Investing in Climate for Growth and Development

5. Ensuring a just and equitable transition

Abstract

Climate change mitigation policies offer long-term benefits – especially for low-income groups, women, the elderly and rural populations, who are likely to be most affected by climate impacts. Yet – these same groups may face short-term challenges from rising costs or reduced incomes. This chapter reviews the distributional and labour market effects of the climate change mitigation policies compatible with the goals of the Paris Agreement on OECD countries. OECD economic and microsimulation modelling shows that climate change mitigation policies can be regressive in the short term, underscoring the need for measures to assess and offset these effects. Although overall labour market impacts are modest, sectoral and occupational shifts will require support, such as skills development. The chapter also explores policy options to help households and workers, and to ensure a fair sharing of costs and benefits of climate mitigation policies.

5.1.  The distributional effects of climate action can be addressed with the right policy mix

5.1.1. Anticipating distributional impacts is crucial to ensure support for climate mitigation

Concerns over costs induced by climate change mitigation policies loom large among households. Households indicate that economic factors (such as unemployment, inflation or poverty) rank higher in public concern than environmental ones (OECD, 2023[1]), and voters may not support climate action if they perceive that it will create significant costs for them. Further, an OECD survey (Dechezleprêtre et al., 2025[2]) suggests that people will oppose climate action when they perceive that the burdens and opportunities are shared unfairly, e.g. with burdens disproportionately falling on disadvantaged groups. However, climate policies are shown to have differentiated effects by income group (Lamb et al., 2020[3]; Markkanen and Anger-Kraavi, 2019[4]), as well as within income groups, e.g. by gender, employment status, age, family size and between urban and rural areas (OECD, 2024[5]). Understanding such distributional effects and how to mitigate them is necessary to ensure support for climate mitigation policy strategy.

There are multiple channels through which climate change mitigation policies impact households. Price changes, especially for essential goods such as energy and food, can affect household consumption and livelihoods. In addition, climate change mitigation policies can alter the incomes of workers and asset owners, by shaping the returns to different production factors, including labour, natural resources and equity held in “green” or “brown” industries (Rausch, Metcalf and Reilly, 2011[6]). As new production technologies are deployed and structural economic changes occur, job markets shift, affecting wages and sectoral employment. These effects are likely to be concentrated in specific groups of workers (Section 5.3), and entail potentially sizeable income changes for them, notably in the case of job loss and extended unemployment.

The effects of climate policies can be regressive and may result in significant short-term burdens, especially for disadvantaged groups. In the short term, low-income households generally see much greater burdens from mitigation policies than others. Higher energy prices pose greater challenges for households who spend a higher share of their consumption on energy (e.g. for heating and cooking) and who may be ill-equipped to face price increases by drawing on savings, cutting back on other expenditures, or reducing reliance on high-emission products (OECD, 2022[7]; OECD, 2024[5]). Limited access to credit can prevent households from switching to cleaner technologies, making the transition more difficult. In certain cases, low‑income groups and workers may fare better from abatement measures than wealthier households and capital owners. For example, higher-income households are more likely to own cars. Consequently, policies targeting emissions from private vehicles could affect higher-income households more, therefore reducing inequality. Additionally, because capital-intensive goods tend to be emission‑intensive, climate policies could reduce returns to capital relative to wage growth. However, even when climate policies do not disproportionately burden low-income households, they can still create social challenges, such as regional job losses. As low‑income households tend to consume little overall, even a minor increase in costs can have a substantial impact on their well-being. This underscores the critical need for carefully assessing these effects and for making support and compensation an integral part of climate policies.

Well-designed policy packages can mitigate adverse effects and may even reduce overall inequality. Accompanying measures, such as the use of carbon-pricing revenue, e.g. via transfers or tax cuts, can mitigate adverse distributional consequences of climate policies. Likewise, choosing the right mix of climate policies can help to address social concerns, as the impact on households varies across different policy instruments and countries (Box 5.1). For instance, in countries where energy is mostly consumed by high income groups – as is the case for Mexico – carbon pricing can be progressive (OECD, 2024[5]). Further, the distributional effect of climate policies will also depend on sectoral responses to the policies as well as on the broader policy environment that influences income. For example, climate policies can be progressive if government transfers to lower‑income households – like pensions or unemployment benefits – are indexed to inflation or cost-of-living increases (Fullerton, Heutel and Metcalf, 2011[8]; Goulder et al., 2019[9]). In such cases, lower- income households are partly shielded from the impacts of climate-policy related price increases.

Box 5.1. Distributional impacts of supply and demand-side mitigation policies

Distributional impact of demand-side policies

Energy efficient and clean technologies play a central role in the climate change – mitigation agenda. Demand-side policies, including subsidies and related incentives (such as preferential feed-in tariffs) tend to accelerate technology adoption and diffusion, and can be politically attractive (Giraudet, Guivarch and Quirion, 2011[10]; Douenne and Fabre, 2022[11]). Yet, assessments of past measures generally show that they are regressive and generally more so than carbon pricing, as they primarily benefit higher-income households with the necessary capital to invest in low-emitting assets (Lihtmaa, Hess and Leetmaa, 2018[12]; Lekavičius et al., 2020[13]; Winter and Schlesewsky, 2019[14]; West, 2004[15]; Levinson, 2019[16]). Findings differ, however, across technologies, with more regressive impacts of subsidies for electric vehicles than for home insulation or solar panels, and little correlation between heat pump adoption and income (Borenstein and Davis, 2016[17]; Davis, 2023[18]). Design features of subsidies or tax credits, such as refundability, timing and targeting, all shape distributional impacts (Giraudet, Bourgeois and Quirion, 2021[19]; Borenstein and Davis, 2016[17]). Outright bans on the demand side are relatively common in Europe, placing restrictions on the use of cars or certain types of residential heating (Braungardt et al., 2023[20]). Bans also raise equity issues, e.g. by creating large and possibly unaffordable asset-replacement costs for the poorest, unless bans are combined with targeted exemptions or compensation (Torné and Trutnevyte, 2024[21]).

Distributional impact of supply-side measures

Supply-side measures shape production processes via regulation or through subsidies, such as those provided for the European Union’s Net-Zero Industry Act. Comprehensive studies are not yet available but there is some initial evidence of progressive impacts of “supply-push” policies that form part of such packages (Brown et al., 2023[22]). Regulatory approaches can take the form of targeted measures, such as building energy codes, fuel economy standards and vehicle pollution-control, including outright bans of high-emission technologies, with some evidence of high burdens for lower-income households (Davis and Knittel, 2019[23]; Jacobsen, 2013[24]; West, 2009[25]; Bruegge, Deryugina and Myers, 2019[26]). Regulation can also take the form of encompassing packages involving multiple levers, such as the US Clean Air Act (Robinson, 1985[27]) and equivalent provisions in other countries. The scope of these packages varies, as do their distributional impacts, with some evidence of regressive effects (Levinson, 2019[16]).

Source: (OECD, 2024[5]).

Regular distributional assessments could inform the design, packaging and sequencing of climate change mitigation policies beyond documenting the burdens from mitigation measures that are already planned or implemented. All climate policies have distributional consequences (Peñasco, Anadón and Verdolini, 2021[28]). However, while the impacts of carbon pricing are often discussed in the literature (OECD, 2024[5]), distributional effects of other mitigation policies have been studied less, especially in the ex-ante literature. As countries’ climate policies evolve, combining a range of different policy instruments, there is a strong and growing need for careful and comprehensive distributional assessments. Without comprehensive, up-to-date evidence on distributional impacts, some provisions, such as exemptions or preferential treatment of certain commodities or sectors, may be implemented on the basis of unspecified or assumed household burdens, which can be costly, and a weak foundation for effective climate action (Box 5.2). Robust and timely evidence on distributional impacts is essential for identifying provisions that may be needed for equity reasons and those that should be adapted or reconsidered.

Box 5.2. Policy choices when there are trade-offs between climate and equity objectives

There can be trade-offs between climate and equity objectives, as the policies that are most effective at reducing emissions are not the most effective at limiting negative distributional impacts. Such trade‑offs shape not only the cost, but also the political feasibility of climate policies and their design.

For instance, equity considerations are one rationale for varying carbon prices by fuel type or sector, even when it increases the cost of achieving a given emissions target. To implement carbon pricing most efficiently, the same price should be levied across sectors and fuel types. This would equalise marginal abatement costs, ensuring an optimal reduction of all emissions sources across greenhouse gases and sectors, considering both their effect on climate change and their abatement costs. Yet, countries have implemented a range of different carbon pricing measures (such as cap and trade emissions certificates, explicit carbon taxes and implicit measures like excise taxes), resulting in different rates and prices across emission sources, fuel types and sectors. As a result, carbon prices and excise taxes vary by industry and fuel type. In the 79 countries covered by the OECD Effective Carbon Rates (ECR) database, the road transport sector faced the highest carbon rates (rates above EUR 60 and EUR 120 per tonne of CO₂ mostly occur in that sector), followed by the electricity and off-road transport sectors. In the industry and building sectors, 72% and 64%, respectively, of emissions remained unpriced, while almost three‑quarters of emissions in the electricity sector faced a positive carbon price (OECD, 2024[5]).

Differential rates and free allocations of emissions permits may be primarily motivated by competitiveness considerations (Böhringer et al., 2017[29]), but they can also reflect equity concerns. For example, the fact that low-income households tend to spend greater shares of their budgets on energy or food but also changes in sectoral employment, which will affect workers disproportionately. While food production accounts for a significant share of GHG emissions globally, GHG emissions from agricultural production are priced at much lower rates than for other sectors (for a discussion see Chapter 4). Denmark is currently the only country that has announced a specific tax on agricultural emissions, and across countries, the size and structure of existing agricultural subsidies can favour high-emission activities in the sector.

Source: (OECD, 2024[5]).

5.1.2.  Distributional impacts of climate policies vary by country but are mostly regressive

Modelling framework

Anticipating the net economic impact of climate policies requires tracing their economy‑wide effects to household living standards. A common approach to quantify distributional effects of climate policies is to link macroeconomic models, which can assess the full set of income and price changes in all sectors, with micro-level data analysis, for a large number of households, representative of countries’ entire population.

