Unlocking global carbon reduction potential by embracing low-carbon lifestyles

Article https://doi.org/10.1038/s41467-025-59269-1 Unlocking global carbon reduction potential by embracing low-carbon lifestyles Yuru Guan 1,2 ,YuliShan 2,3 ,YeHang 2 ,QingyunNie 4 ,YuLiu 5,6 & Klaus Hubacek 1 Low-carbon lifestyles provide demand-side solutions to meet global climate targets, yet the global carbon-saving potential of consumer-led abatement actions remains insufficiently researched. Here, we quantify the greenhouse gas emissions reduction potential of 21 low-carbon expenditures using a global multi-regional input-output model linked with detailed household expenditure data. Targeting households exceeding the global per-capita average required to stay below 2 degrees, our model captures changes in direct energy use, household consumption and upstream intermediate industrial inputs. We find that implementing a combination of low-carbon expenditures among the top 23.7% emitters reduces global carbon footprints by 10.4 gigatons CO 2 e (i.e., 40.1% of the household consumption-basedemissionsofthe116countries analysed in this study or 31.7% of the global total in 2017). Consumption pattern changes related to mobility and services contribute 11.8% and 10.2% of emission reductions. North America shows substantial reduction potential, while some Sub-Saharan African countries present unexpected mitigation possibilities. However, a rebound effect from re-spending income savings from lifestyle changes offsets the expected carbon savings by 6.5% to 45.8% (0.7–4.8 gigatons CO 2 e). Recent work has highlighted the growing importance of demand-side mitigation solutions to achieve global climate targets, as supply-side measures cannot be solely relied upon1–3. From a consumption perspective, accounting for all upstream emissions along global supply chains, household consumption triggers directly and indirectly around two-thirds of total greenhouse gas (GHG) emissions, thus transitioning household consumption patterns towards low-carbon modes should be a critical part of mitigating climate change4,5. The literature on carbon inequality emphasizes the importance of equitable demand-side measures6. Addressing climate change requires a multifaceted approach that includes cutting emissions from high emitters while supporting those who face barriers to low-carbon transitions, such as energy poverty. Research found thatthe top 10% of emitters accounted for 48% of the global emissions in 2019, with the top 1% contributing 16.9% to the global total, while the bottom 50% emitted 12%7. Between 1990 and 2019, the bottom 50% contributed only 16% of global emissions growth, whereas the top 1% accounted for 23% of the total7. This disparity highlights that a relatively small, wealthy part of the global population predominantly drives consumption-based emissions8,9. This situation underscores the urgent need to propose demand-side measures that specifically target carbon-intensive activities among top emitters, as those households have contributed most to climate change and have the greatest capacity for reducing emissions. A recent study by Büchs et al. exemplifies this by investigating the carbon savings by hypothetically reducing the energy consumption of the top 20% of consumers across Received: 12 March 2024 Accepted: 15 April 2025 Check for updates 1 Integrated Research on Energy, Environment and Society (IREES), Energy and Sustainability Research Institute Groningen (ESRIG), University of Groningen, Groningen, the Netherlands. 2 School of Geography, Earth and Environmental Sciences, University of Birmingham, Birmingham, UK. 3 Birmingham Institute for Sustainability and Climate Action, University of Birmingham, Birmingham, UK. 4 School of Management, Nanjing University of Posts and Telecommunications, Nanjing, China. 5 College of Urban and Environmental Sciences, Peking University, Beijing, China. 6 Institute of Carbon Neutrality, Peking University, Beijing, China. e-mail: [email protected];[email protected] Nature Communications | (2025) 16:4599 1 1234567890():,; 1234567890():,; 27 European countries to the level of the 80th percentile. Such an intervention could reduce GHG emissions by 9.7% of these countries’ total emissions10. Additionally, emerging and developing economies (as classified by the International Monetary Fund11), particularly those with large populations and rapidly industrialising sectors, have become significant contributors to global carbon emissions. Within these economies, households with high consumption levels are playing an increasingly prominent role in driving carbon emissions. However, there is limited research focusing on demand-side reduction measures aimed at these high-emitting segments of the