed of throughout the economy and human life impact the evolution of the pressures on and state of the environment. Together with the demand for the various services provided by the environment, including food, drinking water, energy and goods, the evolution of the way these services are provided shape anthropogenic pressures to the environment. They thus combine with the drivers outlined above as crosscutting aspects.
2.3.1. Use of (raw) material resources
The extraction and processing of (raw) material resources10 – biomass, fossil fuels, non-metallic minerals and metals – is a key common driver of climate change, biodiversity loss and pollution. UNEP (2024[68]) find that the extraction of material resources is responsible for over 60% of climate impacts (including land use change) alone whilst the combination of extraction and processing accounts for 40% of health impacts due to emissions of particulate matter (half of which is due to non-metallic minerals). Biomass (agricultural crops and forestry) also accounts for approximately 60% of the total land use related biodiversity loss and water stress although sustainable land and forest management can reduce some of these impacts (FAO, 2024[29]).
All mining on land has increased dramatically and has significant negative impacts on biodiversity, emissions of highly toxic pollutants, water quality and water distribution, and human health (IPBES, 2019[18]). Surface mining leads not only to deforestation (Giljum et al., 2022[69]) but also to other landscape transformations including opening of pits, vast amounts of toxic waste and changes in freshwater streams, all detrimental to biodiversity (Macklin et al., 2023[50]). Subsequent processing of fossil fuels, ores and minerals also releases CO2, SO2, CH4, particulate matter, mercury and other heavy metals, generating acid rain and raising the bioavailability of mercury and other heavy metals (IPBES, 2019[18]). While material extraction of fossil fuels, metal ores and non-metallic minerals is not a major driver of water stress at the global level, it can consume more freshwater than what is naturally available at the local level (Meißner, 2021[70]).
In the baseline scenario, the use of primary materials is projected to increase roughly by half from 96 Gt in 2020 to 145 Gt in 2050. The increase in materials use applies to all material groups considered in the analysis (biomass, fossil fuels, metals and non-metallic minerals) and all regions in the world. Non-metallic minerals remain the largest material group in the three regions. The sustained use in the middle-income region supports strong infrastructure growth. A more gradual ramp-up of infrastructure and minerals use similarly occurs in the lower-income region.
Materials use in Gt and GDP in tln USD
Source: Environmental Outlook modelling toolbox.
As the economies of fast-growing countries mature and build up infrastructure, factories and housing, their use of materials (mostly non-metallic minerals and metals) increases strongly. After the investment boom, materials use tends to stabilise and is directed mostly at investment that replaces existing infrastructure, which tends to involve less intense use of materials.
Despite a projected relative decoupling between economic activity and materials use between 2020 and 2050, materials use is projected to continue rising. The increasing share of services in manufacturing, household and government demand, alongside trends such as digitalisation and an increase in R&D, increases the importance of service sectors in the economy. Since services are less material-intensive than agriculture and industry, this structural shift leads to a decline in global material intensity over the projected period. This “relative decoupling” does not imply a reduction in total materials use. On the contrary, material-intensive sectors continue to expand in absolute terms through 2050, resulting in a substantial overall increase in global materials use. In other words, while each unit of economic output uses fewer materials (including energy), the total volume of materials used still rises due to continued economic and population growth. For instance, under current policies (for example on food loss and waste), global demand for food and agricultural goods is projected to increase by about 38% by 2050 over 2020 levels.
2.3.2. Water use
Water withdrawal is projected to increase by 17%, in line with increasing demand trends for other resources. Water withdrawal constitutes the freshwater taken from ground or surface water sources, either permanently or temporarily.11 Thus water withdrawal differs from the related concept of water consumption (water use). Water withdrawal includes water uptake by economic activities that is later returned to the environment.12 Figure 2.10 details the main applications for water. The largest user of water continues to be the agricultural sector through irrigation: despite some increases in efficiency, total water use for irrigation is projected to increase between 2020 and 2050 as total agricultural land area equipped for irrigation grows. The largest increases in water withdrawal are projected in demand from households and in the industry sector. This is in line with the growing population, increasing affluence and higher consumption that cannot be compensated by increases in water use efficiency (the ratio between water withdrawal and consumption). The demand for water for energy generation is projected to decrease, as renewable electricity sources increase and traditional fossil-fired and nuclear power plants decrease. The latter technology requires large amounts of cooling water while wind and solar electricity do not need this.
Changes in water withdrawal from 2020 to 2050 differ substantially between, but also within, world regions (Figure 2.11). Strong increases are projected in specific parts of South Asia, Middle East, Sub-Saharan Africa, Latin America and the Mediterranean. This is partly due to expansion of irrigated agricultural lands but also to growing populations and increased economic activity, causing higher water demand for households, industry and energy purposes. Substantial decreases in demand are observed in the urban areas of Northern America, Europe and Eurasia, which is related to increased water use efficiency in water demand for industry and households and decreasing demand for energy generation due to shifts towards more renewable energy sources.
Water withdrawal in km3/yr
Source: Environmental Outlook modelling toolbox.
Water intensity, i.e. water withdrawals as ratio to GDP, are projected to decrease across all regions and converge towards levels similar to current levels for North America, Europe and Japan, Korea & Oceania; from a global average of 47 m3/kUSD in 2020 to 24 m3/kUSD in 2050. For comparison, water consumption per capita converges from a global average of 696 m3/cap in 2020 to 659 m3/cap in 2050.
Change in water withdrawal in cubic hectometres per year (hm3/year) between 2020 and 2050
Source: Environmental Outlook modelling toolbox.
To understand and address the future challenges of the triple planetary crisis, the key trends outlined above can be brought together. The current section presents a coherent dashboard of the main trends and drivers in an easily accessible manner. This can support a holistic understanding of the dynamics fuelling environmental degradation and help guide more effective and integrated policy responses. Annex 2.B further decomposes trends in environmental pressures into the contributions of different socioeconomic trends and specific drivers.
The proposed dashboard shows key socio-economic trends and the evolution of a selected set of drivers behind the environmental pressures contributing to the triple planetary crisis (cf. Figure 2.12). A central message of this chapter is the importance of examining cross-cutting issues that drive the crisis, going beyond economic growth and demographics. These transversal drivers relate to agriculture and land use, production and use of energy, and the use of resources, such as materials and water. Accordingly, the dashboard is structured around four categories of indicators. The first category includes GDP and total population. The second category focuses on total final energy use, primary energy use from fossil fuels, and coal use for electricity generation. The third category relates to agriculture, including total food demand, meat and dairy products demand, and fertiliser use. The fourth category captures resource use through water consumption, materials and plastics uses.
Interlinkages are key. First, each single driver contributes to multiple environmental pressures. For example, drivers related to energy production and use are primarily associated with emissions of GHGs and air pollutants, which are environmental pressures that contribute simultaneously to climate change, biodiversity loss and pollution. Similarly, agricultural drivers influence land use, emissions of CH4, N2O, and NH3, as well as nutrient pollution and pesticides use (not shown in the dashboard). Second, the indicator groups are not independent of each other. For instance, food production requires significant water inputs; fossil fuels and food are counted within materials use (alongside with metals and non-metallic minerals); and both the production and transportation of food require energy. These examples illustrate the complex interdependencies that exist even at the level of the drivers, reinforcing the need for an integrated understanding of the forces behind climate change, biodiversity loss and pollution.
Overall, the analysis shows a degree of relative decoupling between the drivers of environmental degradation and economic activity, but pressures to the environment are projected to continue to grow. Figure 2.12 presents the projected growth of the selected indicators from 2020 to 2050 (and the evolution in 15-year increments between 2005 and 2050). At the global level, and under current policies, both economic activity and population are expected to grow significantly between 2020 and 2050 as shown in Figure 2.12, Panel A. Real GDP at purchasing power parity (PPP) and Population bars indeed extend beyond the circle representing an index 100% for the level in 2020 (see also Section 2.2 and Annex 2.A). GDP is projected to more than double, while the global population is expected to increase by 24%. These trends reflect increasing incomes and living standards, accompanied by greater access to energy and food. However, this trajectory may raise environmental concerns. To limit negative impacts, economic growth must become more sustainable, curbing overconsumption and increasingly relying on cleaner energy sources and cleaner sectors, which will require significant technological progress. In other words, to prevent further environmental degradation, or, ideally, to improve the state of the crisis, composition and technique effects must outweigh scale effects. Overall, global projections indicate that, under current policies, this is unlikely to occur. All the selected drivers of climate change, biodiversity loss and pollution are expected to increase, albeit at a slower pace than GDP. This implies a degree of relative decoupling from economic activity, but not absolute decoupling, meaning environmental pressures will continue to rise, but not as rapidly as the economy.