This section presents an assessment of household impacts of the Enhanced NDCs scenario. The Enhanced NDCs scenario assumes accelerated climate action and increased investments across countries, enabling the world to achieve emission reductions in line with the “well-below 2°C” target of the Paris Agreement (see Chapter 2 for additional information on the scenario). This scenario is compared to a Current Policies scenario, which serves as the baseline and reflects policies already in place or legislated, reflecting the existing ambition and implementation gap in current NDCs. For both scenario, the analysis relies on linking macroeconomic projections from the OECD ENV-Linkages Computable General Equilibrium (CGE) model (Château, Dellink and Lanzi, 2014[30]) with household micro-data, as represented in the mSPIN microsimulation model (O’Donoghue and Sologon, 2023[31]; Sologon et al., 2023[32]) (Annex A). For each global region, projections from ENV-Linkages of changes in consumption and prices for each commodity are fed into mSPIN, together with changes in income, sectoral employment, and wages and capital income. The combined macro and micro lens allow quantifying the impact of climate policy‑induced changes in consumer prices (for energy and other items) and factor prices (returns to labour and capital) on the living standards of households with different incomes and spending patterns.1

The analysis focuses on three OECD countries: France, Poland and Türkiye. These countries represent distinct demographic and economic structures, greenhouse gas emissions, as well as different degrees of economic inequality. This section presents income changes in real terms, i.e. accounting for both income streams and price changes, up until the year 2035.2

A case study of three countries: overview

There are large differences in social characteristics, including in terms of employment, demographics and inequality. Employment rates among 15–64-year-olds are currently much higher in France (69%) and in Poland (73%) than in Türkiye (55%). The population in Türkiye is much younger, with a very low ratio of individuals aged 65+ to working age (currently 11%) compared to France (38%) and Poland (30%). Both Poland and Türkiye are aging faster than France, but by 2035, the share of old-age pensioners in both countries is projected to remain highest in France (OECD, 2024[33]). As a result, a larger share of Turks and Poles are potentially exposed to any changes in employment prospects resulting from climate-change and mitigation measures by 2035. With high shares of informal employment in Türkiye (31% according to the World Bank (Elgin et al., 2021[34])), many workers face lower job security than peers in other OECD countries and accessing social protection is difficult for them. Informality also has implications for government strategies to alleviate household burdens from climate action. For instance, using revenues from carbon pricing to reduce income taxes would have little direct benefit for those not paying labour taxes in the first place. Income inequality and poverty headcounts are below OECD average in France (Gini coefficient 0.3, relative income poverty 9%) and Poland (Gini 0.26, income poverty 9%), while income poverty in Türkiye is above average (15%) and inequality is one of the highest in the OECD (Gini 0.4) (OECD, 2024[33]).

The three countries also differ in terms of production structure, a key driver of the economic outcomes of climate mitigation action (Figure 5.1). France’s economy is service based, with services accounting for over 70% of output in 2024. Poland has undergone a significant structural transformation in recent decades, shifting from an agriculture and heavy-manufacturing based economy towards high-value-added production and services. However, it still retains a strong manufacturing base and a large share of fossil-fuel extraction and power generation. Türkiye is an upper middle-income country but with an output structure that is similar to Poland’s in several ways, with comparable shares in manufacturing and fossil fuel production, although its agriculture sector remains larger.

Figure 5.1. The sectoral structure of the economy varies across countries

Sectoral shares in economy-wide gross output (real) by country, 2022

Note: Aggregation of OECD ENV Linkages sectors is detailed in Annex A. EITE industries refer to emission-intensive and trade exposed industries.

Source: OECD ENV-Linkages model, based on the GTAP 11 database (Aguiar et al., 2023[35]).

Statlink

Emission intensities and the energy mix differ across the three countries. France’s emissions intensity is among the lowest in the world, with a very high share of nuclear power in electricity production (emission factor of ca. 0.05 kg CO₂-eq per kWh according to IEA), and with transport as the highest-emitting sector. Poland’s economy still relies strongly on coal power, with coal’s share in electricity generation currently above 60% (and an emission factor of 0.76 kg CO₂-eq per kWh). Its dependence on fossil fuels extends to the production of subsistence goods, such as food, heating and housing, and contributes to some of the highest urban pollution levels in the EU, implying that climate mitigation would yield large co-benefits of better air quality. Around half of Türkiye's energy mix depends on coal and gas (emission factor for electricity at 0.38 kg CO₂-eq per kWh). However, thanks to a deep penetration of renewables in electricity generation and low motorisation rates, the carbon intensity of Türkiye’s power, transport and agriculture sectors are below the EU average. By contrast, Türkiye’s manufacturing sector is more carbon intensive than the average EU country and high‑emissions industries employ a comparatively large share of the workforce.

Household spending and carbon footprints across income groups

High-income groups consume more than the average household, and they therefore emit more GHGs. Earlier OECD calculations, reported in OECD (2024[5]), show that energy-related GHG emissions from household consumption are highest in France, followed by Poland and, with much lower levels, in Türkiye (Figure 5.2).3 Within countries, overall carbon footprints of the top 20% of the income distribution can be two to three times higher than the poorest 20% of the population. Among the three countries, differences across income groups are largest in Türkiye, followed by Poland and France.

Figure 5.2. Emissions from household consumption are highly unequal across and within countries

Emissions from household consumption, tCO₂ per household at different points in the national income distribution, 2015 for EU countries, 2019 for Türkiye

Note: Estimates follow the “consumer responsibility” principle, accounting for all household consumption, including both domestically produced and imported goods. They therefore attribute emissions to the country where the final good is consumed and this differs from the emissions per capita that are physically released in a given country.

Source: (OECD, 2024[5]), using IEA emissions factors for different fuels, using household budget surveys (2015 for EU countries, 2019 for Türkiye) and World Input-Output Database (WIOD) input-output data (for electricity).

Statlink

In the three countries, low-income households devote large portions of their income to fuel and electricity (Figure 5.3). Energy serves basic needs, such as heating and mobility. Spending shares for energy are therefore higher for those at the lower end of the income spectrum. When comparing shares of income spent on energy, differences between income groups are highest in Türkiye, ranging from less than 5% for the richest decile to almost 25% in the poorest decile. Around two thirds of energy-related expenditures in Türkiye are used for heating, followed by a smaller share dedicated to electricity and finally to motor fuels. In Poland, differences across income groups are more limited, ranging from almost 10% for the richest decile to just under 25% for the poorest decile. About half of energy-related expenditures in Poland are spent on heating, across deciles. While expenditure shares on motor fuels are similar across income groups, spending shares on electricity are higher for low-income households. In France, the variation by income group is less notable, with expenditure shares remaining below 10% of income, except for the poorest decile, where households spend 20% of their disposable income on energy. A comparatively small share of disposable income in France is spent on heating, with larger expenditures on electricity and motor fuels, especially among the poorest households.

Figure 5.3. Poorer households typically spend large shares of their income on energy

Household expenditures on fuel and other energy, as a percentage of disposable income, by income decile, 2015 for EU countries, 2019 for Türkiye

Note: Groups 1-10 refer to income deciles (equivalised disposable household income). Heating fuel includes expenditure on gas (natural gas and town gas), liquified hydrocarbons, kerosene and other liquid fuels, coal and other solid fuels. Motor fuel includes expenditure on diesel and petrol for transportation.

Source: (OECD, 2024[5]) using household budget surveys (2015 for EU countries, 2019 for Türkiye) and World Input-Output Database (WIOD) input-output data (for electricity).

Statlink

Distributional impacts of the Enhanced NDCs scenario

In the Enhanced NDCs scenario, projected changes in electricity prices with respect to the Current Policies scenario differ in the three countries. They are moderate in France, since the electricity system is already largely decarbonised, reflecting limited scope for further cost-savings compared to countries with more carbon-intensive energy systems. In coal-reliant Poland, electricity would become significantly more affordable over time despite higher a carbon price than in the Current Policies scenario, due to a large reduction in the cost of producing renewable energy. The energy transition in Türkiye is projected to be slower, as the Enhanced NDCs scenario requires a greater increase in policy stringency relative to Current Policies compared to France and Poland, which, as one example, are already subject to the EU’s Emission Trading Scheme (ETS). As a result of higher increases in fossil energy production costs, electricity prices are projected to rise in the short term, and only converge back to current levels by 2040. Fossil price increases for transportation are projected to have a large impact in France and Poland too, since the transport electrification and the decrease in production costs for Electric Vehicles (EVs) would remain limited in the short run.

Enhanced NDCs are projected to increase national income, but low-income households gain less than others. Relative to Current Policies, GDP per capita is projected to rise by 0.3% in France and by 0.1% in both Poland and Türkiye by 2035. However, by 2035, changes in real disposable income are expected to be regressive in all three countries (Figure 5.4, Panel A). Gains are biggest at the top of the income distribution, the middle-income households see comparatively small changes, while burdens are largest for lower-income households.4

Figure 5.4. Without complementary policies, the Enhanced NDCs scenario has a small but regressive impact on households

Income changes by group, relative to average income change, Enhanced NDCs scenario versus Current Policies scenario, 2035

Note: Income decile group is based on equivalised disposable income in the Current Policies scenario. Gender and age refer to the household head.

Source: mSPIN microsimulation model using inputs from ENV-Linkages model.

Statlink

The patterns of net gains and losses from Enhanced NDCs are complex, reflecting simultaneous changes in prices and incomes:

• Prices and living costs: simulation results indicate a modest increase in living costs by 2035 in France and Türkiye, at 6% or less, depending on the income group. In Poland, households’ GHG footprints are larger, and the cost of consumption baskets are expected to increase much more strongly, by around 20%. For households in Poland to see real-term gains, nominal incomes would therefore need to increase substantially.