population across countries worldwide. To effectively achieve demand-side mitigation, adopting lowcarbon lifestyles that minimise GHG emissions is essential12,13.Lifestyle is a multifaceted construct including behaviours, cognitions, and contextual factors14,15. Household expenditure often serves as a reliable proxy for individual lifestyles, reflecting choices across transportation, food, housing, and consumer goods2,16. By examining consumptionrelated behaviours, such as product selection and usage patterns, we can better understand the drivers of household carbon emissions. For instance, decisions to purchase durable appliances, consume natural fibre clothing or reduce food waste directly impact household carbon footprint. A deeper understanding of these behaviours is crucial for designing effective interventions to promote low-carbon lifestyles. The ‘avoid-shift-improve’framework provides a holistic perspective on effective low-carbon lifestyle changes that correspond to actions through three distinct approaches: absolute reduction, consumption pattern shift, and efficiency improvement1,12. For example, minimising food waste aligns with the ‘avoid’approach, focusing on the absolute reduction of food consumption. Shifting from private vehicles to public transportation represents a ‘shift’approach, and opting for seasonal food consumption thereby reducing required energy inputs in the agricultural sector can be categorised as an ‘improve’approach for upstream industries. These examples illustrate how the ‘avoid-shift-improve’framework can guide us in making informed choices and taking actions that contribute to a more sustainable, low-carbon future. Following this framework, a growing body of literature explores low-carbon lifestyle transitions. Their emphasis primarily centres on a few key domains, including food17,18, mobility19,20, and buildings21–23. However, many of these assessments have a relatively narrow scope, which focuses only on one or a few of the major domains24. A limited number of studies have considered multiple measures, but even these studies examined different measures separately5,16.Moreover,many existing studies have centred on high-income countries (as classified by the World Bank25) such as European countries5,17,theUnited Kingdom26, the United States15,andJapan 27, or selected emerging and developing economies such as India and China28,29. This preference is mainly driven by richer data availability, higher environmental awareness, and stricter environmental policies in these regions5,15,17,26,27. While the measures proposed in these studies, applied uniformly to all populations of these countries, can influence the consumption of carbon-intensive products, they may disproportionately affect vulnerable populations, particularly those already struggling to achieve decent living standards30,31. Furthermore, given that top global emitters are not confined to high-income countries but come from all world regions7, it becomes increasingly important to explore the mitigation possibilities of high-carbon households in emerging and developing economies5. When discussing lifestyle changes through consumption-related behaviours, it is crucial to acknowledge that carbon savings are frequently counteracted by rebound effects32–34. An example is that residents adopting home insulation measures may result in direct rebounds, such as turning up the thermostat to increase comfort, and/ or indirect rebounds, that is spending the remaining savings on other products and services35. The direct rebound effect, primarily focusing on direct energy consumption36, has been extensively studied and factored in the design of energy-saving guidelines or policies, exemplified by initiatives in the United Kingdom and Ireland32,35.These guidelines mainly recommend considering additional energy use (e.g., a 20% increase in heating demand35) when estimating potential energy savings from interventions like the installation of energy-efficient household boilers. However, there exists a notable gap in comprehending and quantifying broader, indirect rebounds encompassing all upstream processes, mainly due to uncertainties in re-spending patterns and limitations in detailed data on household expenditures33,37,38. In this paper, we conduct an assessment of the household carbon footprints across various consumption levels, using a modified household expenditure database derived from the World Bank Global Consumption Dataset (WBGCD) and a global multi-regional inputoutput dataset from the Global Trade Analysis Project (GTAP)8,31.Our analysis covers both direct GHG emissions from home fuel use and private transport and indirect or upstream GHG emissions resulting from household consumption activities creating