A key takeaway of this chapter is that improved efficiency plays a crucial role in moderating the rising pressures driven by multiple socio-economic and sectoral trends. As the bar representing the evolution of GDP is longer than the ones representing materials use and water use in Figure 2.12, Panel A, materials use and water use efficiency are projected to improve. This means that less water and materials will be required to produce each dollar of GDP in 2050 than in 2020. However, under current policies, these efficiency gains are not sufficient to offset the overall scale of economic and population growth. Efficiency improvements enable a relative decoupling, which helps slow the increase in energy, food, materials, and water demand, but are not sufficient to halt their continued rise at the global level in the coming decades. Thus, efficiency improvements need to be combined with upstream policies to curb demand and downstream policies to deal with environmental burdens.
Most drivers shown in the dashboard exhibit an upward trend in per capita terms. Notable exceptions include water consumption and use of fossil fuels to produce primary energy. Although global water consumption (not to be confused with water withdrawal, which includes water temporarily withdrawn from the environment but returned later) is projected to rise by 22%, per capita water consumption is expected to decline slightly by 1%. Similarly, while the use of fossil fuels for primary energy increases by 16%, per capita use declines by 6%, driven by greater energy efficiency, increased electrification of transport, and the growing share of renewables. However, coal use for electricity generation stands out: it is projected to increase substantially, by 86% in absolute terms and 50% per capita, despite a 17% decline in its intensity relative to GDP.
While still showing growth, comparing the bar lengths in Figure 2.12, Panel A with Panel B illustrates a slight slowdown for the projected 2020-2050 evolution of climate change, biodiversity loss and pollution compared with 2005-2020. Some drivers (e.g. plastics use) show limited signals of slower growth in the future. Coal-fired electricity generation, on the other hand, is set to grow faster in the 2020-2050 period.
Global aggregates mask significant regional heterogeneity. To provide a more detailed perspective, Figure 2.13 shows the same set of indicators as in Figure 2.12 but disaggregated across nine regions, covering the world. Two additional elements enrich the regional plots: a black dot, which shows the evolution of the global average for each indicator (as in Figure 2.12), and bar colours, which reflect the relative intensity of each indicator in 2050 compared to the region with the highest value. For example, GDP per capita in 2050 is highest in North America (represented in black) and lowest in Sub-Saharan Africa (indicated in light blue), where it amounts to about 8% of North America’s level.
The evolution of key socio-economic indicators varies significantly across regions. GDP is projected to grow between 2020 and 2050 in all regions, with the fastest growth occurring in lower-income regions (see the length of the bar beyond the circle in the South Asia and Sub-Saharan Africa panels of Figure 2.13).13 Population growth is more uneven: it is expected to be more rapid in South Asia, Sub-Saharan Africa and in the Middle East and North Africa, while remaining stable or declining in Europe, East Asia, Eurasia and in Japan, Korea, and Oceania. While the global analysis does not reveal any cases of absolute decoupling in the coming decades, regional-level data show a few such instances. In such instances, the bars in Figure 2.13 are lower than the outer grey line meaning that the level of the indicator is below its value in 2020. Absolute decoupling cases are mostly related to the energy transition, with total energy and fossil energy use, as well as coal-based electricity supply all projected to decline in Europe (by -10%, -50% and -45% respectively), North America (-14%, -28% and -100%), and Japan, Korea, and Oceania (-9%, -50% and -63%). Another example is the projected 11% decrease in nitrogen fertiliser use in Europe. Nonetheless, such instances of absolute decoupling remain exceptions, with most regions continuing to experience rising trends across the drivers of climate change, biodiversity loss and pollution.
The majority of indicators show increasing trends across all regions. As incomes and access to energy improve, energy use more than doubles in South Asia and the Middle East and North Africa, and rises by more than 70% in Sub-Saharan Africa. These increases are largely met by expanded use of fossil fuels and coal-fired electricity. South Asia, in particular, stands out for showing sharp growth across all energy-related indicators. Sub-Saharan Africa also displays high projected growth in total energy demand, fossil fuel consumption, coal use in electricity generation, plastics, materials, and food demand. However, as indicated by the colour scale in Figure 2.13Error! Reference source not found., the intensity of these indicators in both South Asia and Sub-Saharan Africa remains among the lowest globally, even by 2050. Conversely, several region-indicator pairs combine both high growth rates and high per capita levels in 2050, posing particular concerns. These include plastics use in North America and Japan, Korea and Oceania, and coal-fired electricity generation in East Asia. Such combinations of high intensity and strong growth suggest heightened environmental pressures, reinforcing the triple planetary crisis.
In addition to the major trends already discussed, such as the absolute increase in key drivers of the triple planetary crisis, relative decoupling from GDP and gains in efficiency, the proposed dashboard also highlights notable sectoral developments projected under current policies. In the agricultural sector, current production and consumption patterns and their projected trends suggest the sector will not reduce its adverse impacts on the environment. As a consequence, the related emissions of GHGs and pollutants, as well as biodiversity loss, are expected to increase. Without shifts in production technologies, farm practices and food consumer behaviours, the climate mitigation potential of the agricultural sector will remain untapped. The projected continued reliance on fossil fuels is incompatible with efforts to address the triple planetary crisis. The transition to electrification powered by renewables, implemented in ways to minimise potential risks such as increased pollution and biodiversity loss (see Chapter 6), is a key component to mitigate the challenge. In parallel, total materials use and water consumption are also projected to grow, continuing to follow an unsustainable trend and adding further strain on ecosystems and natural resources. However, under current policies, the pace and scale of this transition are insufficient.
Global evolution for selected drivers
Note: The length of the bars represents evolution over time (between the two dates in each panel). For instance, for Panel A, the circle indicates the 2020 level indexed to 100%, against which to compare the evolution in the bar. For example, plastics use is projected to increase by 106% over the 2020-2050 period at the global level. Annex 2.C provides details on the values and the units of reported indicators.
Source: Environmental Outlook modelling toolbox.
Regional evolution for selected drivers, 2020-2050
Note: The length of the bars represents the evolution from 2020 to 2050. The circle indicates the 2020 level. The colour indicates the relative intensity of a given indicator in each region (e.g. per capita GDP in 2050) compared to the region with the highest intensity in 2050. For example, the light blue bar for Real GDP (PPP) in Sub-Saharan Africa shows that GDP per capita in this region will be the lowest in 2050, relative to other regions. In contrast, the dark bar for North America indicates that GDP per capita will be the highest in this region. Black dots represent evolution over time at the global level. Values are capped at +140% above 2020 levels for visual appearance (a triangle effect is present when values exceed this threshold). Annex 2.C provides details on the values and the units of reported indicators.
Source: Environmental Outlook modelling toolbox.
The world population is expected to continue to grow and reach 8.5 billion by 2030 and 9.6 billion by 2050 (Annex Figure 2.A.1). The global population growth rate is projected to decline gradually, implying a peak population level somewhere in the third quarter of the century. The decline in population growth applies to all countries. Yet the total population in lower-income regions is projected to continue to rise significantly in the coming decades, reaching more than 5 billion by mid-century. Average population growth in the coming decades is projected to be highly heterogenous across countries, ranging from population declines in East Asia to sustained growth above 2% in many less-developed Sub-Saharan African countries.
Population in billion people in 2020 and 2050, and total growth between 2020 and 2050
Source: Central projection from UN (2022[30]).
Furthermore, the composition of populations changes over time, with significant aging in many countries in the higher- and middle-income regions, leading to an increased share of elderly people (Annex Figure 2.A.2). This has a wide range of consequences, including different consumption patterns, among which an increased demand for health care services (de la Maisonneuve and Oliveira Martins, 2014[71]) and a reduction in the ratio of working age population to total population (and thus labour supply).14 In contrast, in many less-developed countries populations are significantly younger. Education levels are projected to continue to converge (see e.g. (KC and Lutz, 2017[72])). Education – and especially female education – in turn affects birth rates and thus future population size (Adhikari, Lutz and Kebede, 2024[73]).
Global population in million people by sex and age cohort
Note: M stands for male population, F for female population.
Source: Central projection from UN (2022[30]).
Global economic activity, measured through GDP, is projected to grow substantially faster than population: on average by 2.7% per year, versus 0.7% for population (Annex Figure 2.A.3). Thus, global GDP, at constant prices, is projected to grow from USD 126 trillion in 2020 to USD 283 trillion in 2050. There is significant regional variation, with steady but rather low growth in higher-income region, a declining growth rate in the middle-income region (not least related to a fast decline in annual growth in the People’s Republic of China), and faster growth in lower-income regions. The latter is a combination of fast growth in a number of emerging economies in the short run, not least India, plus increasing growth rates in many developing countries, including in Sub-Saharan Africa.