• Employment and wages: in all three countries, wages would increase at a faster pace under Enhanced NDCs than under Current Policies. By 2035, per-capita labour income would be significantly higher in real terms, with Enhanced NDCs, with gains ranging from +1.3% in Türkiye to +1.8% in Poland and 2% in France. The effects of any employment and earnings losses resulting from policy constraints are thus smaller than the job and wage gains from investment, a lower tax wedge financed with additional revenues (notably from carbon pricing) and productivity effects from energy efficiency gains.5

• Capital income: incomes from property and financial capital are a sizeable component of household disposable income, especially in Türkiye (and other middle-income countries), where self-employment, including in the agricultural sector, tends to be more common than in high-income countries. For instance, up to one fourth households in Türkiye report capital incomes as their main source of income, especially among middle-income earners. In France and Poland, this share is smaller. Investments are a key component of projected GDP growth in the Enhanced NDCs scenario and half of climate-related investment is assumed to be sourced from the private sector. Changes in capital income are therefore a key factor shaping the distribution of gains and losses.6

Beyond income levels, other factors impact the distributional effects of climate policies. Policy studies as well as debates on household gains and losses from climate action frequently focus on differences by income group, and on whether overall impacts are regressive or progressive. Beyond income levels, however, many other characteristics drive people’s consumption habits, the way they source their income, and, therefore, the exposure to price or income changes resulting from climate action. These include gender, age, household size or living in a rural area rather than a city. Additional results for the three case‑study countries confirm findings from other studies that have shown gains and losses to vary strongly within income groups.7 As a result, some households may experience substantial losses, even when the overall impact on their income group appears modest or positive. These patterns need to be understood to design policy packages that do not disproportionately harm disadvantaged groups. Although existing studies find that carbon footprints - and thus the potential impacts of pricing instruments – differ significantly across demographic groups (OECD, 2024[5]; Semet, 2025[36]), results from the Enhanced NDCs scenario (Panels B and C of Figure 5.5) do not show simple or systematic patterns by gender and age in the three countries. In France, women would, on average, fare better than men under Enhanced NDCs, but they would do worse than men in Poland and Türkiye (Panel B). Working-age individuals in Türkiye would fare worse than retirees, while older individuals would do worse in France and Poland (Panel C). In practice however, the effects by gender, age and other characteristics can be correlated and simple averages cannot capture details (results for female-headed households can be very different for lone parents and widows, for example). Future work could further analyse policy impacts for a range of household characteristics, while controlling for others.

5.1.3. Support policies can help to make the transition fair and equitable

Compensation policies can offset some of the burdens of climate policies on low-income or disadvantaged households. Carefully tailored support to households can facilitate mitigation approaches, even when they have undesired distributional impacts in the short term. In this context, a key advantage of carbon pricing and of phasing out existing fossil-fuel subsidies, is that they generate public revenue which governments can use to finance such support. Current carbon prices remain well below the levels that could ensure meeting existing climate change mitigation commitments, with modest net carbon revenues estimated at 1% of GDP on average across OECD countries (OECD, 2024[37]). In some OECD countries, revenues are already comparable in size to major categories of social spending, such as income support for the working-age population or social services excluding health.8

As part of broader policy packages, channelling some or all revenues from carbon pricing or other revenue-raising policies (e.g. removal of fossil fuel subsidies) to households provides governments scope for cushioning losses and shaping distributional outcomes. Such compensation can reduce possible regressive impacts of climate action. It can also help address equity concerns when burdens are not regressive, as carbon pricing can pose affordability challenges for some households, even when its impact across households is uniform or progressive. Even a very simple compensation programme, in the form of a uniform lump-sum transfer to everyone, can limit household losses and make parts of the population better off than without carbon pricing measures (OECD, 2024[5]). Figure 5.5 illustrates this for Poland and France, by considering carbon-pricing measures implemented during the period 2012-2021.9

Avoiding unwanted burdens for disadvantaged households requires compensation strategies that are tailored to households’ needs and circumstances. Full revenue redistribution in the form of uniform lump-sum payments helps to mitigate regressive impacts of climate policies (Figure 5.5, Panel A).10 A majority towards the top of the income spectrum lose out, because, for households spending large absolute amounts on fuel and other goods, carbon price burdens are likely to exceed the lump sum. By contrast, most (70% or more) households in the bottom income decile benefit or have no additional burden. These households have low expenditures (in absolute terms), and the flat-rate transfer therefore offsets or exceeds the effect of carbon prices for many of them.11 Even among low-income groups, the lump-sum transfer leaves some losers, highlighting the limitations of simple across-the board compensation schemes. Tailoring compensation strategies, by linking transfer amounts to households’ support needs, helps to ensure that they are sufficient to protect disadvantaged households.

Tailored compensation strategies are also more cost-effective than untargeted ones. Fiscal, equity, efficiency and effectiveness considerations call for carefully tailored compensation strategies. Panel B of Figure 5.5. shows the percentage of households with net losses, when different shares of carbon-pricing revenues are paid out in the form of lump-sum transfers. With no compensation at all, higher carbon prices translate into higher expenditures, making all households worse off. As transfers increase, the share of losers eventually declines. Results indicate that, when compensation takes the form of uniform flat-rate transfers, averting losses for most of the population is expensive In France, a uniform lump-sum transfer sufficient to offset losses for most households would have consumed nearly all carbon pricing revenue.

Household support needs can be substantial but, in practice, less than the full revenue from carbon pricing may be available for financing direct compensation measures. Where compensation consumes all or most of additional revenues, the potential for financing other programmes (such as clean energy deployment subsidies or energy efficiency investments) may be limited. Linking transfer amounts to households’ support needs can reduce the fiscal costs from direct compensation measures, while maintaining the price signals consistent with the objective of lowering emissions. Many countries that have introduced some form of recycling of carbon price revenues have indeed targeted transfers to those most in need (OECD, 2024[5]).

Figure 5.5. To be effective, support payments financed from carbon-pricing revenues should be tailored to household needs

Share of individuals with net income losses, 2012-21 carbon pricing measures with revenue recycling (lump-sum payments)

Note: Excludes Türkiye, as the increase in carbon-price burdens for households was very small during the period. Household compensation takes the form of uniform lump-sum transfers to individuals. Income deciles in Panel A refer to equivalised disposable household income. Individuals occur net losses when the additional cost for their consumption basket induced by carbon pricing exceeds the lump-sum transfer from full/partial revenue recycling.

Source: (OECD, 2024[5]) using OECD Effective Carbon Rates data, IEA emissions factors for different fuels, World Input-Output Database (WIOD) and household budget surveys (2015).

Statlink

The Enhanced NDCs scenario includes revenue recycling,12 but the revenue raised is not sufficient to fully offset regressive effects. The limited effect of revenue recycling (Chapter 2) is due to the specific policy mix used in the scenarios, which relies only partially on carbon pricing and other revenue-generating measures. The effect of revenue recycling is therefore limited, even for countries that do include carbon pricing, such as France and Poland via the EU ETS. This finding highlights the importance of governments incorporating distributive concerns into their policy decisions, particularly regarding household support. In some cases, new revenues, e.g. from carbon pricing, can fully finance such compensation. When this is not the case, e.g. because there are competing demands on carbon pricing revenues, it may be possible to mobilise additional support by deploying existing social protection budgets. Indeed, as for other megatrends and key societal transitions and emergencies, such the COVID-19 pandemic, ageing and digitalisation, there can be a good case for leveraging existing social protection systems to cushion the effects of the transition.

5.2. Managing the impacts of accelerated climate action on labour markets is key to a just transition

5.2.1. Climate action is already changing labour markets

Climate action is already changing the job market, bringing new job opportunities. The OECD Employment Outlook 2024 (2024[38]), on which much of this section is based, has shown that across the OECD, between 2015 and 2019, around 20% of workers were already employed in “green-driven occupations”, meaning not only “green jobs” (Box 5.3) per se but also those jobs that are likely to be in demand because they provide goods and services required by green activities (OECD, 2024[5]).13 Between 2011 and 2022, the share of green-driven occupations in total employment has increased by 2% on average in European-OECD countries and by 15% in the United States (Figure 5.6). This effect was driven by green new and emerging occupations, which increased by 12.9% over the same period. Jobs that are likely to be in demand because they provide goods and services required by green activities also increased but at a significantly slower rate.

In the energy sector, which has undergone the most rapid changes so far, the transition towards net-zero emissions is already having significant consequences on energy-related jobs. Despite the emission reduction targets and regulations in place, the International Energy Agency (IEA) estimates that, more people worked in the energy sector in 2022 than in 2019, almost exclusively due to growth in clean energy, which now employs more workers than fossil fuels (IEA, 2023[39]). Energy employment reached nearly 67 million in 2022, growing by 3.4 million from 2019, with the clean energy sectors adding 4.7 million jobs globally over the same period, mostly thanks to employment in solar PV, wind, electric vehicles (EVs) and battery manufacturing, heat pumps and critical minerals mining (IEA, 2023[39]).

Figure 5.6. Green new and emerging occupations are growing fast

Percentage change in the share of green-driven (green) and GHG-intensive occupations (pink) in total employment

Note: Europe: unweighted average of Austria, Czechia, Estonia, Finland, France, Hungary, Lithuania, Luxembourg, the Netherlands, Norway, Poland, the Slovak Republic, Slovenia, Sweden and Switzerland. Green residual occupations are green-driven occupations excluding those occupations that can be both green-driven and GHG-intensive.

Reading: Between 2011 and 2022, the share of green-driven occupations in total employment has increased by about 1% in 2011-15 and about 1% in 2015-2019 followed by a small decrease in 2019-2022 for a total of about 2% on average in European-OECD countries over the entire period. In the United States, the share of green-driven occupations in total employment has increased by about 3% in 2011-15, 4% in 2015‑2019 and 7% in 2019-2022 for a total of about 15% over the entire period.

Source: Secretariat’s estimates based on version 24.1 of the O*NET database and the following country-specific sources: United States: Current Population Survey; All other countries: EU Labour Force Survey, see Figure 2.4 in OECD (2024[38]).

Statlink

Box 5.3. Measuring “green jobs”

Despite the prominence in the public debate, there is no universally accepted definition1 of green jobs and estimates of the incidence of green jobs in OECD countries vary greatly, depending on the adopted definition – see OECD (2024[38]) for a discussion of measurement approaches. One of the reasons for the lack of consensus on the definition of “green job” is that the concept is used to try to address very different policy issues. On the one hand, the concept of green jobs can be used to measure the contribution of the labour market to the transition, which is a narrow focus. On the other hand, the term green job is also used to identify the type of jobs that are likely to expand because of the transition, hence also including jobs that support green or low-carbon activities but are not necessarily green per se. Only10 OECD countries out of the 35 countries that responded to an OECD Policy questionnaire on the net-zero transition have attempted to define what a green job is (OECD, 2025[40]). In most cases, such exercise had a statistical purpose, i.e. gauge the size of the jobs likely to be positively affected by the transition (e.g. in Austria, Italy or Switzerland). In other cases, the definition reflects broad policy objectives: in Canada, for instance, the definition of a “sustainable job” follows the one proposed by the International Conference of Labour Statisticians to include “any job that is compatible with Canada’s path to a net-zero emissions and climate resilient future” but also that reflects the concept of a “decent job”. Finally, in few cases, the definition is used for operational purposes, such as guiding the work of the PES or targeting hiring incentives: in Slovenia, for instance, employers could receive a subsidy when hiring an unemployed person for a “green job”, defined on the basis of the firm’s activity but also its status (e.g. social enterprise).