emissions along global supply chains. This allows us to pinpoint carbon footprint hotspots among population groups, regions, and consumption categories. Households that exceed the global average carbon targets (aiming to stay below 2 degrees)are modelled in low-carbon lifestylechanges. We select 21 low-carbon expenditures across food, diet, mobility, buildings, clothing, manufactured products, and services. We simulate the carbon reduction potential of these 21 lifestyle changes on household direct energy use and other final consumption and upstream emissions along the entire global supply chain39. Our analysis quantifies the aggregated carbon reduction potentials achieved through a combination of low-carbon expenditures for 116 countries and looks closer into household-specific mitigation outcomes. Furthermore, we discuss how much of the expected mitigation benefits attributed to these lifestyle changes may be offset by indirect rebound effects under three re-spending scenarios. We aim to enhance the understanding of rebound effects throughout the entire supply chains and shed light on the magnitude of these effects utilising our detailed household expenditure data. Results Hotspots of global household carbon footprints We calculated the carbon footprint in 2017 for households with varying consumption patterns of 201 expenditure groups across 116 countries representing 79.5% of global GDP and 87.3% of the global population including 86 emerging and developing economies. This calculation encompasses direct GHG emissions from fuel use in home and private transport and indirect GHG emissions associated with household consumption activities, emitted in production processes along global supply chains. Figure 1a shows that higher expenditures translate into higher carbon footprints among household deciles on a global scale7,9.The carbon footprint of the poorest decile amounts to 0.5 t CO 2 epercapita in 2017, while the wealthiest decile has an average carbon footprint of 15.6 t CO 2 e per capita. The size of the household carbon footprint is closely linked with their consumption patterns. In general, expenditure on buildings and food tends to be the largest contributors to GHG emissions. Poorer households predominantly contribute GHG emissions through food consumption, while wealthier households exhibit greater shares of emissions from services and mobility. The distribution of carbon footprints among the four gases varies significantly: food consumption primarily drives CH 4 and N 2 O emissions, while emissions from F-gases are predominantly linked to the consumption of manufactured products (Fig. 1b). We derived a range of targets for global annual consumptionbased carbon emissions informed by 2-degree-consistent pathways projected by integrated assessment model scenarios40. 23.7% of the global population (1.6 billion) have a per-capita carbon footprint Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 2 exceeding the global annual average target (i.e., the 2020 upper limit of 4.6 t CO 2 e per capita), 89.0% of whom live in highand uppermiddle-income countries. The carbon-exceeding households are responsible for about 78.1% of global expenditure and contribute 63.7% of consumption-based emissions of the 116 countries analysed in this study. Figure 2shows disparities in carbon footprints across regions. The proportion of carbon-exceeding households varies considerably across regions. North America has the highest average footprint (17.2 t CO 2 e per capita), with 85.4% of its population exceeding the global 2020 target. Lower-footprint regions such as Sub-Saharan Africa show substantial variation, with 5.4% of the population having a much larger footprint (9.7 t CO 2 e per capita) compared to its carbon-compliant households. Turning our attention to individual countries, India, the third-largest global consumption-based GHG emitter in 2017 (2.2 gigatons (Gt) CO 2 e), has only 3.7% of its population comprising carbonexceeding households. In contrast, Luxembourg, a high-income country with low total emissions (14.2 Mt CO 2 e), have the largest share of carbon-exceeding households (99.7% of its population). China exhibitsbothhighcarbonfootprints(5.2GtCO 2 e) and a substantial share of carbon-exceeding households (24.0%). This highlights the complex relationship between regional development, income levels, and carbon footprints. The occurrence of carbon exceedance extends beyond just high-emitting or advanced economies, highlighting the necessity for nuanced strategies that aim to reduce household carbon footprints globally, with a focus on demand-side mitigation measures. Reduction potential from engaging in various low-carbon expenditures Here, we present the results of implementing lifestyle-oriented