The average (per capita) income of the lower-income region in 2020 equalled USD 6 700; by 2050, this is projected to rise to USD 15 900; a 2.4-fold increase (Annex Figure 2.A.3). For the middle-income region, income is projected to increase from USD 16 000 on average in 2020 to USD 34 100 on average by 2050 (a bit more than doubling). Finally, for the higher-income region, growth is slower but starts from a higher level: from USD 45 300 in 2020 to USD 75 000 in 2050 (a growth of 65%). As with population, the income growth projections are fairly similar to those in the literature, and especially SSP2.
Per capita gross domestic product by region, levels in 2020 and 2050, and total growth between 2020 and 2050
Source: Environmental Outlook modelling toolbox.
The faster income growth in emerging and developing economies (Annex Figure 2.A.4) is in line with the concept of conditional convergence: countries that are further from their technology frontier will grow faster, as they catch up in technology to more developed economies (Johansson et al., 2013[74]; Dellink et al., 2017[75]). Population growth influences income in two distinct ways: per capita income lowers with higher population (resources are shared across more people), but more people in the working age population leads to higher labour supply, which is a component of GDP. This is most clearly seen in Sub-Saharan Africa: the currently young population implies a rapid increase in the working age population in the coming decades, and thus fast growth of labour supply. But total population grows almost as fast and dampens growth of income (GDP per capita). Investments and capital growth and technological developments have a positive effect on income growth. Emerging economies, such as the South Asia region, can sustain fast growth with a significant contribution from technological progress as they catch up. In the higher-income region, the contribution of the demographic factors is rather small, and both capital and technology drivers dominate overall income growth.15
Average annual income growth between 2020 and 2050 and contribution of the various drivers
Source: Environmental Outlook modelling toolbox.
As explained in the previous section, economic growth is driven by a combination of using more inputs (more labour, more capital) and increasing the production outputs that can be made with the same amount of inputs.16 The latter aspect, technological improvements, includes efficiency improvements in the use of specific production inputs, e.g. energy efficiency, as well as improvements in output per unit of primary inputs, especially labour, i.e. labour productivity growth (e.g. from learning-by-doing). These technology effects can differ substantially between sectors, and between regions. Generally, productivity growth is relatively fast in industrial sectors compared to agriculture and services, and faster in fast-growing emerging economies as their technology levels catch up to the most developed countries.
These productivity increases give rise to a change in the structure of GDP, as measured by the contribution of value added generated in the different sectors to economy-wide value added generated, i.e. GDP. For least developed countries, agriculture represents a large share, while production in emerging economies is characterised by industrialisation, including a significant growth in construction to support infrastructure development. Sectoral shares are more stable in developed economies, with high levels of services and relatively low levels of agricultural production. Consumption patterns do not have to follow these same trends, and the transformation of consumption patterns often lags behind transformation of production, as countries engage in international trade (see below) to specialise production where they have a comparative advantage, while consumer preferences gradually evolve as incomes rise.
In all three regions services make up the majority of value added generated, even more so in 2050 than in 2020 (Annex Figure 2.A.5). As expected, the share is lowest in the lower-income group, but even that group contains several countries that have fairly diversified economies. This domination of the services sector is linked to a range of socioeconomic trends: (i) the income elasticity of services tends to be higher than for most other commodities, and thus rising income leads to a rising share of household consumption spent on services; (ii) aging populations lead to increased demand for healthcare services; (iii) productivity growth is on average lower than for industry, and thus in relative terms expenditures on services need to rise compared to expenditures on goods to keep a balanced basket of consumption; (iv) manufacturing tends to be increasingly “servitised” (Pilat and Nolan, 2016[76]), and finally (v) digitalisation of the economy, including a swich to e-commerce, imply less need for physical commodities and more for services. Digitalisation contributes significantly to the servitisation of both industrial production and consumer demand, induces a shift away from equipment towards electronics, reduces international transport while simultaneously lowering the cost of international trade, and e.g. increases energy use for data centres (OECD, 2019[77]; OECD, 2024[78]).
In contrast, the share of agriculture in total production declines over time – food is an essential good and food demand rises only modestly as incomes increase. The share of value added generated in energy declines as well, reflecting continued energy efficiency improvements. The decline in energy value added is especially pronounced in the middle-income region, which contains many of the fossil fuel exporting countries. These shifts in the structure of the economy are much less pronounced in the higher-income region, whose countries tend to have more diversified economies and more stable demand structures.
Sectoral composition of value added (in constant 2017 USD), in shares of total value added
Source: Environmental Outlook modelling toolbox.
The faster population and economic growth in the middle- and lower-income regions also implies a shift in consumption towards those regions. Thus, the share of the higher-income region in global production and consumption is projected to decline over time. This also affects the balance of trade: in 2020, the largest trade flow occurs within the countries in the higher-income region, representing more than one-third of total trade. But this share declines over time, with exports by especially the lower-income region expanding (from 16% in 2020 to 20% in 2050); most of that expansion involves the rapidly increasing trade between countries in the lower-income region (from less than 4% in 2020 to 6% in 2050).
As the structure of the economy differs across regions, different countries have different trade specialisations. Thus, the increase in exports by the lower-income region does not extend proportionately to all commodities. Rather, the biggest increases in the share of the global market are projected for energy – relatively easy to achieve as export prospects for the fossil fuels are rather weak and thus there is very little growth of the traditional fuel trade flows – as well as industry, matching the theory on economic development. Specifically, industrial production in the lower-income region increases more rapidly than consumption, and development of export opportunities is a key part of the development strategy of many countries.
Share of regions in global exports by commodity group
Source: Environmental Outlook modelling toolbox.
International trade thus plays a key role in reconciling trends on the production side and trends on the consumption side. However, long-term imbalances on the trade balance, with a significant surplus on the current account, put pressure on exchange rates and in most cases such surpluses are projected to fade – though not necessarily fully disappear.
Technological and structural change also impact energy efficiency. Energy efficiency for freight and passenger transport is projected to improve as shown in Annex Figure 2.A.7. For instance, between 2020 and 2050, the energy intensity of aviation for freight transport (the amount of kilojoules (kJ) required to transport a tonne of freight by one kilometre (tkm)) is expected to fall by over 30% in the higher- and middle-income regions and 12% in the lower-income region. Similar improvements are also observed for the energy intensity of aviation for passenger transport (the amount of kJ required to transport a passenger by one kilometre (pkm)) with all regions projected to undergo reductions above 35%. The energy intensity of light-duty vehicles also falls by 36% (in the middle-income region) to 55% in the higher-income region. Most regions are projected to also have energy intensity reductions for buses, navigation and railways, although these are much smaller.
Residential, industry and commercial energy intensity is projected to decrease in all regions between 2020 and 2050. This is most pronounced for residential energy intensity (measured in kJ by m2) which decreases by approximately 50% in the higher-income region. This could be explained by the developments in building standards made in higher-income regions such as the standardisation of modern insulation methods. Industry and commercial energy intensity is projected to drop in all regions with the exception of commercial energy intensity in the lower-income region, in link with the development of services in this region.
Changes in energy intensity of key sectors: transport, buildings, and industry
Note: Freight transport energy intensity is the ratio of energy use in kilojoules (kJ) to tonne-kilometres travelled (tkm). Passenger transport energy intensity is the ratio of energy use in kJ to passenger-kilometres travelled (pkm). Residential is the ratio of energy use in kJ to m2, Industry and Commercial are measured as the ratio of energy use in kJ to value added in 2010 USD.
Source: Environmental Outlook modelling toolbox.
Other sectors, including plastics and other chemicals: regional and sectoral shifts in economic growth drive demand
The structure of the economy is changing over time. Such changes happen more rapidly for fast-developing countries that industrialise and diversify their economies than for more developed countries that already have relatively mature, diversified economies. Nonetheless, there is a trend towards servitisation (the increased reliance on services in manufacturing as an input or as bundled output) in all countries, with services making up a larger share of final consumption, e.g. health expenditures for health care in an aging population, but also increased use of services in industry, either by bundling outputs (a service contract combined with the purchase of a good) or through leasing commodities. These economic dynamics affect environmental pressures and are captured in the analytical framework through their effect on structural change and thus on the composition of total demand.