Based on the O*NET project, the OECD (2024[38]) distinguishes three groups of occupations of this type:

• Green new and emerging occupations: new occupations (entirely novel or “spinoffs” from an existing occupation) with unique tasks and worker requirements (e.g. Biomass Plant Engineers; Carbon Trading Analysts; Solar Photovoltaic Installers).

• Green-enhanced skills occupations: existing occupations whose tasks, skills, knowledge and external elements, such as credentials, tend to be altered because of the net-zero transition (e.g. Arbitrators, Mediators, and Conciliators; Architects; Automotive Specialty Technicians; Farmers and Ranchers).

• Green increased demand occupations: existing occupations in increased demand due to the net-zero transition but with no significant changes in tasks or worker requirements. Some occupations in this group can be considered as directly contributing to low emissions and clearly involve green tasks (e.g. Environmental Scientists and Specialists; Forest and Conservation Workers) but most are not and should be seen as in support of green economic activities (e.g. Construction Workers; Drivers; Chemists and Materials Scientists).

1. The 19th Conference of Labour Statisticians Guidelines (ICLS) has adopted a definition which differentiates between employment in the environmental sector (employment in environmental processes and employment in the production of environmental outputs) and green jobs, which includes these categories but must also fulfil the criteria for decent work. Conceptual and practical difficulties in its operationalisation have limited the adoption of this definition in the literature.

5.2.2.  Climate action will create new jobs, but also cause job losses and require workforce reallocation

Aggregate employment effects of ambitious climate policies are expected to be limited. The transition will reduce output and therefore jobs in high emitting sectors, which employ a small fraction of the working population, and will create new jobs in more labour-intensive and cleaner sectors, such as the services and clean energy sectors. High-emission industries, responsible for 80% of emissions, employ just 7% of the workforce on average across OECD countries (OECD, 2024[5]). This chapter focuses on the labour market impacts of the Enhanced NDCs scenario (presented in Chapter 2), which shows that accelerated policy action is compatible with continuous economic growth but entails substantial structural changes in the economy (Chapter 4). The analysis of labour market impacts of the Enhanced NDCs scenario presented in this Section is based on the OECD ENV‑Linkages model (Château, Dellink and Lanzi, 2014[30]).

In the Enhanced NDCs scenario, global employment is projected to continue growing in the next fifteen years. Compared to the Current Policies scenario, in the Enhanced NDCs scenario aggregate global employment is projected to be virtually unaffected and is even projected to be slightly higher. This outcome reflects the interplay of several key mechanisms assumed in the scenario (Figure 5.7). More ambitious climate mitigation policies, such as carbon pricing and regulations that facilitate the transition through technology constraints in production and consumption, lead to output contractions in some sectors, resulting in a modest decline in employment (Policy constraints effect). Conversely, the Enhanced NDCs scenario features increased public and private investments in energy and clean technologies (Increased investments effect), which slightly boost aggregate economic activity and therefore employment. Likewise, the scenario reflects the benefits from the improved technologies in terms of energy-efficiency gains (Energy transition effect), which can also slightly boost economic activity and employment. Finally, while half of the generated revenues from carbon pricing, fossil fuel subsidy reforms as well as other revenue‑rising. policies are recycled to compensate households via direct transfers, the other half is used to lower labour income tax rates, increasing labour demand (Revenue recycling effect).14

Figure 5.7. The net impact of Enhanced NDCs on global employment is marginally positive compared to the Current Policies scenario

Employment (billions of persons)

Note: This figure presents projected changes in global aggregate employment under the Enhanced NDCs scenario for the years 2030, 2035 and 2040 decomposed by the main mechanisms connecting climate action to economic outcomes. The bars for Current Policies and Enhanced NDCs show the absolute level of global employment (in billions of persons) in each respective year (with a different y-scale for each year). The intermediate bars show the incremental effect of the key mechanisms that drive the Enhanced NDCs scenario on aggregate output and thus global employment relative to the Current Policies scenario.

Reading: In 2040, the projected effect of policy constraints with Enhanced NDCs alone reduces global employment by 0.13% relative to the Current Policies counterfactual in 2040. At the same time - in isolation - energy efficiency gains, revenue recycling and increased investments raise employment by 0.04%, 0.09% and 0.03%, respectively - resulting in a projected net employment increase of 0.03% in 2040 under Enhanced NDCs compared to Current Policies.

Source: OECD ENV-Linkages model, with inputs from NIESR’s NiGEM and IEA’s GCEM models.

Statlink

Even if the net aggregate economic effects are marginally positive, accelerated climate action will imply changes in sectoral employment. Following structural changes in the economy, jobs will be created in expanding low-emission activities, while others will be lost in shrinking GHG-intensive industries (OECD, 2023[41]; OECD, 2024[5]). Many jobs will also be maintained and transformed as tasks and working methods become greener. The main challenge for labour market policies will therefore be to help workers and companies transition from emission-intensive sectors and occupations to other jobs. According to the modelling analysis, An Enhanced NDCs scenario would imply changes in sectoral employment, with a shift in jobs from high-emitting to lower-emitting sectors, including renewables as well as services (Figure 5.8). Employment in construction will also benefit from increased demand for renovation and green infrastructure.

Figure 5.8. Jobs will be created in green energy sectors, services and construction, but will be suppressed in fossil fuel sectors

Global net employment gains or losses by sector in the Enhanced NDCs scenario compared to the Current Policies scenario, 2040 

Note: The results in this graph also depend on the technology set up of the model, which does not include some technologies, such as coal‑with‑CCS. The development of such technologies, though fossil fuel based, could limit employment losses in sectors relying on coal.

Source: OECD ENV-Linkages model, with inputs from NIESR’s NiGEM and IEA’s GCEM models.

Statlink

5.2.3. Green jobs will bring about additional changes to the labour market

Changes to the labour market will not only affect the quantity but also the quality of jobs. In OECD countries, green-driven occupations, in particular the green new and emerging occupations, tend to be characterised by higher wages and fewer temporary contracts than other occupations (Figure 5.9), although workers in these jobs are often exposed to larger unemployment risk as they are more likely to be employed in innovative startups or activities characterised by high average growth but also high failure rates and shakeouts. However, the OECD Employment Outlook 2024 (2024[38]) shows that, across the three groups of green-driven jobs, the job-quality advantage, relative to other jobs, tends to be concentrated in high-skill occupations, suggesting that those workers who have the specific competences required by these expanding jobs have a competitive edge in the labour market and thrive compared to their peers. Emerging economies face similar if not more serious issues because of labour informality which entails lower wages and lack of access to social protection resulting in low earnings quality, high risk of falling into extreme low pay while employed and a generally lower quality of the working environment (OECD, 2015[42]). However, OECD (2023[43]) finds that, at least in Latin-American countries, relatively higher skill green jobs have an important potential contribution to increasing formalisation, particularly if paired with active labour market policies and with policies to improve female employability.

Figure 5.9. Green new and emerging occupations benefit from higher wages and a better quality of the working environment

Estimated percentage difference in the share of green-driven occupations

Note: This figure reports the point estimate (and 95% confidence intervals) of the percentage difference in the incidence of each type of occupation between high and low-wage employees, unemployed and employed workers and job strained and not-strained workers. High (resp. low) wage is defined as hourly wage above one and-a-half times (resp. below two thirds of) the median wage. Job strain results from insufficient resources in the workplace (e.g. work autonomy, social support at work or learning opportunities) to meet job demands (e.g. work intensity or physical health risk factors).

Reading: Controlling for demographic characteristics, the percentage share of green new and emerging occupations is, on average, 19% higher among high-wage employees than among low-wage employees. A larger incidence of a given type of occupation among high-wage workers than among middle or low-wage workers is indicative of a positive wage gap between that occupation and the others.

Source: Secretariat’s estimates based on version 24.1 of the O*NET database and several national and international data sources, see Figures 2.9, 2.12 and 2.15 in OECD (2024[38]).

Statlink

On average, across OECD countries, the transition will not equally affect employment for men and women. Men are more likely to hold green-driven occupations but also more often employed in occupations concentrated in high-emission sectors (OECD, 2024[5]). This suggests that men are both more exposed to the risk of job loss in high-emission industries, but also better placed to reap the benefits of the transition. Conversely, women’s concentration in the service sector puts them at less risk of transition‑related job losses, but also raises concerns about their ability to benefit from the job opportunities that will open. In particular, the current under-representation of women in science, technology, engineering and mathematics (STEM) education and the persistence of gender stereotypes raise concerns about women's ability to benefit from the growing employment opportunities in the highest paying expanding sectors (OECD, 2024[5]).

The transition comes at a time when labour markets around the world are experiencing other changes, such as technological advances, especially around generative artificial intelligence, the reorganisation of global value chains and rapid demographic ageing (OECD, 2024[5]). Ambitious NDCs should therefore place measures to address the social impact of the transition at the core of climate strategies to ensure social acceptability and minimise the social costs of more ambitious NDCs. A coherent set of labour market policies to prevent mass redundancies, accompany workers and steer them towards the growing sectors of the economy is key to help workers navigate within and across affected sectors and regions.

5.2.4. Policies can help manage structural adjustments in labour markets

The transition to new jobs can be facilitated with dedicated policies. A mix of reskilling and financial incentives to support the livelihoods of displaced workers using unemployment insurance and well-targeted wage insurance schemes can steer them towards the expanding sectors of the economy, with effective active labour market policies and geographical mobility policies. However, retraining needs can be high and particularly needed in sectors, locations and communities that are hit hardest by the transition, such as in fossil fuel extraction. While most high-skill jobs in emission-intensive industries share similar skill requirements with occupations in other industries, this is less the case for low-skill jobs who will require more intensive and targeted retraining (Box 5.4). Recent estimates based on data from selected OECD countries15 shows that it is on average 24% more costly to lose a job in a high-emission sector than in other sectors, because the majority of workers in these sectors tend to be older, less educated and tend to have wage premia that are difficult to match in new jobs they can get access to (Figure 5.10. ) (OECD, 2024[5]). However, a recent assessment from Canada, shows that not all workers displaced from fossil fuel industries experience similar earnings trajectories after job loss: five years after job loss, 1 in 4 workers displaced from coal mining from 2004 to 2011 saw their annual wages fall by at least CAD 19,000; while, 1 in 4 workers saw their annual wages rise by at least CAD 31,000 (Chen and Morissette, 2020[44]). A challenge is that new jobs brought about by the green transition may not be in the same local market as the high‑emission jobs that are lost.16

Figure 5.10. Workers in high emission industries face large and persistent job displacement costs

Difference in annual earnings between displaced workers and their matched counterparts relative to the time of displacement, average across countries, %

Note: The figure plots the average coefficients and the corresponding 90% confidence intervals across countries based on an event-study regression in in OECD (2024[38]). The coefficients capture the earnings losses of displaced workers relative to observationally comparable non-displaced workers. The point estimates show the impact of job loss on earnings in event time, where workers are displaced between time 0 and time 1, such that time 1 is the first post-displacement year. Related to this, earnings losses present a drop by construction at time 0, as earnings capture the sum of labour payments over the entire year and consequently already capture part of the displacement effect at time 0. The reference period for earnings losses is k=3. Point estimates and confidence intervals from country-level regressions are averaged with equal weights. The countries included are Australia, Austria, Canada, Denmark, Estonia, Finland, Germany, Hungary, the Netherlands, Norway, Portugal, Spain, France and Sweden.