mitigation measures for carbon-exceeding households across 116 countries. We modelled the reduction potential of 21 low-carbon expenditures separately (see detailed descriptions in Supplementary Data 2). These expenditure-related mitigation measures align with the ‘avoid-shift-improve’framework1,12, addressing carbon reduction through absolute reduction, consumption shift and efficiency improvement. Our results provide a deeper understanding and enable the comparison of the mitigation benefits associated with the adoption of low-carbon expenditures for specific household groups. Figure 3shows considerable emission reduction outcomes from the implementation of these lifestyle changes for carbon-exceeding households. The 21 selected low-carbon expenditures offer global GHG reduction potentials ranging from −0.01% (‘Natural Materials’)to 10.9% (‘Nonmarket Services’) including avoided upstream emissions. Given that higher costs on leisure activities among carbon-exceeding households, low use of commercial services (abbreviated as ‘Nonmarket Services’) presents a mitigate potential of 10.9%. In the diet category, shifting towards a healthy vegan diet—reducing consumption of animal-based food, sugar, and unhealthy processed food products—presents a promising reduction potential to reduce global GHG emissions by 8.3%. The four diet-related expenditures focus on reducing specific food consumption (primarily animal-based food) and substituting them with plant-based alternatives, which overlap in their approach but differ in the targeted products (see more in Supplementary Data 2). In the buildings category, implementing passive house standards could result in a 6.0% carbon reduction. Turning to manufactured products, implementing sharing and repair initiatives for home appliances (‘Share & Repair’) could contribute to a 3.0% reduction. Regarding mobility, adopting modes such as transitioning from private vehicles to public transportation (‘Less Cars’), working from home, and halving air travel could potentially reduce carbon emissions by 1.4–3.6%. It is worth noting that working from home reduces emissions from land transport but leads to an increase in home energy use, thus weakening the overall mitigation. In the food category, reducing food waste yields more modest results (1.3%), while opting for seasonal and organic food has minimal impacts (0.1–0.8%). Fig. 1 | Global household carbon footprints per population decile in 2017. a Per capita carbon footprint by consumption categories (stacked bars). Mobility covers direct emissions from private transport and indirect emissions induced by public transport. Buildings encompass direct emissions from residential fossilfuel use and indirect emissions from home electricity and energy services use, construction, and building material use. The other four categories cover indirect emissions from all corresponding upstream emissions (see details in Supplementary Data 1). The added dashed line shows a global annual target of 4.6 t CO 2 e per capita per year for household carbon footprints in 2020, aimed at limiting climate warming to below 2 degrees. bHousehold carbon footprint distribution by categories and gases (CO 2 , CH 4 ,N 2 O, and F-gases). Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 3 In clothing, changing to fibres has limited impact but extending the lifespan of garments through practices such as swapping and repairing (‘Durable Fashion’) could achieve a 1.2% reduction in global GHG emissions. Overall, ‘Avoid’measures show higher mitigation potential, followed by ‘Shift’strategies with moderate impact. ‘Improve’measures contribute to reductions but largely rely on production-based advancements for deeper decarbonisation. We also found that mitigation potentials differ among regions (Fig. 4). The effectiveness of lifestyle changes arises from differencesin infrastructural conditions, supply chains with respective production patterns and energy mix, as well as consumption patterns among households and countries. North America and Europe and Central Asia consistently exhibit higher relative reduction potentials across various low-carbon expenditures compared to the global average, whereas South Asia demonstrates the lowest mitigation potential in most changes. Adopting a healthy vegan diet offers considerable potential for carbon reduction across regions. For example, in Latin America & Caribbean, such a diet could achieve a 17.4% reduction in carbon emissions. This substantial potential aligns with the region’s existing dietary challenges, characterised by unhealthy dietary patterns and reliance on carbon-intensive food consumption patterns18. Similarly, at the country level, Mongolia stands out with notable reduction potential by adopting different diets. Households in Mongolia contribute considerably to emissions from food consumption, accounting for 59.8% of its total carbon footprint in 2017. When comparing three low-carbon lifestyles related to mobility in each country, 63 out of 116 countries have greater reduction potentials by adopting ‘Less Cars’compared to the other two mobility-related expenditures. 