As the global economy grows between 2020 and 2050, the sectoral composition of chemical demand is projected to evolve. Around 60% of chemical demand is projected to continue to be driven by chemicals, other manufacturing and rubber and plastics by 2050, as illustrated in Annex Figure 2.A.8. The chemicals sector itself is projected to continue to be the leading consumer of chemical products (here excluding pharmaceuticals, rubber and plastics), accounting for over USD 1 trillion in 2020 and USD 1.7 trillion in 2050. While in 2020, over half of chemical demand by the chemicals sector is concentrated in North America, Europe, and Japan, Korea and Oceania, most of the projected growth in chemical demand to 2050 is driven by other regions. South Asia (+149%), Eurasia (+253%) and the Middle East and North Africa (+381%) see the highest growth rates driving growth in chemical demand by the chemicals sector. By 2050, an increasing proportion of chemical demand will be accounted for by other manufacturing increasing from 16% in 2020 to 18% in 2050 in contrast to the chemicals sector which will account for 27% in 2050 relative to 31% in 2020.
Chemical demand in USD bln
Note: Chemical demand represents the intermediate demand for chemicals (excluding rubber, plastics and pharmaceuticals) by sector (excluding services) in each region. Chemical demand is expressed in constant 2017 USD bln for both 2020 and 2050. Only sectors with a share higher than 2% in 2020 are shown.
Source: Environmental Outlook modelling toolbox.
Another example of the structural shifts occurring in the baseline scenario can be drawn with plastics. Three main applications account for approximately 60% of plastics use: packaging, construction and vehicles, which includes transport equipment (Annex Figure 2.A.9, Panel A). With limited substitution away from plastics foreseen under current policies, plastics use follows increases in economic activity across the economy and is thus projected to increase for all sectors (Annex Figure 2.A.9, Panel B). The largest increases in plastics use are for packaging, construction, transportation and consumer products, linked to soaring demand for consumer goods and transportation as economies develop. With the digitalisation and electrification trends, plastics use dedicated to electrical and electronic products increase strongly as well.
Despite the services sector having a relatively low plastic intensity (the amount of plastic per unit of output), the shift towards a service-based economy will result in the services sector using the most plastics. This trend manifests with the growing use of plastic products in service industries, such as packaging and consumer items (e.g., takeaway food containers, healthcare and medical products, art supplies, credit cards, and luggage). The rise in plastics used for packaging indicates that current policies are inadequate to counteract the increased plastics use by key sectors dependent on packaging, including business services, food products, and trade.
The increase of plastics use for other applications is slower. The growth of the textile sector in middle- and lower-income economies drive plastics use for clothes. In those countries, investment in infrastructure lead construction activities to use plastics, as necessary to economic development (OECD, 2019[15]). Plastics use for industrial sectors and machinery undergoes a slower growth, following the structural shifts away from industry towards services.
Source: Environmental Outlook modelling toolbox.
The projections include (mostly implicit) policies included in the macroeconomic projections and (mostly explicit) environmental policies in each of the domains of the triple planetary crisis and beyond. Annex Table 2.A.1 summarises the policies included in these two domains by wide issues.
Source: Authors’ own elaboration.
Environmental policies
Policies related to mitigating climate change are based on implementation of specific sectoral targets directly on appropriate model variables or appropriate proxies (renewable energy targets, efficiency improvements, etc.). Targets on decarbonisation rates are emulated through sectoral carbon prices, which result in a target emission level over a given time horizon. The carbon price, imposed on all GHG emissions, induces a response of the energy system where investments in energy efficiency, fossil fuel substitution and additional investments in non-fossil options increase (van Vuuren et al., 2017[86]). These carbon prices are differentiated at regional and sectoral levels. Land based climate change mitigation efforts are represented in the model by afforestation and reduced deforestation and forest degradation policies. Only policies that are currently embedded in national laws were taken into account and only countries with considerable AFOLU emission reductions were considered, including Brazil, Canada, the People’s Republic of China, European Union, India and Indonesia (Nascimento et al., 2023[81]).
The energy policies are specified by sector. In the electricity sector, policies implemented explicitly include capacity targets for power technologies, renewable energy supply targets as a percentage of the total electricity production, electricity capacity caps, coal phase out policies, and efficiency goals on existing and new power plants. Policies in the transport sector include fuel efficiency car standards, targets for the share of biofuels and electric vehicles, fuel taxes and subsidies. Regulations on CH4 fugitive emissions and F-gases are applied. In the building sector, energy efficiency is improved via imposing a premium factor on building insulation levels, providing subsidies for insulation of buildings, imposing appliance standards, and regulations for residential PVs and/or heat pumps uptake.
Current policies are implemented until 2030, after which they are continued at that level. The year 2030 was chosen since that is the target year for almost the whole policy ensemble implemented in the IMAGE model. Post 2030 it is assumed that policies remain constant, with sectoral policy targets maintained at the current policy level where applicable, using the following methods: All (sectoral) carbon prices, fuel prices, subsidies, premiums on power production etc. are kept at the same level until the end of the century. Policies that have a certain target level for a certain year, i.e. renewable capacity targets for power supply technologies, are treated as the minimum threshold in the model. Thus, if in the post-2030 period the endogenous model solution leads to an outcome that is more ambitious than the current policies, the model adopts the improved outcome.
Policies related to combatting pollution link to three elements in this report: air pollutants, plastic leakage, and nutrient pollution. First, policies related to reducing outdoor air pollution are implicitly represented through improvements in emission factors. The underlying assumption is improvements in pollutant end-of-pipe control, growing more ambitious up to 2100: emission factors assume efficient implementation of current environmental legislation (CLE) and thus describe a scenario of pollution control where countries implement all planned legislation until 2030 with adequate institutional support. The CLE emission factors are “fleet average” values that are the aggregate emission factor of all ages of equipment operating in the given year (Rao et al., 2017[87]). The higher- and middle-income regions implement CLE in 2030, reaching 2030 West Europe emission factors by 2100; the lower-income region implements Western Europe factors increased by 10% in 2030, reaching 2030 West Europe emission factors by 2100). Second, policies related to plastic leakage to the environment are represented through improvements in waste management infrastructure, leading to less waste, as well as explicit bans on some microplastics use. Third, policies related to fertiliser use are implicitly represented through improvements in nitrogen and phosphorus use efficiencies.
Policies to protect biodiversity in the baseline take the form of protected areas. The baseline projections include the assumption that current protected areas are stable and effectively protected until 2050, represented on a map in Annex Figure 2.A.10. Terrestrial protected areas in IMAGE and GLOBIO are based on the World Database on Protected Areas (UNEP-WCMC, 2020[80]). Agricultural expansion cannot occur in any type of protected area (Van der Esch et al., 2022[88]), but already existing agricultural land inside protected areas can remain. These protected areas represent 17% of the global land area (Kok et al., 2023[89]), while pressure on land increases at the expense of (unprotected) natural areas in the baseline scenario.
Source: Environmental Outlook modelling toolbox.
Other policies
The macro-economic projections include a number of policies other than environmental policies, directed towards several domains including economic, labour, education, monetary, trade. These trends, which policies underlie, are used to project the drivers of growth to produce long run macro-economic projections.
The demographic projections include the policies geared towards population and education from the UN database (UN, 2022[30]). The labour policies are implicitly taken onboard from (ILO, 2022[82])’s active population prospects and OECD Labour Force Statistics and Projections (2022).
The economic projections include all the policies described in the short- and medium-term forecasts (from OECD Economic Outlook (OECD, 2023[83]), the World Economic Outlook database (IMF, 2023[84]), and World Bank (2023[85])). These sources reflect the forecast of economic growth to 2028 as of late 2023. The projections further build on the long-term macroeconomic projections provided in (OECD, 2023[83]).
Decomposing the key factors of environmental pressures behind climate change, biodiversity loss and pollution is a complex endeavour, but a vital one to understand the underlying dynamics at play. The previous sections examined the various drivers in the form of the scale, composition and efficiency effects and provided sectoral perspectives. This Annex brings together the underlying socioeconomic trends and specific drivers of the triple planetary crisis to identify how specific indicators of environmental pressures are affected by the various aspects.
The scale of environmental pressures has been growing and is projected to continue growing. In almost all cases, this is (and has been) only partially compensated by composition and efficiency effects, with the former having an ambiguous effect depending on the specific changes in economic structures and geography of production, but especially the latter alleviating environmental pressures. These trends for the drivers of environmental pressures are consistent with most of the literature (UNEP, 2024[68]). Annex Figure 2.B.1 presents decompositions of the major environmental pressures discussed in this chapter, using the basic decomposition into scale, composition and efficiency effects as discussed throughout the chapter. This decomposition is challenging for biodiversity loss, as climate change and pollution are among the key drivers of biodiversity loss (see also Chapter 3). For GHG emissions, plastics, and materials use, a similar story emerges: population and GDP per capita growth (the scale factors) increase the pressures, while structural change (the composition factor) and technology change (the efficiency factor) alleviate environmental pressures both in 2020 and in 2050. The overall impact of structural change for plastics between 2020 and 2050 is positive.17 The contribution of structural change to plastics use growth reflects the evolution of growth projections, the pervasiveness of plastics in sectors that continue to grow, in particular industrial sectors and demand for packaging. In the case of GHG emissions, structural change and technology change are measured by primary energy over GDP (the amount of energy to produce one dollar of value added) and the ratio of GHG to primary energy (the amount of emissions for each GWh of energy used).