Reading: Workers in in high emission industries face greater earnings losses after job displacement, averaging a 36% decrease over 5-6 years after job loss compared to 29% in other sectors.

Source: National linked employer employee data, see Figure 3.6 in OECD (2024[38]).

Statlink

Most OECD countries already have a range of instruments for managing structural adjustments. In some countries trade unions and employers’ organisations have a long-standing experience in anticipating structural change. However, some instruments may need to be adapted – e.g. in terms of targeting and resources and new ones may be needed, for example to make low-skilled green jobs more attractive, or to address the strong regional disparities in the effects of the transition. The extent to which expansion or innovation of existing policies, or new instruments, will be required will depend on the country context and the specific design of climate change policies. For example, a comprehensive exercise to inform public policy under different transition scenarios is the Australian Clean Energy Workforce Assessment, completed in late 2023 (OECD, 2025[40]).

Policies focusing on training can help workers in high-emission sectors to make the transition to other jobs. Effective and targeted training programmes are needed to provide workers, especially the low‑skilled, with reskilling opportunities to help them move out of high-emitting sectors (OECD, 2024[45]), and to acquire a set of skills that matches the needs of the green transition (Box 5.4). Early interventions focused on workers at high risk of redundancy, such as those undertaken by the Job Security Councils in Sweden, are particularly effective, as they reduce the length of time workers are out of work.

Box 5.4. Skills for Enhanced NDCs

The OECD (2024[38]) shows that the most required skills for green-driven occupations are those linked to the knowledge economy, such as critical thinking, monitoring, active learning, complex problem solving and decision making. Furthermore, newer jobs emerging because of the transition demand higher proficiency across all skills compared to established green-driven occupations.

Comparing the skill requirements of green-driven occupations with those of emission-intensive and environmentally neutral occupations, Figure 5.11 shows that new green-driven occupations with low education and experience requirements (“Low-skill”) generally require higher skill levels than jobs in GHG‑intensive and neutral occupations with similar education and experience requirements. Conversely, the skill requirements of green-driven and GHG-intensive occupations with high education and experience requirements (“High-skill”) are very similar.

Figure 5.11. Transitions out of emission-intensive occupations are possible with the necessary retraining

Skill requirements by type of occupation

Note: The figure shows the level at which a group of skills is needed to perform the occupation. For an easier interpretation, means have been standardised to a scale ranging from 0 to 100, where greater values implies that a given skill category is required at higher levels For comprehensive details on how occupations were classified as Low-skill (requiring little to moderate preparation) or High-skill (requiring considerable preparation), see OECD (2024[38]).

Reading: Within Low-skill occupations, new green-driven occupations – such as solar photovoltaic installers or geothermal technicians – require much higher levels of system skills. Within High-skill occupations, in the extreme case of process skills (e.g. critical thinking, active learning), there is no gap in skills requirements between green driven and emission-intensive occupations.

Source: OECD elaboration based on O*NET data, see Figure 4.2 in OECD (2024[38]).

Statlink

While OECD countries recognise the need for reskilling in the green transition, workers in emission-intensive sectors receive less training and information on changing skill demands is often unclear. To address this, the OECD (2025[40]) shows that eight OECD countries offer employer incentives for green training, while others provide direct financial support to workers. Nineteen countries fund training providers to develop green skills programs, but broader early interventions are needed to prevent job displacement. In addition, some countries offer outplacement services to assist workers in transitioning to green jobs. For example, Germany’s “transfer companies” provide short-term employment, training and job placement for those laid off from carbon-intensive industries. Similarly, Poland supports SMEs and jobseekers with training and financial aid. Finally, 11 OECD countries use job retention schemes to support businesses transitioning to net-zero. Spain’s RED mechanism, for instance, allows temporary work reductions while requiring retraining.

An early assessment of labour market consequences will be important to accompany workers who might face most difficulties and help companies get the workers they need. The consequences of changes in the labour markets vary by sector and by profession and inevitably some professions will be negatively affected. About one-third of the OECD countries that have responded to the OECD Policy questionnaire on the net-zero transition (OECD, 2025[40]) have undertaken some form of assessment of the labour market implications of climate policies, although few have looked closely at the readiness and adequacy of existing policies and programmes. As part of these stress tests, consultations with key stakeholders, starting with the social partners, will provide essential input for the design of the transition and the measures needed to support workers and their families. Clear and well‑designed NDC targets will help better anticipate labour market consequences, encourage the development of skills and competences needed for the transition (see Box 5.5 on the specific labour market bottlenecks that may hinder the delivery of more ambitious climate targets).

Box 5.5. Addressing labour shortages to deliver more ambitious climate targets

To be able to deliver more ambitious climate targets, policies should help sectors that are expanding to find the workers they need. In some cases, workers from high-emission sectors will easily find new green-driven jobs, i.e. green jobs as well as jobs that do not directly contribute to emission reductions but are in demand because they provide goods and services required by green activities. Box 5.4 shows that the skill requirements of green-driven and GHG-intensive occupations with high education are very similar. However, this is not the case for low- and medium-skilled workers, for whom the skills gap is larger and who therefore require significantly more retraining to move into green-driven jobs. In addition, the transition will increase the overall demand for skilled workers in the labour market, in particular in science, technology, engineering and mathematics (STEM), Alexander et al. (2024[46]) show that economies with a robust supply of STEM workers, in particular those that have managed to better include women, are better placed in the transition because they generate more green innovation and face lower bottlenecks in expanding the green workforce.

However, skills are not the only bottleneck. Labour shortages in expanding green sectors may also be related to the lower attractiveness of some green-driven jobs compared to other jobs. The findings discussed above concerning job quality suggest that without policy measures, low-skill green-driven occupations may be an unattractive option for low-skilled workers, even if they require little or no training. Moreover, it is to be noted that while, on average, middle- and high-skill green-driven occupations come with a wage premium, STEM graduates may find it more attractive to work in other sectors (e.g. finance or tech) that pay higher wages (Popp et al., 2022[47]).

Finally, while both green-driven and emission-intensive occupations tend to be concentrated in rural areas, they are not necessarily in the same regions. The geographical mismatch between declining and expanding sectors is a third bottleneck to consider. Geographical mobility policies may be needed to support people interested in moving to other areas, in addition to targeted green subsidies and investments in the regions that are most vulnerable to the transition (OECD, 2023[41]).

Even with effective training and early intervention, some displaced workers may struggle to find suitable jobs immediately. Income support schemes can cushion the financial impact on workers who have lost their jobs, while strengthening incentives to move into new jobs. Unemployment insurance (UI) remains the primary tool for income support, but wage insurance schemes – which compensate workers for wage losses when transitioning to lower-paying jobs – could help accelerate reemployment. Currently, only 7 out of 35 OECD countries use such schemes, mainly as part of trade adjustment programs (OECD, 2025[40]). Wage insurance could be particularly beneficial for low-skill workers, who often earn less in green jobs. However, concerns exist about fairness and potential lock-in effects in low-wage jobs. Still, evidence from the United States (Hyman, Kovak and Leive, 2024[48]) suggests that well‑targeted, temporary wage insurance can be cost-effective.

Displaced workers, as well as those at risk of displacement and job loss, will need support in the job search process and steering them towards opportunities in expanding segments of the economy. Job search assistance has proven effective in increasing re-employment rates and stability, with countries like Spain, Colombia and Slovakia using it to direct vulnerable workers towards in-demand green jobs (OECD, 2025[49]). To be effective, public employment services (PES) must be well-staffed and continuously upskilled to understand shifting labour market needs. Some countries, like Australia and France, develop regional roadmaps to align skills training with emerging job opportunities. Additionally, employment incentives, such as Australia’s New Energy Apprentice Support Payment, help encourage workers to enter clean energy careers. In regions heavily affected by declining emission-intensive industries, job opportunities may be limited in the short term, necessitating geographical mobility policies. Yet only 7 out of 35 OECD countries have relocation support programs and none are directly tied to the green transition. To prevent regional disparities, place-based policies, such as green subsidies and targeted investments, may be needed (OECD, 2023[50]; OECD, 2025[40]). Another positive example is that of the transition out of coal in Germany (Box 5.6).

Robust stakeholder engagement is essential for effective management of the transition. While some OECD countries include trade unions, employers or other partners in consultations, only a few closely involve the social partners in managing the transition. A successful example of stakeholder involvement is Sweden's Job Security Councils, which are established through collective agreements between employers and workers in different sectors. They are an example of preventive measures that generally lead to rapid re-employment of most displaced workers. In addition, past OECD evidence suggests that collective bargaining and social dialogue can have a positive impact on working conditions, yet workers in low‑emission activities are less well represented in collective bargaining (Zwysen, 2024[51]). Initiatives to promote collective bargaining and social dialogue in these industries and companies would therefore have an important role to play and could improve their attractiveness for low- and medium-skilled workers.

Box 5.6. The transition out of coal in Germany

Long a cornerstone of European heavy industry, the Ruhr region in Germany has undergone a profound economic transformation which can serve as an inspiration for the transition to net zero. Once dominated by coal, steel and chemical production, the region saw a sharp decline in coal-related employment from over 600,000 workers in 1957 to fewer than 6,000 in 2016 due to automation, productivity gains and competition from alternative energy sources.