33 countries benefit more from ‘Working from Home’ and 20 countries see more advantages from ‘Less Flying’. Implementing passive house standards can significantly cut residential energy use, leading to considerable carbon reductions across many countries, particularly in North America, and Europe and Central Asia. In some nations, such as Kyrgyzstan and Sweden, ‘Durable Fashion’exhibits a relatively higher potential for carbon emissions reduction, with reductions of 3.6% and 2.7%, respectively. Furthermore, embracing ‘Share & Repair’practices for home appliances could result in considerable emissions reductions in 94 out of 116 countries, surpassing the potential reductions achieved by adopting ‘No Chemicals’and ‘Durable Appliances’. Households in North America and Europe & Central Asia can achieve greater carbon reductions compared to other regions through three services-related expenditures: lower use of commercial services, decreased longdistance leisure travel, and proximity-based services. Mitigation potentials from combining low-carbon expenditures When assessing the cumulative effects of implementing these actions simultaneously, we assumed widespread adoption of all low-carbon expenditures by global carbon-exceeding households. To avoid double counting, we accounted for overlapping impacts, particularly where multiple actions target similar household activities. For example, we selected the diet with the highest reduction potential to avoid overestimating carbon savings due to the overlap among the four dietrelated expenditures. Additionally, we excluded actions unlikely to reduce emissions, such as using natural building materials in certain Fig. 2 | Regional carbon emission distribution in 2017. a Regional population distribution within global emitter groups, with the carbon-exceeding population shares indicated. bRegion-average carbon footprints for household groups (i.e., total population, carbon-exceeding households, and carbon-compliant households). Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 4 countries. Our aim is to estimate the maximum achievable carbon savings rather than to provide precise predictions. Globally, implementing a combination of expenditure-focused mitigation measures could potentially lead to a 10.4 Gt CO 2 ereduction in the carbon footprint, representing 40.1% of the household consumption-based emissions of the 116 countries analysed in this study or 31.7% of the global household carbon footprint in 2017 (Fig. 5). The reduction primarily originates from changes in household consumption volumes and patterns among carbon-exceeding populations, particularly in mobility (3.0 Gt CO 2 e or 11.8%), services (2.6 Gt CO 2 e or 10.2%) and food (2.1 Gt CO 2 e or 8.2%). Changes in clothing expenditures contribute the least (0.2 Gt CO 2 e or 0.9%). Breaking down by gases, CO 2 accounts for the largest share at 29.8%, followed by CH 4 (7.9%), N 2 O (1.8%), and F-gases (0.6%). The relative aggregated GHG reduction varies across 116 countries, ranging from 2.3% in Democratic Republic of the Congo to 72.3% in Malta. We observe large relative mitigation potentials for countries in North America (3.7 Gt CO 2 e or 66.7%), Europe & Central Asia (3.3 Gt CO 2 e or 53.8%), and Latin America & Caribbean (0.8 Gt CO 2 eor41.0%). The United States could achieve the largest absolute reduction (3.7 Gt CO 2 e) mainly through expenditure changes in mobility and services (see details in Supplementary Fig. 2). For European countries, a set of expenditure-focussed mitigation measures could result in substantial decreases in the carbon footprint, particularly in Luxembourg (67.2%), Denmark (64.6%), and Greece (64.5%). Luxembourg shows huge potential, with reductions of 9.5 Mt CO 2 ecomparedtoitsGHGemissions in 2017, primarily due to the extensive engagement of a substantial portion of its population (99.7% are carbon-exceeding households) in such demand-side mitigation. Figure 6illustrates significant deviations (ranging from −7% to 33%) between carbon savings achieved by targeting carbon-exceeding households with a combination of low-carbon expenditure measures and those obtained by applying the same changes to a corresponding proportion of “average consumers”in each country—a common practice in the literature41,42. These deviations highlight the heterogeneity in household consumption patterns, with high-emitting