The drivers of land and fertiliser use show different dynamics, where specific drivers take precedence. For example, for fertilisers, it is more insightful to take land use as the basic unit rather than agricultural production. Population growth remains a positive driver across all decompositions shown in the figure. The largest driver of both fertiliser use and land use depicts (in opposite directions) the current trend of intensification of agriculture, where the land-saving efficiency is increased through an increase in fertiliser. However, this dynamic does not constitute a free arbitration as the expansion of land is constrained by its availability. Finally, for land use, the largest projected driver of land use expansion corresponds to diet intensity between 2020 and 2050, revealing the efficiency of the food transformation industry as well as the evolution of diets towards more meat.
Contribution of the various factors (in the unit of the indicator) to the 2005-2020 (upper panel) and 2020-2050 (lower panel) evolution of indicators
Note: Scales differ by panel. The term pc stands for per capita.
Source: Environmental Outlook modelling toolbox.
The global story hides significant regional disparities as illustrated in Annex Figure 2.B.2 for GHG emissions. While scale (population growth) and income effects (GDP per capita growth) increase GHG emissions, composition (Primary energy/GDP growth) and technology effects (GHG/Primary Energy) reduce GHG emissions in all regions, although regional differences in the relative magnitude of the drivers are large. For instance, higher-income regions have negligible increases in GHG emissions from population growth. In contrast, population growth in lower-income regions is high, putting upward pressure on the drivers of environmental impacts and therefore GHG emissions. Whilst the population of middle-income regions also grows, their main increase in environmental impacts is driven by economic growth. Economic growth is also a main driver of environmental impacts in lower-income regions, although their growth effect is lower than in middle-income regions. GDP per capita growth drives emissions up in higher-income countries but to a lesser degree than other regions as the economy and consumption turns to more services. Structural change (primary energy to GDP ratio) and technological change (GHG to primary energy ratio) both slow emission growth in all regions. The wealthier regions show stronger decoupling through structural change, increasingly specialising in services, while the less wealthy regions show a larger influence of technological change (illustrating the switch from cheap carbon-intensive energy carriers used for fast economic growth towards a more sustainable energy system, which is already much more advanced in wealthier economies).
Contribution of the various factors (in Gt CO2e)
Source: Environmental Outlook modelling toolbox.
This decomposition is useful to paint broad strokes but has limitations. For instance, a caveat of the analysis for GHG emissions is that this decomposition mostly shows the influence of energy, when other sector-specific emissions (i.e. from agriculture and industrial processes) also contribute to total emissions, whose drivers do not appear in the current graph. The more diverse the system, the harder it becomes to show synthetic indicators.
Overall, the comparisons of GDP growth and drivers of environmental pressures paint a stark picture of the challenge to overcome: in most regions and for most indicators, the environmental pressures continue to rise in the coming decades. This should however be seen in a context of continued economic growth. Most regional dynamics follow a relative decoupling, i.e. environmental degradation continues at a slower pace than that of economic growth.18 This implies that under current policies the pressures on the environment continue to increase, even if the pressure per unit of economic activity declines over time. Consequently, the risks of crossing important tipping points also continue to increase, even if the environmental degradation trend is lower than that of economic growth.
Most drivers of environmental pressures are projected to reach a relative decoupling, with primary PM2.5 emissions projected to show absolute decoupling (a decrease from 2020 to 2050). The first line of Annex Figure 2.B.2 shows the evolution of the indexes of the main drivers of climate change, biodiversity loss and pollution (with a global value of 1 in 2020). The indicators related to biodiversity loss, climate change, and plastic and nutrient pollution show an increase by 2050, at various degrees, but lower than that of GDP. This implies a relative decoupling. However, the degree of this relative decoupling varies by indicator: land use growth is small (13%), being very constrained, primary energy grows by 36%, nitrogen (N) fertiliser use grows by 43%, while plastic pollution more than doubles (+106%), almost on par with GDP. In contrast, the index for primary PM2.5 shows an absolute decrease globally.
This result overall holds over the 3 aggregate regions presented in this report. The second to fourth lines of Annex Figure 2.B.2 shows the evolution of the indexes of these main drivers (the initial value in 2020 representing the share in the global driver, the lines comparing to GDP). However, primary energy shows an absolute decoupling in the higher-income region. As at the global level, PM2.5 decrease occurs in all regions.
The largest increases for these drivers of environmental pressures are shown for the lower-income region. In that region, cropland increases by 18%. Plastics use more than triples, while primary energy more than doubles and nitrogen fertiliser increases by 52%. The only decrease occurs for PM2.5 (-36%).
Index 1 for the global value in 2020 (for each indicator)
Note: Initial value for other regions represents the share in the global value, growth is by region compared to GDP growth (compared to line).
Source: Environmental Outlook modelling toolbox.
The main implications of economic growth, at least since the industrial revolution, have been growing demand for fossil fuels and other resources, significant land use changes and larger generation of waste. However, in the long run, higher GDP per capita could lead to larger willingness to pay to reduce environmental impacts as basic needs are being fulfilled and poverty reduced. The relationship between environmental outcomes and income has been studied for decades. Notably, it was empirically investigated by Grossman and Krueger (1991[3]) and described later as environmental Kuznets curves (EKC) by Panayotou (1993[90]). The traditional model of EKC is an inverted U-shaped curve where environmental impacts tend to grow as income levels rise, reaching a peak, and subsequently declining after exceeding a certain income threshold.
GHG emissions per capita and GDP per capita follows an inverted U-shaped curve as illustrated by Annex Figure 2.B.2 based on country level data on the period 1990-2020. This result for GHG is consistent with the majority of previous work (Shahbaz and Sinha, 2019[91]; Wang, Wang and Jiang, 2024[92]). Each country follows its own path and has a specific initial position but on average, GHG emissions per capita rapidly surge then decelerate then stop increasing at USD 45 000 2017 PPP per capita, after which they decreases.
The shape of the EKC depends very much on the environmental impacts and pollutants analysed as illustrated by Grossman and Krueger (1995[93]) for urban air pollution, oxygen regime, faecal contamination, heavy metal contamination and nutrients in rivers using data at the city-country level. Another example of how EKC are pollutant dependent is plastic leakage. While macroplastic leakage increases with GDP per capita and then rapidly decreases after USD 10 000 per capita, microplastic leakage continues to grow with GDP per capita with decreasing marginal increases (OECD, 2022[12]). For biodiversity conservation, the evidence of an inverted U-shape relation with income is mixed and inconclusive (Dietz and Adger, 2003[94]; McPherson and Nieswiadomy, 2005[95]; Mills and Waite, 2009[96]).
Note: Total GHG emissions in Gt of CO2 equivalent are composed of CO2 totals excluding short-cycle biomass burning (such as agricultural waste burning and savanna burning) but including other biomass burning (such as wildfires, post-burn decay, peat fires and decay of drained peatlands), all anthropogenic CH4 sources, N2O sources and F-gases (HFCs, PFCs and SF6). The world average curve is estimated using a quadratic function and using data from all countries via an OLS regression including country and year fixed-effects. The right shift in income observed for various countries in 2020 is due to the covid-19 crisis.
Source: Authors calculation based on data from the World Development Indicators 2022 (https://data.worldbank.org/products/wdi) and from the ENV-Growth model.
Environmental Kuznets curves are seen more as tools to visualise the joint evolution of environmental outcomes and income. However their main limitation is that they do not provide clear information on the causal effect of income as stressed by Stern (2004[97]; 2017[98]) nor on interlinked drivers behind this evolution, which is highly important for analysing the triple planetary crisis. For example, the reduction in domestic environmental impacts can be either due to the adoption of stricter environmental policies, cleaner technologies or the outsourcing abroad of economic activities that have high environmental impacts, all correlated with higher income.
Figure 2.12 and Figure 2.13 are based on several indicators reporting the global level and the regional intensities of selected drivers of environmental changes. Annex Table 2.C.1 summarises the units of these indicators. and Annex Table 2.C.2 provides the values of these indicators.