Substantial investments in education and innovation-friendly policies with the establishment of 22 universities and over 250,000 students by 2014 were key to ensure a successful transition. In addition, tripartite agreements between coal companies, trade unions and different levels of government facilitated measures such as retraining programmes and relocation of workers, thereby mitigating the social costs of deindustrialisation. By 2009, the region's growing renewable energy sector employed 24,000 people and generated EUR 7 billion in revenue.

Urban regeneration initiatives, investment in infrastructure, cultural development and environmental innovation helped stem out-migration and transformed the Ruhr into a leader in green industry technology. Today, the region employs more people in environmental industries than in coal and steel, positioning it at the forefront of the transition.

Source: (OECD, 2023[41])

Emerging economies could face stronger challenges. While strengthened quality training and apprenticeship programmes, as well as carefully designed active labour market programmes are also relevant issues for most emerging economies, specific issues require careful consideration. Social protection systems (e.g. unemployment compensation, social assistance programmes, such as cash transfers, and health care benefits) tend to be less developed than in many OECD countries. Moreover, in the context of a large informal economy and often weak enforcement, another important area for policy action is improving the effectiveness of labour laws in protecting workers (e.g. working-time regulations, health and safety legislation, employment protection legislation). Reducing informality is therefore a priority also to provide effective protection for workers affected by more ambitious climate action.

Overall, countries around the world can withstand the labour market impacts of ambitious NDCs, even when accounting for the necessary job reallocations. To do so, policies to address the social impacts of the transition should be at the core of climate strategies in both OECD and emerging economies. The key difference between the challenges posed by technological advances or rapid demographic ageing and those posed by climate action is the political nature of climate targets, which are set and implemented by government action. The measures discussed in this section together with the guidelines for a just transition adopted at the International Labour Organization in 2015 (ILO, 2015[52]) outline a coherent set of labour market policies to help workers meet the challenges associated with ambitious NDCs (Table 5.1).

Table 5.1. Labour market policies for a fair climate transition

Objective	Instruments
Stress-testing existing policies	Assessment of the labour market implications of the transition Skills assessment and anticipation (SAA) Consultations with key stakeholders
Helping workers in high-emission sectors to make the transition to other jobs	Reskilling Financial support/incentives and career guidance Outplacement services Job retention schemes
Supporting the livelihoods of displaced workers	Unemployment insurance Wage insurance schemes Strengthen the social protection system (in emerging economies)
Steering towards the expanding sectors of the economy	Job search assistance (need for adequate PES staff) Upskilling and reskilling programmes Career guidance Geographical mobility policies & place-based policies Increase the benefits of formality, decrease the costs of formalisation and improve enforcement methods (emerging economies)

5.3. Accelerated climate action and associated reduced climate damages have a net positive impact on people in the long run

The impacts of climate change and of climate change mitigation policies on welfare unfold over different timelines. Reduced climate damages imply large benefits from policy action in the long term. However, climate policies can create trade-offs in the short-term, competing with other priorities like public spending on education and employment, highlighting the need to balance climate action with inclusive social outcomes (Baumol and Oates, 1988[53]; Baranzini, Goldemberg and Speck, 2000[54]; Vona, 2021[55]). The difference in timing between the costs and benefits of the transition also influences distributional effects. Lower-income groups tend to be more exposed and vulnerable to climate change and would thus benefit significantly from policy action in the long term. While high-income groups usually have more means to shield themselves from the effects of rising temperatures, lower-income households are more exposed and lack the resources to adapt (Hodok and Kozluk, 2024[56]). For Latin America and the Caribbean, a recent study by the Inter-American Development Bank (IBRD) suggests that more than 78 million poor people live in areas that are highly exposed to climate related‑shocks (Inter-American Development Bank, 2023[57]). In the medium to long term, large sections of the global and national populations will be significantly better off with effective climate change mitigation that averts rapid-onset disasters (floods, hurricanes, wildfires) and slow-onset events (desertification, heat waves, rising sea levels, etc.) attributable to climate change. Low-income households, who are usually more exposed to climate change risks, will benefit from reduced climate damages. Similarly, as they are also more exposed to other environmental risks, they are more likely to experience climate co-benefits, such as from improved air quality. While high‑income households might be better able to adapt to climate damages, economic losses can be consistent for high-income groups, when taken in aggregate.

Accelerated climate action will help alleviate harmful effects of climate change on labour productivity, benefiting workers and the economy. Climate change has led to increasing frequency, intensity and duration of heatwaves (Domeisen et al., 2022[58]), and they now also occur in places where they were previously untypical, such as the 2021 heatwave in the Pacific Northwest of North America (White et al., 2023[59]). Rising temperatures and more frequent heatwaves will have a direct impact on the labour market, as heat exposure is a well-recognised and documented occupational health and safety risk.17 As a result, heat stress has been shown to lower productivity (Day et al., 2019[60]; ILO, 2019[61]; Burke et al., 2023[62]; Cachon, Gallino and Olivares, 2012[63]; Zhang et al., 2018[64]), increase absenteeism (Somanathan et al., 2021[65]), heighten the risk of work-related accidents, including fatal ones (Fatima et al., 2021[66]; Froom et al., 1993[67]; Park, Pankratz and Behrer, 2021[68]) as well as affect the functioning of machinery and infrastructure (Benhamou and Flamand, 2023[69]), increasing risks related to self-heating materials and chemical substance. Without ambitious policies, lower labour productivity caused by the increasing incidence of heatwaves can result in high economic costs (Costa et al., 2024[70]) (Box 5.7).

Accounting for the costs of delayed policy action would provide additional arguments in support of accelerated policy action. Accelerated climate action undertaken in an orderly manner ensures that necessary policy and technological changes are less disruptive and less costly, especially for vulnerable people. As the time available for bridging gaps between current and required climate change abatement efforts becomes shorter, the prospect of drastic and fast-paced policy changes increases. Drastic and fast‑paced policy shifts pose increasing risks, including sudden adjustment burdens for households, trade‑offs between carbon pricing and living standards, and rising pressures on public finances (Pisani-Ferry, 2021[71]). The social and labour market costs of ambitious action should also be compared with the costs of delayed action. Through this approach, what may seem like a trade-off between tackling climate change and protecting jobs and livelihoods can be resolved, aligning climate action with economic stability and job protection.

Box 5.7. The economic costs of climate-related labour productivity due to increasing temperatures

Recent cross-country firm-level evidence (Costa et al., 2024[70]) quantifies the effect of heat stress on labour productivity, covering both gradual temperature increases and extreme weather events. The analysis builds on a unique dataset gathering detailed weather and financial information for more than 2.7 million manufacturing and services firms across 23 advanced economies between 2000 and 2021.

The analysis shows that both more frequent high-temperature days and the occurrence of heatwaves lead to reduced labour productivity (Figure 5.12). Ten extra days above a temperature of 35°C in a year result in a 0.3% reduction in firms’ annual labour productivity. This effect is comparable to the decrease in productivity from a 5% rise in energy prices (André et al., 2023[72]). One additional heat wave lasting at least five days causes a 0.2% reduction in firms’ annual labour productivity.

Figure 5.12. High temperatures and heatwaves reduce labour productivity in OECD countries

Note: Bars represent estimated coefficients and vertical lines the respective confidence intervals. In Panel A each estimation differs with respect to the definition of the temperature variable (number of days above 30ºC, or above 35ºC or above 40ºC). In Panel B, each estimation differs with respect to the definition of heat wave, varying both the temperature threshold above which temperature must rise for a heat wave to have occurred and the minimum number of consecutive days this temperature needs to have occurred.

Source: Costa et al. (2024[70]).

Statlink

The negative productivity effects of rising temperatures are more pronounced in smaller, less productive firms and are exacerbated by prolonged heat waves, high humidity and low wind speeds. Larger firms tend to be more resilient, benefiting from greater financial resources, advanced technology and adaptation strategies. Some adaptation is already occurring, as firms in warmer regions and those with prior heatwave exposure suffer smaller productivity losses. National Adaptation Plans and firm-level investments help mitigate the effects of heat stress. However, adaptation remains limited – higher temperatures relative to an already warm average result in more significant productivity losses and there is no evidence of adaptation to severe extremes.

Source: Costa et al. (2024[70])

References

[35] Aguiar, A. et al. (2023), “The Global Trade Analysis Project (GTAP) Data Base: Version 11”, Journal of Global Economic Analysis, Vol. 2/7, https://doi.org/10.21642/jgea.040101af.

[46] Alexander, N. et al. (2024), “Green Jobs and the Future of Work for Women and Men”, IMF Staff Discussion Notes, No. 2024/003,, International Monetary Fund, Washington, D.C.

[72] André, C. et al. (2023), “Rising energy prices and productivity: short-run pain, long-term gain?”, OECD Economics Department Working Papers, No. 1755, OECD Publishing, Paris, https://doi.org/10.1787/2ce493f0-en.

[78] Araar, A., Y. Dissou and J. Duclos (2011), “Household incidence of pollution control policies: A robust welfare analysis using general equilibrium effects”, Journal of Environmental Economics and Management, Vol. 61/2, pp. 227-243, https://doi.org/10.1016/j.jeem.2010.12.002.

[54] Baranzini, A., J. Goldemberg and S. Speck (2000), “A future for carbon taxes”, Ecological Economics, Vol. 32/3, pp. 395-412, https://doi.org/10.1016/s0921-8009(99)00122-6.

[53] Baumol, W. and W. Oates (1988), The theory of environmental policy, Cambridge University Press.

[69] Benhamou, S. and J. Flamand (2023), Le travail à l’épreuve du changement climatique, France Strategie, Paris.

[29] Böhringer, C. et al. (2017), “The efficiency cost of protective measures in climate policy”, Energy Policy, Vol. 104, pp. 446-454, https://doi.org/10.1016/j.enpol.2017.01.007.

[17] Borenstein, S. and L. Davis (2016), “The Distributional Effects of US Clean Energy Tax Credits”, Tax Policy and the Economy, Vol. 30/1, pp. 191-234, https://doi.org/10.1086/685597.

[20] Braungardt, S. et al. (2023), “Banning boilers: An analysis of existing regulations to phase out fossil fuel heating in the EU”, Renewable and Sustainable Energy Reviews, Vol. 183, p. 113442, https://doi.org/10.1016/j.rser.2023.113442.

[22] Brown, M. et al. (2023), Tax Credits for Clean Electricity: The Distributional Impacts of Supply-Push Policies in the Power Sector, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w31621.