households demonstrating a higher potential for reducing their carbon footprint. In North American and European nations, the observed deviations remain relatively modest despite a high potential for emission reductions. This can be primarily attributed to the substantial presence of carbon-exceeding households within the randomly selected average consumers in these affluent nations. Notably, we found that some Sub-Saharan African countries, such as Mauritius (achieving a 52.5% reduction in this study), Namibia (45.6%), and Chad (44.7%), often overlooked in previous studies, display relatively higher deviations. Namibia’sreductionsaredrivenbychangesinfood, mobility and services expenditures, due to its heavy dependence on the tourism industry. Figure 7shows different relative reduction potentials across population groups with varying expenditure levels. For instance, households in China display a median reduction of 52.13% (with the 25th–75th percentile a reduction ranging from 52.07% to 52.14%), while in Angola, this median stands at 54.5% (with the 25th–75th percentile between 44.0% and 69.8%). We also observe a regressive distribution pattern in several countries where lower household expenditure levels correlate with higher relative reduction potential in countries such as Angola, South Africa, Luxembourg, and Eswatini. This pattern doesnot universally apply, as seen in other nations such as the United States, Finland, Kazakhstan, and China, where wealthier households may benefit more from lifestyle-oriented measures in terms of carbon reduction than others. Additionally, certain countries show unique distributions; for instance, in Belgium, Austria, Greece, and Malta, households with higher consumption levels could achieve mid-level mitigation results. Fig. 3 | Global carbon reduction potentials of 21 low-carbon expenditures. Letters ‘A’,‘S’,and‘I’refer to three low-carbon approaches: avoid, shift, and improve, respectively. Overlaps between different expenditures and potential rebound effects are not considered in this figure. Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 5 Offsetting potential reductions through rebound effects The adoption of low-carbon expenditures can lead to money savings. However, unintended rebound effects, where these saved expenditures are re-spent, can offset the initial mitigation benefits38.Ourstudy excludes direct rebound effects, as it is unlikely for consumers to increase their consumption of the same fuel product due to energy efficiency gains within the context of our low-carbon lifestyle scenarios33,36. Instead, we focus on indirect rebound effects among final consumers, exploring how saved money is reallocated into consumption of other goods and services and the subsequent rebound in upstream carbon emissions throughout the supply chain. To estimate these reallocations, a double-semi-log regression model38,43 is used to calculate marginal expenditure shares across different household groups and countries based on expenditure BuildingsClothing Manufactured Products Services FoodMobility Carbon footprint reduction within region (in %) Middle East& North Africa Sub-Saharan Africa Latin America &Caribbean Europe& Central Asia East Asia &Pacific North America South Asia Proximity Services Community Services Nonmarket Services Share & Repair Durable Appliances No Chemicals Natural Fiber Durable Fashion Natural Materials Min Construction Work Passive House Less Cars Less Flying Seasonal Food Organic Food Healthy Vegan Vegan Vegetarian Mediterranean Work from Home Less Waste S S S S A I A A S A A S S I A A A A S S I Diet 0% 5% 10% 15% 20% 3.5% 2.0% -0.0% 0.5% 0.0% 1.5% 1.9% 2.3% 4.0% 0.7% 0.8% 0.0% 0.9% 1.2% 1.6% 1.5% 0.5% 2.0% 7.3% 2.8% 1.8% 2.6% 0.5% -0.0% 0.3% 0.0% 5.6% 6.8% 7.2% 8.4% 1.2% 0.4% 0.0% 0.8% 0.4% 1.0% 0.7% 0.5% 1.0% 3.4% 2.1% 1.0% 1.8% 0.6% 0.0% 0.4% 0.0% 11.6% 14.5% 15.2% 17.4% 2.3% 1.0% 0.0% 2.9% 0.8% 3.3% 3.1% 1.0% 2.2% 7.2% 6.6% 2.8% 1.2% 0.3% -0.0% 0.4% 0.0% 1.2% 1.5% 2.4% 2.6% 0.4% 0.2% 0.1% 0.2% 0.2% 0.8% 0.3% 0.3% 0.4% 2.5% 1.0% 0.5% 9.3% 1.8% -0.0% 1.5% 0.0% 4.2% 5.3% 6.9% 10.0% 1.6% 1.0% 0.1% 1.8% 1.5% 4.8% 2.9% 2.1% 4.4% 13.8% 8.3% 4.3% 5.3% 0.5% -0.0% 1.0% 0.0% 2.0% 2.5% 3.0% 6.8% 1.1% 1.1% 0.1% 0.5% 1.2% 1.7% 0.2% 0.4% 2.1% 6.9% 1.8% 1.7% 8.6% 0.7% 0.0% 1.1% 0.0% 3.4% 4.2% 4.9% 8.4% 1.2% 0.6% 0.0% 1.5% 1.6% 4.5% 7.1% 2.9% 7.7% 20.9% 9.8% 5.6% Fig. 4 | Regional carbon reduction potentials through 21 low-carbon expenditures. The colours indicate the reduction compared to the baseline (i.e., carbon footprints of 2017 within each region). The darker the colour, the greater the reduction potential. Letters ‘A’,‘S’,and‘I’refer to three low-carbon approaches: avoid, shift, and improve, respectively. Specific numeric results are presented. Overlaps between different expenditures and potential rebound effects are not considered in this figure. Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 6 elasticities36,43,44 (see more details in the Methods section). These regression results describe the additional expenditure on a specific commodity for every one-dollar increase in total expenditure, offering a comprehensive insight into how households distribute additional spending across various consumption categories, thereby contributing to the evaluation of potential rebound effects. Our initialevaluation focuses on rebound effects across 21 distinct low-carbon expenditure measures, where we assumed that all saved money is re-spent on products unaffected by specific low-carbon lifestyle changes, following consumers’marginal expenditure patterns. For example, reduced air travel decreased spending on airfare but left expenditures on other non-aviation goods constant. The freed-up funds from air travel savings were then distributed proportionally (based on their marginal propensity to spend) among these nonaviation travel items and other goods. We found considerable disparities in rebound effects across different lifestyles (see Supplementary Fig. 4). Service-related expenditures tend to have greater backfire effects, as savings from cutting back on leisure activities are often redirected to areas with higher carbon intensity, such as increased spending on food or extended time at home (raising residential energy use). When exploring the potential magnitude of rebound effects resulting from combined lifestyle changes, we designed various alternative re-spending patterns, taking inspiration from refs. 33,37,38. As shown in Fig. 8, our analysis involves three rebound scenarios (SC1SC3). SC1 assumes that all saved money is re-spent on products unaffected by low-carbon expenditures. In this case, it is estimated that 4.8 Gt CO 2 e GHG reductions are offset, resulting in a rebound effect of 0% 20% 40% 60% 80% 20% 40% 60% 80% 20% 40% 60% 20% 40% 60% 80% a cd ef CO2reduction (in %) CH4reduction (in %) F-gases reduction (in %) N2O reduction (in %) b GHG reduction (in %) 0% 20% 40% 60% Buildings 6.1% Clothing, 0.9% Services 10.2% Manufactured Products, 2.9% Mobility 11.8% Food 8.2% GHG reduction 10.4 Gt CO2e Fig. 5 | Cumulative carbon savings from a combination of low-carbon expenditures. The colours of the map in subplots (a,c–f) show relative reduction potentials of GHG, CO 2 ,CH 4 ,N 2 O, and F-gas compared to their 2017 emissions across 116 countries. The pie chart in subplot (b) shows the contributions of changes in product consumption to global GHG reduction. The basemap layer is derived from Runfola, D. et al. geoBoundaries: A global database of political administrative boundaries. PloS one 15, e0231866 (2020), published under the CC BY 4.0 license. Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 7 45.8%. Re-spending on the buildings category leads to a reduction loss of 2.1 Gt CO 2 e. Drawing inspiration from Druckman et al. 33 and Grabs et al.38, SC2 and SC3 focus on redirecting all saved money towards the six lowest carbon-intensive products across six consumption categories (SC2) and the least carbon-intensive products among all expenditure items (SC3). While the outcomes of SC2 and SC3 are unlikely in reality, they provide insights into the lower bounds of the rebound effects. Under SC2, we found that 4.5 Gt or 43.8% of the expected global carbon reductions resulting from low-carbon lifestyles are eroded due to such re-spending. Notably, 3.1 Gt CO 2 eof this loss is induced by re-spending on mobility. Meanwhile, SC3 reflects the best case, indicating that only 0.7 Gt CO 2 e of carbon savings would be lost, resulting in a rebound effect of 6.5%. SC3 involves the re-spending of saved expenditures primarily on services-related products characterised by lower embodied carbon intensity compared to products in other categories. We also discovered diverse impacts and potential magnitude of rebound effects experienced by different countries, stemming from 0% 20% 40% 60% 80% 0% 20% 40% 60% 80% Reduction from average consumers Reduction from carbon-exceeding households North America Europe & Central Asia East Asia & Pacific Latin America & Caribbean Middle East & North Africa South Asia Sub-Saharan Africa N a mibi a E s watin i C ha d Mauritiu s Ni g e r M a lt a Mon g olia So uth Af ric a Fig. 6 | Comparison of national carbon savings through low-carbon expenditures implemented by carbon-exceeding households and randomly selected average consumers. The y-axis represents national carbon savings achieved by carbon-exceeding households through low-carbon expenditures. The x-axis shows the reductions resulting from applying the same changes to an equivalent proportion of “average consumers”in each country. 