Note: Indicators on the level of the pressures determine the length of the bars, while indicators on the regional intensity determine the colour shade of the bars in regional plots.
Global evolution for selected drivers
Note: The length of the bars represents evolution over time indexed at 100% in 2020. For instance, for Panel A, the circle indicates the 2020 level indexed to 100%, against which to compare the evolution in the bar in Panel B. For example, plastics use is project to increases by 106% over the 2020-2050 period at the global level. Annex 2.C provides details on the values and the units of reported indicators.
Source: Environmental Outlook modelling toolbox.
Note: Values for 2020 and intensities for 2050 are reported in the units listed in Annex Table 2.C.1, while the relative change in 2050 compared to 2020 is presented as a percentage. For example, in the Central & South America region, the GDP in 2020 was USD 9.5 trillion (2017 USD); this is projected to increase by 128% by 2050, resulting in an intensity indicator measured as GDP per capita of USD 28 829 (2017 USD) per person in 2050. In Figure 2.12, the length of the bars represents the relative change between 2020 and 2050, while in Figure 2.13, the colour shading of the bars reflects the intensity in 2050.
References
[73] Adhikari, S., W. Lutz and E. Kebede (2024), “Forecasting Africa’s fertility decline by female education groups”, Proceedings of the National Academy of Sciences, Vol. 121/46, p. e2320247121, https://doi.org/10.1073/PNAS.2320247121.
[51] Baldé, C. et al. (2024), “Global E-waste Monitor 2024”, International Telecommunication Union (ITU) and United Nations Institute for Training and Research (UNITAR).
[64] Benton, T. et al. (2021), “Food system impacts on biodiversity loss: Three levers for food system transformation in support of nature”, Chatham House, Research Paper, Energy, Environment and Resources Programme.
[35] Bren d’Amour, C. et al. (2016), “Future urban land expansion and implications for global croplands”, Proceedings of the National Academy of Sciences, Vol. 114/34, pp. 8939-8944, https://doi.org/10.1073/pnas.1606036114.
[65] Campbell, B. et al. (2017), “Agriculture production as a major driver of the Earth system exceeding planetary boundaries”, Ecology and Society, Vol. 22/4, https://doi.org/10.5751/ES-09595-220408.
[49] Cárdenas Rodríguez, M., L. Dupont-Courtade and W. Oueslati (2015), “Air Pollution and Urban Structure Linkages: Evidence from European Cities”, OECD Environment Working Papers, No. 96, OECD Publishing, Paris, https://doi.org/10.1787/5jrp6w9xlbq6-en.
[46] Castells-Quintana, D., E. Dienesch and M. Krause (2021), “Air pollution in an urban world: A global view on density, cities and emissions”, Ecological Economics, Vol. 189, p. 107153.
[57] Coelho, P., J. Teixeira and O. Gonçalves (2011), “Mining Activities: Health Impacts”, in Encyclopedia of Environmental Health, Elsevier, https://doi.org/10.1016/b978-0-444-52272-6.00488-8.
[71] de la Maisonneuve, C. and J. Oliveira Martins (2014), “The future of health and long-term care spending”, OECD Journal: Economic Studies, https://doi.org/10.1787/eco_studies-2014-5jz0v44s66nw.
[75] Dellink, R. et al. (2017), “Long-term economic growth projections in the Shared Socioeconomic Pathways”, Global Environmental Change, Vol. 42, pp. 200-214, https://doi.org/10.1016/j.gloenvcha.2015.06.004.
[58] Dietler, D. et al. (2021), “Impact of mining projects on water and sanitation infrastructures and associated child health outcomes: a multi-country analysis of Demographic and Health Surveys (DHS) in sub-Saharan Africa”, Globalization and Health, Vol. 17/1, https://doi.org/10.1186/s12992-021-00723-2.
[94] Dietz, S. and W. Adger (2003), “Economic growth, biodiversity loss and conservation effort.”, Journal of environmental management, Vol. 68/1, pp. 23-35.
[62] Eldridge, D., J. Ding and S. Travers (2022), “A Global Synthesis of the Effects of Livestock Activity on Hydrological Processes”, Ecosystems, Vol. 25/8, pp. 1780-1791, https://doi.org/10.1007/s10021-022-00746-9.
[28] FAO (2024), Land statistics 2001–2022, FAO, https://doi.org/10.4060/cd1484en.
[29] FAO (2024), The State of the World’s Forests 2024, FAO, https://doi.org/10.4060/cd1211en.
[63] FAO (2023), “Pathways towards lower emissions – A global assessment of the greenhouse gas emissions and mitigation options from livestock agrifood systems”, Rome, https://doi.org/10.4060/cc9029en.
[16] FAO (2023), “The State of Food and Agriculture 2023”, in The State of Food and Agriculture 2023 – Revealing the true cost of food to transform agrifood systems, FAO, https://doi.org/10.4060/cc7724en.
[21] FAO (2020), Valuation of the health and climate-change benefits of healthy diets, FAO, https://doi.org/10.4060/cb1699en.
[66] FAO (2019), “Water use in livestock production systems and supply chains - Guidelines for assessment (Version 1)”, Livestock Environmental Assessment and Performance (LEAP) Partnership. Rome..
[69] Giljum, S. et al. (2022), “A pantropical assessment of deforestation caused by industrial mining”, Proceedings of the National Academy of Sciences, Vol. 119/38, https://doi.org/10.1073/pnas.2118273119.
[54] Goldenman, G. et al. (2019), The cost of inaction, Nordic Council of Ministers, Copenhagen, https://doi.org/10.6027/tn2019-516.
[60] Gouel, C. and H. Guimbard (2018), “Nutrition Transition and the Structure of Global Food Demand”, American Journal of Agricultural Economics, Vol. 101/2, pp. 383-403, https://doi.org/10.1093/ajae/aay030.
[93] Grossman, G. and A. Krueger (1995), “Economic growth and the environment.”, The quarterly journal of economics, Vol. 110/2, pp. 353-377.
[3] Grossman, G. and A. Krueger (1991), Environmental Impacts of a North American Free Trade Agreement, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w3914.
[38] Haddad, N. et al. (2015), “Habitat fragmentation and its lasting impact on Earth’s ecosystems.”, Science advances, Vol. 1/2, p. e1500052.
[27] Hansen, M. et al. (2013), “High-Resolution Global Maps of 21st-Century Forest Cover Change”, Science, Vol. 342/6160, pp. 850-853, https://doi.org/10.1126/science.1244693.
[10] Hellweg, S. et al. (2023), “Life-cycle assessments to guide solutions for the triple planetary crisis”, Nature Reviews Earth & Environment, Vol. 4, pp. 471-486, https://doi.org/10.1038/s43017-023-00449-2.
[26] Houghton, R. and A. Nassikas (2017), “Global and regional fluxes of carbon from land use and land cover change 1850–2015”, Global Biogeochemical Cycles, Vol. 31/3, pp. 456-472, https://doi.org/10.1002/2016gb005546.
[67] Huber, V. et al. (2014), “Will the world run out of land? A Kaya-type decomposition to study past trends of cropland expansion”, Environmental Research Letters, Vol. 9/2, p. 024011, https://doi.org/10.1088/1748-9326/9/2/024011.
[25] Hurtt, G. et al. (2011), “Harmonization of land-use scenarios for the period 1500–2100: 600 years of global gridded annual land-use transitions, wood harvest, and resulting secondary lands”, Climatic Change, Vol. 109/1-2, pp. 117-161, https://doi.org/10.1007/s10584-011-0153-2.
[82] ILO (2022), Indicators and data, International Labour Organization, https://rplumber.ilo.org/files/website/bulk/indicator.html.
[84] IMF (2023), World Economic Outlook: Navigating Global Divergences, International Monetary Fund, https://www.imf.org/-/media/Files/Publications/WEO/2023/October/English/text.ashx.
[6] Iovino, L. (2023), Corporate Taxation and Carbon Emissions, https://cepr.org/publications/dp17915.
[9] IPBES (2024), The thematic assessment report on interlinkages among biodiversity, water, food and health. Summary for policymakers, https://ipbes.canto.de/pdfviewer/viewer/viewer.html?v=IPBES11Media&portalType=v%2FIPBES11Media&column=document&id=cj0uc5396d1ed5418tsuic2r45&suffix=pdf&print=1.
[18] IPBES (2019), “Global assessment report of the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services, Brondízio, E. S., Settele, J., Díaz, S., Ngo, H. T. (eds).”, IPBES secretariat, Bonn, Germany. 1144 pages. ISBN: 978-3-947851-20-1, https://doi.org/10.5281/zenodo.3553579.