[26] Bruegge, C., T. Deryugina and E. Myers (2019), “The Distributional Effects of Building Energy Codes”, Journal of the Association of Environmental and Resource Economists, Vol. 6/S1, pp. S95-S127, https://doi.org/10.1086/701189.

[62] Burke, M. et al. (2023), Game, Sweat, Match: Temperature and Elite Worker Productivity, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w31650.

[63] Cachon, G., S. Gallino and M. Olivares (2012), “Severe Weather and Automobile Assembly Productivity”, SSRN Electronic Journal, https://doi.org/10.2139/ssrn.2099798.

[30] Château, J., R. Dellink and E. Lanzi (2014), “An Overview of the OECD ENV-Linkages Model: Version 3”, OECD Environment Working Papers, No. 65, OECD Publishing, Paris, https://doi.org/10.1787/5jz2qck2b2vd-en.

[44] Chen, W. and R. Morissette (2020), How Do Workers Displaced from Energy-producing Sectors Fare after Job Loss? Evidence from the Oil and Gas Industry, https://www150.statcan.gc.ca/n1/pub/11-626-x/11-626-x2020021-eng.htm.

[70] Costa, H. et al. (2024), “The heat is on: Heat stress, productivity and adaptation among firms”, OECD Economics Department Working Papers, No. 1828, OECD Publishing, Paris, https://doi.org/10.1787/19d94638-en.

[79] Costantini, V. et al. (2025), “The welfare impact of climate action: A distributional analysis for Italy”, Energy Economics, Vol. 143, p. 108181, https://doi.org/10.1016/j.eneco.2025.108181.

[18] Davis, L. (2023), The Economic Determinants of Heat Pump Adoption, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w31344.

[23] Davis, L. and C. Knittel (2019), “Are Fuel Economy Standards Regressive?”, Journal of the Association of Environmental and Resource Economists, Vol. 6/S1, pp. S37-S63, https://doi.org/10.1086/701187.

[60] Day, E. et al. (2019), “Upholding labour productivity under climate change: an assessment of adaptation options”, Climate Policy, Vol. 19/3, pp. 367–385, https://doi.org/10.1080/14693062.2018.1517640.

[2] Dechezleprêtre, A. et al. (2025), Fighting climate change: International attitudes toward climate policies, American Economic Review, https://doi.org/10.1257/aer.20230501.

[58] Domeisen, D. et al. (2022), “Prediction and projection of heatwaves”, Nature Reviews Earth & Environment, Vol. 4/1, pp. 36-50, https://doi.org/10.1038/s43017-022-00371-z.

[11] Douenne, T. and A. Fabre (2022), “Yellow Vests, Pessimistic Beliefs, and Carbon Tax Aversion”, American Economic Journal: Economic Policy, Vol. 14/1, pp. 81-110, https://doi.org/10.1257/pol.20200092.

[34] Elgin, C. et al. (2021), “Understanding Informality”, CERP Discussion Paper, Vol. 16497, https://cepr.org/publications/dp16497.

[66] Fatima, S. et al. (2021), “Extreme heat and occupational injuries in different climate zones: A systematic review and meta-analysis of epidemiological evidence”, Environment International, Vol. 148, p. 106384, https://doi.org/10.1016/j.envint.2021.106384.

[67] Froom, P. et al. (1993), “Heat Stress and Helicopter Pilot Errors”, Journal of Occupational and Environmental Medicine, Vol. 35/7, pp. 720-724, https://doi.org/10.1097/00043764-199307000-00016.

[8] Fullerton, D., G. Heutel and G. Metcalf (2011), Does the Indexing of Government Transfers Make Carbon Pricing Progressive?, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w16768.

[19] Giraudet, L., C. Bourgeois and P. Quirion (2021), “Policies for low-carbon and affordable home heating: A French outlook”, Energy Policy, Vol. 151, p. 112140, https://doi.org/10.1016/j.enpol.2021.112140.

[10] Giraudet, L., C. Guivarch and P. Quirion (2011), “Comparing and Combining Energy Saving Policies: Will Proposed Residential Sector Policies Meet French Official Targets?”, The Energy Journal, Vol. 32/1_suppl, pp. 213-242, https://doi.org/10.5547/issn0195-6574-ej-vol32-si1-12.

[9] Goulder, L. et al. (2019), “Impacts of a carbon tax across US household income groups: What are the equity-efficiency trade-offs?”, Journal of Public Economics, Vol. 175, pp. 44-64, https://doi.org/10.1016/j.jpubeco.2019.04.002.

[75] Hamilton, K. and G. Cameron (1994), “Simulating the Distributional Effects of a Canadian Carbon Tax”, Canadian Public Policy / Analyse de Politiques, Vol. 20/4, p. 385, https://doi.org/10.2307/3551997.

[56] Hodok, J. and T. Kozluk (2024), “Distributional impacts of energy transition pathways and climate change”, OECD Economics Department Working Papers, No. 1820, OECD Publishing, Paris, https://doi.org/10.1787/56e93f3e-en.

[48] Hyman, B., B. Kovak and A. Leive (2024), Wage Insurance for Displaced Workers, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w32464.

[39] IEA (2023), World Energy Employment 2023, https://www.iea.org/reports/world-energy-employment-2023.

[61] ILO (2019), Working on a warmer planet. The impact of heat stress on labour productivity and decent work., International Labour Office, Geneva.

[52] ILO (2015), Guidelines for a just transition towards environmentally sustainable economies and societies for all, International Labour Organization.

[57] Inter-American Development Bank (2023), “Social protection and climate change: How can we protect the most vulnerable households against new climate threats?”, Policy Brief, Vol. IDB-PB-00375.

[24] Jacobsen, M. (2013), “Evaluating US fuel economy standards in a model with producer and household heterogeneity”, American Economic Journal: Economic Policy, Vol. 5/2, pp. 148-187, https://doi.org/10.1257/pol.5.2.148.

[77] Labandeira, X., J. Labeaga and M. Rodríguez (2009), “An integrated economic and distributional analysis of energy policies”, Energy Policy, Vol. 37/12, pp. 5776-5786, https://doi.org/10.1016/j.enpol.2009.08.041.

[3] Lamb, W. et al. (2020), “What are the social outcomes of climate policies? A systematic map and review of the ex-post literature”, Environmental Research Letters, Vol. 15/11, p. 113006, https://doi.org/10.1088/1748-9326/abc11f.

[13] Lekavičius, V. et al. (2020), “Distributional impacts of investment subsidies for residential energy technologies”, Renewable and Sustainable Energy Reviews, Vol. 130, p. 109961, https://doi.org/10.1016/j.rser.2020.109961.

[16] Levinson, A. (2019), “Energy Efficiency Standards Are More Regressive Than Energy Taxes: Theory and Evidence”, Journal of the Association of Environmental and Resource Economists, Vol. 6/S1, pp. S7-S36, https://doi.org/10.1086/701186.

[12] Lihtmaa, L., D. Hess and K. Leetmaa (2018), “Intersection of the global climate agenda with regional development: Unequal distribution of energy efficiency-based renovation subsidies for apartment buildings”, Energy Policy, Vol. 119, pp. 327-338, https://doi.org/10.1016/j.enpol.2018.04.013.

[4] Markkanen, S. and A. Anger-Kraavi (2019), “Social impacts of climate change mitigation policies and their implications for inequality”, Climate Policy, Vol. 19/7, pp. 827-844, https://doi.org/10.1080/14693062.2019.1596873.

[76] Metcalf, G. et al. (2008), Analysis of U.S. Greenhouse Gas Tax Proposals, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w13980.

[31] O’Donoghue, C. and D. Sologon (2023), The Transformation of Public Policy Analysis in Times of Crisis – A Microsimulation-Nowcasting Method Using Big Data.

[40] OECD (2025), A Fair Net-Zero Transition: Labour Market Policies to Meet Climate Targets, OECD Publishing, Paris, https://doi.org/10.1787/98a67eef-en.

[82] OECD (2025), Social Expenditure Database (SOCX), https://www.oecd.org/en/data/datasets/social-expenditure-database-socx.html.

[49] OECD (2025), The pivotal role of active labour market policies and public employment services in the green transition, OECD Publishing, Paris, forthcoming.

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

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

[33] OECD (2024), Society at a Glance 2024: OECD Social Indicators, OECD Publishing, Paris, https://doi.org/10.1787/918d8db3-en.

[45] OECD (2024), Training Supply for the Green and AI Transitions: Equipping Workers with the Right Skills, Getting Skills Right, OECD Publishing, Paris, https://doi.org/10.1787/7600d16d-en.

[5] OECD (2024), “Who pays for higher carbon prices? Mitigating climate change and adverse distributional effects”, in OECD Employment Outlook 2024: The Net-Zero Transition and the Labour Market, OECD Publishing, Paris, https://doi.org/10.1787/9138d7e3-en.

[50] OECD (2023), A Territorial Approach to Climate Action and Resilience, OECD Regional Development Studies, OECD Publishing, Paris, https://doi.org/10.1787/1ec42b0a-en.

[1] OECD (2023), How Green is Household Behaviour?: Sustainable Choices in a Time of Interlocking Crises, OECD Studies on Environmental Policy and Household Behaviour, OECD Publishing, Paris, https://doi.org/10.1787/2bbbb663-en.

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

[7] OECD (2022), “Coping with the cost-of-living crisis: Income support for working-age individuals and their families”, OECD, Paris, https://www.oecd.org/social/Income-support-for-working-age-individuals-and-their-families.pdf.

[42] OECD (2015), OECD Employment Outlook 2015, OECD Publishing, Paris, https://doi.org/10.1787/empl_outlook-2015-en.

[43] OECD et al. (2023), Latin American Economic Outlook 2023: Investing in Sustainable Development, OECD Publishing, Paris, https://doi.org/10.1787/8c93ff6e-en.

[68] Park, R., N. Pankratz and A. Behrer (2021), “Temperature, Workplace Safety, and Labor Market Inequality”, DP, No. 14560, IZA, Bonn.

[28] Peñasco, C., L. Anadón and E. Verdolini (2021), “Systematic review of the outcomes and trade-offs of ten types of decarbonization policy instruments”, Nature Climate Change, Vol. 11/3, pp. 257-265, https://doi.org/10.1038/s41558-020-00971-x.

[73] Pereira da Silva, L., F. Bourguignon and M. Bussolo (eds.) (2008), The Impact of MacroEconomic Policies on Poverty and Income Distribution, The World Bank, https://doi.org/10.1596/978-0-8213-5778-1.

[71] Pisani-Ferry, J. (2021), “Climate policy is macroeconomic policy, and the implications will be significant”, PIIE Policy Brief 21-20, https://www.piie.com/publications/policy-briefs/climate-policy-macroeconomic-policy-and-implications-will-be-significant.

[47] Popp, D. et al. (2022), The Next Wave of Energy Innovation: Which Technologies? Which Skills?, National Bureau of Economic Research, Cambridge, MA, https://doi.org/10.3386/w30343.

[6] Rausch, S., G. Metcalf and J. Reilly (2011), “Distributional impacts of carbon pricing: A general equilibrium approach with micro-data for households”, Energy Economics, Vol. 33, pp. S20-S33, https://doi.org/10.1016/j.eneco.2011.07.023.

[27] Robinson, H. (1985), “Who pays for industrial pollution abatement?”, The Review of Economics and Statistics, pp. 702-706.

[81] Saez, E. and G. Zucman (2016), “Wealth Inequality in the United States since 1913: Evidence from Capitalized Income Tax Data *”, The Quarterly Journal of Economics, Vol. 131/2, pp. 519-578, https://doi.org/10.1093/qje/qjw004.

[36] Semet, R. (2025), “Unravelling the Influence of Household Characteristics and Decisions on their Carbon Footprint: A Quantile Regression Analysis”, Economie et Statistique / Economics and Statistics 545, pp. 27-46, https://doi.org/10.24187/ecostat.2024.545.2127.

[32] Sologon, D. et al. (2023), Spatial Economics of Income Distribution Across Borders: Drivers of Spatial Inequalities using Microsimulation (SPIN-CORE, FNR), https://liser.elsevierpure.com/en/projects/spatial-economics-of-income-distribution-across-borders-drivers-o.

[65] Somanathan, E. et al. (2021), “The Impact of Temperature on Productivity and Labor Supply: Evidence from Indian Manufacturing”, Journal of Political Economy, Vol. 129/6, pp. 1797-1827, https://doi.org/10.1086/713733.

[21] Torné, A. and E. Trutnevyte (2024), “Banning fossil fuel cars and boilers in Switzerland: Mitigation potential, justice, and the social structure of the vulnerable”, Energy Research & Social Science, Vol. 108, p. 103377, https://doi.org/10.1016/j.erss.2023.103377.

[74] van Ruijven, B., B. O’Neill and J. Chateau (2015), “Methods for including income distribution in global CGE models for long-term climate change research”, Energy Economics, Vol. 51, pp. 530-543, https://doi.org/10.1016/j.eneco.2015.08.017.

[80] Vandyck, T. et al. (2021), “Climate policy design, competitiveness and income distribution: A macro-micro assessment for 11 EU countries”, Energy Economics, Vol. 103, p. 105538, https://doi.org/10.1016/j.eneco.2021.105538.

[55] Vona, F. (2021), “Managing the distributional effects of environmental and climate policies: The narrow path for a triple dividend”, OECD Environment Working Papers, No. 188, OECD Publishing, Paris, https://doi.org/10.1787/361126bd-en.

[25] West, S. (2009), Distributional effects of alternative vehicle pollution control policies, Routledge, London, https://doi.org/10.4324/9781315257570.

[15] West, S. (2004), “Distributional effects of alternative vehicle pollution control policies”, Journal of Public Economics, Vol. 88/3-4, pp. 735-757, https://doi.org/10.1016/s0047-2727(02)00186-x.

[59] White, R. et al. (2023), “The unprecedented Pacific Northwest heatwave of June 2021”, Nature Communications, Vol. 14/1, https://doi.org/10.1038/s41467-023-36289-3.

[14] Winter, S. and L. Schlesewsky (2019), “The German feed-in tariff revisited - an empirical investigation on its distributional effects”, Energy Policy, Vol. 132, pp. 344-356, https://doi.org/10.1016/j.enpol.2019.05.043.

[64] Zhang, P. et al. (2018), “Temperature effects on productivity and factor reallocation: Evidence from a half million chinese manufacturing plants”, Journal of Environmental Economics and Management, Vol. 88, pp. 1-17, https://doi.org/10.1016/j.jeem.2017.11.001.

[51] Zwysen, W. (2024), Green transition and job quality: risks for worker representation, European Trade Union Institute, Brussels.

Notes

← 1.  Such “soft-linking” of CGE models and micro-simulation has some tradition in distributional analysis (Pereira da Silva, Bourguignon and Bussolo, 2008[73]; van Ruijven, O’Neill and Chateau, 2015[74]) including for climate-change mitigation (Hamilton and Cameron, 1994[75]; Metcalf et al., 2008[76]; Labandeira, Labeaga and Rodríguez, 2009[77]; Araar, Dissou and Duclos, 2011[78]). But unlike the comparative analysis of comprehensive mitigation packages here, both older and more recent studies typically focus on a single country (Costantini et al., 2025[79]), or on individual types of climate action (Vandyck et al., 2021[80]).

← 2.  Headline figures in the rest of the report are for 2040. For a distributional analysis, a shorter timespan (2035) is seen as more informative, given uncertainties in the precise pattern of key societal developments (e.g. ageing, labour markets) in the longer term.

← 3.  Differences in household carbon footprints reflect levels of development, population density, consumption patterns, production technology, and other factors. Average emissions (not shown explicitly in Figure 5.2. Emissions from household consumption are highly unequal across and within countries ranging between around 1 tonne of CO₂ per household and year Türkiye, to 6 tonnes in Poland, and 8 in France.

← 4.  At the same time, while not quantified here, more ambitious climate policies also imply reduced climate damages with potentially higher benefits for low-income households. Lower-income households are often more vulnerable to climate impacts due to limited adaptive capacity and greater exposure, particularly to health‑related damages losses (Hodok and Kozluk, 2024[56]). Thus, they could benefit significantly from climate action and its co-benefits (Section 5.3).

← 5.  In both the Current Policies and Enhanced NDC scenarios, aggregate employment growth is expected to continue globally in the next 10 to 15 years (Section 5.2) also for two of the case-study countries, with employment under Current Policies increasing by 6% in France by 2035 relative to 2024. The baseline employment gain is much bigger in Türkiye (+15%) due to a young population and the associated demographic dividend. In fast-ageing Poland, employment is expected to remain virtually unchanged under Current Policies. In the Enhanced NDCs scenario, employment would be slightly higher in France (+0.1%) and in Türkiye (+0.3%), but lower in Poland (-0.3%), compared with Current Policies.

← 6.  The quality of information on capital incomes is comparatively poor in household survey data, with known under-reporting, especially among top-income earners (Saez and Zucman, 2016[81]). As a result, capital-income aggregates produced by CGE models and survey data do not match, which is a well-known challenge for linking macro and micro models, including in the present context. The approach taken here is to maintain the information in the microdata as closely as possible, by letting household-reported capital incomes change in parallel with the CGE-estimated change for each type of income, but without attempting to correct for misreporting.

← 7.  In all three countries, a common measure of the spread within income groups (the inter-quartile ratio comparing the 25% biggest gainers in an income decile to the 25% with the biggest losses), is 4 to 8 times as large as differences of average income changes across income deciles.

← 8.  Across OECD countries, 2019 expenditures on income support for the working-age population and on non-health social services average, respectively, 3.6% and 2.3% of GDP. Eight countries spent less than 2% of GDP on working‑age income support and 18 spent less than 2% on non-health social services. See the OECD Social Expenditure Database (SOCX) (OECD, 2025[82]).

← 9.  For tractability reasons, estimates are based on the simplest of the revenue recycling scenarios, an equal lump‑sum transfer to all households. The lump-sum transfers in these cases are therefore the same as the average carbon price burden. Results on gainers and losers account for carbon price burdens at the household level, and therefore capture the variability of gains and losses across and within income groups, which remain hidden when assessing average burdens by decile.

← 10.  Across populations, recycling all the revenue from 2012‑21 carbon pricing reforms mostly creates more gainers than losers, with a smaller share of reform losers than winners in France (42%). In Poland, carbon price burdens were highly concentrated at the bottom of the income distribution and a lump-sum transfer would offset the burdens of slightly less than half of the population, leaving 53% worse off.

← 11.  The shape of the curves reflects the incidence of carbon price burdens shown earlier. They are also driven by inequalities within income groups. For instance, greater disparities of spending on fuel and other high‑emission consumption items in the lower parts of the income distribution can translate into significant numbers of people with sizeable burdens, who may then be net losers even after a lump-sum transfer.

← 12.  The Enhanced NDCs scenario assumes that governments which rely on carbon taxation or fossil fuel subsidy reforms in their climate policy packages can recycle the resulting additional fiscal revenues by allocating 50% to labour tax reductions and 50% to direct lump-sum household transfers.

← 13.  Out of 20% of workers employed in green-driven occupations:

• 46% are existing occupations whose skill set is being altered because of the green transition (referred to here as “green-enhanced skills occupations”).

• 40% are existing jobs that will be in demand because they provide goods and services required by green activities (referred to here as “green increased demand occupations”).

• Only 14% are what can properly be described as “green new or emerging occupations”.

← 14.  See Chapter 2 for a more in-depth discussion of the main mechanisms connecting climate action to economic outcomes in the Enhanced NDCs scenario.

← 15.  The selected countries are Australia, Austria, Canada, Denmark, Estonia, Finland, France, Germany, Hungary, the Netherlands, Norway, Portugal, Spain and Sweden.

← 16.  For example, the Podlaskie Voivodeship in Poland or the state of Wyoming in the United States have a share of emission-intensive occupations compared to the national average significantly higher than that of green-driven occupations and, therefore, as things stand, are areas more at risk of being left behind in the transition compared to the national average. On the opposite, the Lower Silesian Voivodeship in Poland (Dolnośląskie) or Attica (Attiki) in Greece in the bottom-right quadrant have a share of emission-intensive occupations significantly lower than that of green-driven occupations and, therefore, as things stand, are better placed to benefit from the transition compared to the national average – see discussion in OECD (2024[38]).

← 17.  Heat-stress – which depends on metabolic heat, environmental factors such as temperatures, humidity or wind speed and the clothing worn – can cause fatigue, reduced alertness and concentration, poorer information processing quality, increased reaction times, blurred vision, irritability and mood changes. Heat therefore interferes with the performance of physical tasks as well as complex and cognitively demanding activities or relatively simple and routine tasks that require special attention and vigilance (such as monitoring and control).