[8] IPCC (2023), Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland., Intergovernmental Panel on Climate Change (IPCC), https://doi.org/10.59327/ipcc/ar6-9789291691647.
[45] IPCC (2022), “Climate Change 2022: Impacts, Adaptation and Vulnerability. Contribution of Working Group II to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change”, Cambridge University Press. Cambridge University Press, Cambridge, UK and New York, NY, USA, 3056, https://doi.org/10.1017/9781009325844.
[31] IPCC (2022), “Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change”, Cambridge University Press, Cambridge, UK and New York, NY, USA., https://doi.org/10.1017/9781009157926.
[7] IPCC (2021), “Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change”, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2391, https://doi.org/10.1017/9781009157896.
[74] Johansson, Å. et al. (2013), “Long-Term Growth Scenarios”, OECD Economics Department Working Papers, No. 1000, OECD Publishing, Paris, https://doi.org/10.1787/5k4ddxpr2fmr-en.
[2] Kaya, Y. and K. Yokobori (1997), “Environment, energy, and economy : strategies for sustainability”, p. 381.
[72] KC, S. and W. Lutz (2017), “The human core of the shared socioeconomic pathways: Population scenarios by age, sex and level of education for all countries to 2100”, Global Environmental Change, Vol. 42, pp. 181-192, https://doi.org/10.1016/j.gloenvcha.2014.06.004.
[43] Ke, X. et al. (2018), “Direct and indirect loss of natural habitat due to built-up area expansion: a model-based analysis for the city of Wuhan, China”, Land Use Policy, Vol. 74, pp. 231–239.
[89] Kok, M. et al. (2023), “Assessing ambitious nature conservation strategies in a below 2-degree and food-secure world”, Biological Conservation, Vol. 284, p. 110068, https://doi.org/10.1016/j.biocon.2023.110068.
[52] Landrigan, P. et al. (2018), “The Lancet Commission on pollution and health”, The Lancet, Vol. 391/10119, pp. 462-512, https://doi.org/10.1016/s0140-6736(17)32345-0.
[4] Lankoski, J., E. Nales and H. Valin (2025), “Assessing the impacts of agricultural support policies on the environment: Economic analysis, literature findings and synthesis”, OECD Food, Agriculture and Fisheries Papers, No. 223, OECD Publishing, Paris, https://doi.org/10.1787/808f110c-en.
[33] Lemonsu, A. et al. (2015), “Vulnerability to heat waves: Impact of urban expansion scenarios on urban heat island and heat stress in Paris (France)”, Urban Climate, Vol. 14, pp. 586-605.
[44] Liu, X. et al. (2019), “Global urban expansion offsets climate-driven increases in terrestrial net primary productivity.”, Nature communications, Vol. 10/1, p. 5558.
[39] Liu, Z., C. He and J. Wu (2016), “The relationship between habitat loss and fragmentation during urbanization: an empirical evaluation from 16 world cities.”, PloS one, Vol. 11/4, p. e0154613.
[24] Li, W. et al. (2018), “Gross and net land cover changes in the main plant functional types derived from the annual ESA CCI land cover maps (1992–2015)”, Earth System Science Data, Vol. 10/1, pp. 219-234, https://doi.org/10.5194/essd-10-219-2018.
[99] Ludemann, C. et al. (2024), “A global FAOSTAT reference database of cropland nutrient budgets and nutrient use efficiency (1961–2020): nitrogen, phosphorus and potassium”, Earth System Science Data, Vol. 16/1, pp. 525-541, https://doi.org/10.5194/essd-16-525-2024.
[50] Macklin, M. et al. (2023), “Impacts of metal mining on river systems: a global assessment”, Science, Vol. 381/6664, pp. 1345-1350, https://doi.org/10.1126/science.adg6704.
[48] McCarty, J. and N. Kaza (2015), “Urban form and air quality in the United States”, Landscape and Urban Planning, 139, Vol. 139, pp. 168-179.
[37] McDonald, R. et al. (2020), “Research gaps in knowledge of the impact of urban growth on biodiversity”, Nature Sustainability, Vol. 3/1, pp. 16–24.
[95] McPherson, M. and M. Nieswiadomy (2005), “Environmental Kuznets curve: threatened species and spatial effects”, Ecological Economics, Vol. 55/3, pp. 395-407.
[70] Meißner, S. (2021), “The impact of metal mining on global water stress and regional carrying capacities—a GIS-based water impact assessment.”, Resources, Vol. 10/12, p. 120.
[11] Meng, F., M. Brandão and J. Cullen (2024), “Replacing Plastics with Alternatives Is Worse for Greenhouse Gas Emissions in Most Cases”, Environmental Science & Technology, Vol. 58/6, pp. 2716-2727, https://doi.org/10.1021/acs.est.3c05191.
[96] Mills, J. and T. Waite (2009), “Economic prosperity, biodiversity conservation, and the environmental Kuznets curve.”, Ecological Economics, Vol. 68/7, pp. 2087-2095.
[81] Nascimento, L. et al. (2023), Greenhouse gas mitigation scenarios for major emitters - Analysis of current climate policies and mitigation commitments: 2023 update, https://newclimate.org/sites/default/files/2023-11/NewClimate_PBL2023_CurrentPolicies.pdf.
[79] Nascimento, L. et al. (2024), Progress of major emitters towards climate targets: 2024 Update, NewClimate Institute, Cologne, Germany, https://www.pbl.nl/en/publications/progress-of-major-emitters-towards-climate-targets.
[42] Newbold, T. et al. (2015), “Global effects of land use on local terrestrial biodiversity”, Nature, Vol. 520/7545, pp. 45–50.
[20] OECD (2025), Environmental Performance of Agriculture in OECD Countries: Key Trends and Insights, OECD Publishing, Paris, https://doi.org/10.1787/2679ba38-en.
[17] OECD (2025), “Targeting methane emissions to mitigate the risk of climate overshoot”, OECD Net Zero+ Policy Papers, No. 14, OECD Publishing, Paris, https://doi.org/10.1787/5fa37719-en.
[22] OECD (2024), Agricultural Policy Monitoring and Evaluation 2024: Innovation for Sustainable Productivity Growth, OECD Publishing, Paris, https://doi.org/10.1787/74da57ed-en.
[78] OECD (2024), OECD Digital Economy Outlook 2024 (Volume 1): Embracing the Technology Frontier, OECD Publishing, Paris, https://doi.org/10.1787/a1689dc5-en.
[53] OECD (2023), Economic instruments to incentivise substitution of chemicals of concern – a review, OECD Series on Risk Management of Chemicals, OECD Publishing, Paris, https://doi.org/10.1787/d73c0786-en.
[83] OECD (2023), “OECD Economic Outlook No 114 (Edition 2023/2)”, OECD Economic Outlook: Statistics and Projections (database), https://doi.org/10.1787/62ec50a4-en (accessed on 12 January 2024).
[5] OECD (2022), Agricultural Policy Monitoring and Evaluation 2022: Reforming Agricultural Policies for Climate Change Mitigation, OECD Publishing, Paris, https://doi.org/10.1787/7f4542bf-en.
[12] OECD (2022), Global Plastics Outlook: Economic Drivers, Environmental Impacts and Policy Options, OECD Publishing, Paris, https://doi.org/10.1787/de747aef-en.
[56] OECD (2022), Global Plastics Outlook: Policy Scenarios to 2060, OECD Publishing, Paris, https://doi.org/10.1787/aa1edf33-en.
[15] OECD (2019), Global Material Resources Outlook to 2060: Economic Drivers and Environmental Consequences, OECD Publishing, Paris, https://doi.org/10.1787/9789264307452-en.
[77] OECD (2019), Measuring the Digital Transformation: A Roadmap for the Future, OECD Publishing, Paris, https://doi.org/10.1787/9789264311992-en.
[40] OECD (2018), “Rethinking Urban Sprawl: Moving Towards Sustainable Cities”, OECD Publishing, Paris, https://doi.org/10.1787/9789264189881-en.
[59] OECD/FAO (2025), OECD-FAO Agricultural Outlook 2025-2034, OECD Publishing, Paris/Food and Agriculture Organization of the United Nations, Rome, https://doi.org/10.1787/601276cd-en.
[19] OECD/FAO (2024), OECD-FAO Agricultural Outlook 2024-2033, OECD Publishing, Paris/Food and Agriculture Organization of the United Nations, Rome, https://doi.org/10.1787/4c5d2cfb-en.
[90] Panayotou, T. (1993), “Empirical tests and policy analysis of environmental degradation at different stages of economic development”, International Labour Office Working Paper.
[76] Pilat, D. and A. Nolan (2016), “Benefiting from the next production revolution”, in Debate the Issues: New Approaches to Economic Challenges, OECD Publishing, Paris, https://doi.org/10.1787/9789264264687-22-en.
[61] Poore, J. and T. Nemecek (2018), “Reducing food’s environmental impacts through producers and consumers”, Science, Vol. 360/6392, pp. 987-992, https://doi.org/10.1126/science.aaq0216.
[87] Rao, S. et al. (2017), “Future air pollution in the Shared Socio-economic Pathways”, Global Environmental Change, Vol. 42, pp. 346-358, https://doi.org/10.1016/j.gloenvcha.2016.05.012.
[1] Riahi, K. et al. (2017), “The Shared Socioeconomic Pathways and their energy, land use, and greenhouse gas emissions implications: An overview”, Global Environmental Change, Vol. 42, pp. 153-168, https://doi.org/10.1016/j.gloenvcha.2016.05.009.
[41] Seto, K., B. Güneralp and L. Hutyra (2012), “Global forecasts of urban expansion to 2030 and direct impacts on biodiversity and carbon pools”, Proceedings of the National Academy of Sciences, Vol. 109/40, pp. 16083-16088, https://doi.org/10.1073/pnas.1211658109.
[91] Shahbaz, M. and A. Sinha (2019), “Environmental Kuznets curve for CO2 emissions: a literature survey.”, Journal of Economic Studies, Vol. 46/1, pp. 106-168.
[97] Stern, D. I. (2004), “The rise and fall of the environmental Kuznets curve.”, World development, Vol. 32/8, pp. 1419-1439.
[98] Stern, D. (2017), “The environmental Kuznets curve”, In Oxford Research Encyclopedia of Environmental Science.
[47] Stone, J. (2008), “Urban sprawl and air quality in large US cities”, Journal of environmental management, Vol. 86/4, pp. 688-698.
[13] Tacker, M. et al. (2025), “Life cycle assessment of polyethylene packaging and alternatives on the European market”, Cleaner Environmental Systems, Vol. 17, p. 100270, https://doi.org/10.1016/j.cesys.2025.100270.
[30] UN (2022), World Population Prospects 2022. Summary of Results, https://www.un.org/development/desa/pd/sites/www.un.org.development.desa.pd/files/wpp2022_summary_of_results.pdf.
[68] UNEP (2024), “Global Resources Outlook 2024: Bend the Trend – Pathways to a liveable planet as resource use spikes.”, International Resource Panel. United Nations Environment Programme. Nairobi, https://wedocs.unep.org/20.500.11822/44902.
[55] UNEP (2019), Global Chemicals Outlook II: From Legacies to Innovative Solutions, https://www.unep.org/resources/report/global-chemicals-outlook-ii-legacies-innovative-solutions (accessed on 22 January 2025).
[32] UNEP (2019), Global Environment Outlook (GEO-6): Healthy Planet, Healthy People.
[80] UNEP-WCMC (2020), Protected areas map of the world, http://www.protectedplanet.net.
[88] Van der Esch, S. et al. (2022), The global potential for land restoration: Scenarios for the Global Land Outlook 2, PBL Netherlands Environmental Assessment Agency, The Hague, https://www.pbl.nl/en/publications/the-global-potential-for-land-restoration-scenarios-for-the-global-land-outlook-2.
[36] van Vliet, J. (2019), “Direct and indirect loss of natural area from urban expansion”, Nature Sustainability, Vol. 2/8, pp. 755-763., https://doi.org/10.1038/s41893-019-0340-0.
[86] van Vuuren, D. et al. (2017), “Energy, land-use and greenhouse gas emissions trajectories under a green growth paradigm”, Global Environmental Change, Vol. 42, pp. 237-250, https://doi.org/10.1016/j.gloenvcha.2016.05.008.
[92] Wang, Q., X. Wang and X. Jiang (2024), “Reinvestigating the environmental Kuznets curve (EKC) of carbon emissions and ecological footprint in 147 countries: a matter of trade protectionism.”, Humanities and Social Sciences Communications, Vol. 11/1, pp. 1-17.
[23] Winkler, K. et al. (2021), “Global land use changes are four times greater than previously estimated”, Nature Communications, Vol. 12/1, https://doi.org/10.1038/s41467-021-22702-2.
[14] Woods, J. et al. (2021), “A framework for the assessment of marine litter impacts in life cycle impact assessment”, Ecological Indicators, Vol. 129, pp. 1-15, https://doi.org/10.1016/j.ecolind.2021.107918.
[85] World Bank (2023), World Development Indicators, https://datatopics.worldbank.org/world-development-indicators/.
[34] Zhang, W. et al. (2018), “Urbanization exacerbated the rainfall and flooding caused by hurricane Harvey in Houston”, Nature, Vol. 563, pp. 384–388.
← 1. Socioeconomic trends can also affect environmental pressures through indirect effects. For instance, landscape fragmentation and climate change are major drivers of biodiversity loss. Clarifying these indirect relations is fundamental in mapping out the synergies and trade-offs between climate change, biodiversity loss, and pollution. However, as they involve interactions between environmental states rather than drivers, they are addressed in the next chapters.
← 2. Figure 1.3 in Chapter 1 provides the fuller picture, including the feedback effects of environmental degradation.
← 3. The challenge for agricultural policymaking is to address multiple challenges simultaneously: supporting the food system and ending hunger, while addressing several global issues such as environmental degradation and distortions in international markets (OECD, 2022[5]).
← 4. Agricultural systems with lower input use often have better overall environmental performance per hectare (e.g. lower pollution to water bodies). However, these systems typically have lower yields as well. This implies that all else equal, more land would be needed to produce the same amount of output.
← 5. These numbers differ from e.g. those by FAO (Ludemann et al., 2024[99]) that focus on nutrient surpluses for croplands, not accounting for grassland surpluses.
← 6. These aggregate results show overall trends, but extract from the variability at the country level and even further when looking at spatial results. Parts of Africa in particular can show negative phosphorus surplus, in line with the current scientific findings (Ludemann et al., 2024[99]). While all numbers are subject to uncertainty, the contribution of grasslands to phosphorus surpluses is highly uncertain because the uptake of grass is not very well known.
← 7. Such practices, and other land use changes, can also have mixed effects – and thus trade-offs – on different ecosystem services; for instance between carbon storage and plant diversity.
← 8. This selection of sectors does not imply that the environmental impact of other economic activity is negligible. Rather, they are meant as illustrative for the link between economic activity and the pressures on the environment, highlighting a number of salient issues. When considering substitution to alternatives to the highlighted commodities, a full LCA is required to ensure that the overall outcome improves; in some cases, the environmental profile of substitutes is better, but in other cases it may be worse.
← 9. While chemical intensity is an imperfect measure (being based on monetary values, a small intensity can simply reflect relatively high output unitary prices), it provides useful comparative insights when interpreted alongside other indicators.
← 10. While biomass and fossil fuels are partly covered in the section on agriculture and energy, material resources include all materials, including these categories. In this section, these categories have a broader scope: fossil fuels include all resources used (and not just energy), including chemicals and plastics. Similarly, biomass include textiles, grazed biomass, fishing in addition to conventional agricultural commodities. This definition of material resources is in line with the literature on the topic (OECD, 2019[15]; UNEP, 2024[68]).
← 11. See https://www.oecd.org/en/data/indicators/water-withdrawals.html for more information.
← 12. A typical example is water withdrawal for cooling of nuclear power plants.
← 13. As the visual depiction of growth factors is capped at 140%, Figure 2.13 does not fully represent the scale of the growth factors in lower-income regions.
← 14. Increases in the retirement age, as implemented recently in some countries, can alleviate this, but in most cases life expectancy continues to outpace increases in retirement age. Furthermore, longer schooling means young people enter the labour market at a later age.
← 15. In a long-term balanced growth path, capital accumulation would adjust to match growth in labour and technology. This only happens when countries are at the technology frontier and when capital supply is not limited. The fact that the contribution of technology growth is higher than that of capital even in higher-income countries shows that for the coming decades, all regions are still in transition towards such a (very) long-term balanced growth path.
← 16. Sustainable economic growth achieves this by not increasing the associated environmental inputs, nor the environmental pressures associated with production. This subsection does not aim to calculate an environmentally-adjusted productivity growth but rather presents the economic drivers such that later subsections can illuminate the associated environmental aspects.
← 17. This shows a slight change in trends since the Global Plastics Outlook (OECD, 2022[54]) which highlights that plastics are used in a wide variety of economic sectors with differing growth patterns.
← 18. Sometimes decoupling can be described in the form of Environmental Kuznets Curve, illustrated in Annex Box 2.B.1 based on GHG emissions.