5 54 562 5810 Carbon footprint reduction (in %) Per-capita expenditure (2017US$ per capita per day) Malta Greece Austria Belgium United States Finland Kazakhstan China Eswatini Luxembourg South Africa Angola 40% 50% 60% 70% 80% Fig. 7 | Relative carbon reduction potentials of households from lifestyle changes in selected countries. The x-axis displays the relative carbon reduction potentials of household groups, indicating the reduction potential of each household group in relation to its 2017 baseline carbon footprint. The median and 25th–75th percentiles (bars) are shown. Sample sizes (n) = 116, 114, 95,109, 67, 83, 99, 105, 95, 91, 90, and 112 for Angola, South Africa, Luxembourg, Eswatini, China, Kazakhstan, Finland, United States, Belgium, Austria, Greece, and Malta, respectively. Sample sizes refer to the number of household groups per country; the number of people per household group varies. See Supplementary Fig. 3 for results for all 116 countries studied in this paper. Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 8 various re-spending patterns (ranging from 12.8% to 228.8% in SC1, 3.8% to 135.0% in SC2, and 0.3% to 17.1% in SC3) (see Supplementary Fig. 5). Due to relatively higher energy costs or the absence of consumption level saturation32, emerging and developing economies are prone to experiencing more pronounced backfire situations caused by such re-spendings than advanced economies. That underscores the heightened importance of implementing rebound mitigation policies specifically tailored for emerging and developing economies. Discussion This research shows the potential of adopting low-carbon lifestyle changes among high-carbon households, leading to a considerable contribution to GHG emission mitigation at a global scale. Our estimations reveal that implementing a combination of low-carbon lifestyles in 23.7% of the top-emitting population could potentially result in a 40.1% reduction in household consumption-based GHG emissions in the 116 analysed countries (equivalent to 31.7% of the global household carbon footprint in 2017). We observed that nations in North America, Europe & Central Asia, and Latin America & Caribbean exhibit substantial potential for reducing GHG emissions mainly due to their high per-capita carbon footprints and the large number of households involved in our lifestyle-oriented mitigation actions. An interesting finding of this article is the unexpected demand-side mitigation possibilities observed in some countries in Sub-Saharan Africa, such as Mauritius, Namibia, and Chad, which have been overlooked in previous studies. When exploring emission hotspots among carbon-exceeding populations, we identified key consumption categories with high climate relevance such as buildings and food. Within the buildings category, major contributors to emissions encompass home energy use, construction works, and building material. Our mitigation scenarios highlight the large potential for carbon reductions through lowcarbon lifestyle changes, particularly in services, diet, and buildings. Interventions such as reducing food waste, limiting long-distance travel and leisure activities, and shifting towards plant-based diets offer immediate and tangible benefits. These ‘low-hanging fruits’represent key opportunities for policymakers to achieve progress with relatively straightforward measures. However, measures targeting the clothing sector, yield only little impact on GHG reductions primarily stemming from the inherently low-carbon-intensive characteristics of the clothing sectors. Thus, while impactful and easily implementable interventions are available, effective emissions reduction requires prioritising the most significant sources and opportunities for change. The question of how to realise low-carbon lifestyle changes is pivotal, as an individual’s choices are intricately linked to income and consumption levels, willingness, resource accessibility, and fiscal and policy frameworks2. A combination of regulatory, economic, and information-based instruments, referred to as a “policy package”,is generally more effective in achieving these transitions than relying on single policy instruments alone45. Governments worldwide have initiated various policies to support lifestyle changes, especially in critical areas of buildings, diet, and mobility. For example, countries such as Spain and Pakistan have promoted shorter working times and encouraged remote work to conserve energy46. Investments in transport infrastructure, such as cycle lanes and high-speed rail systems, facilitate a shift towards more sustainable modes of mobility47.In response to the energy price crisis triggered by the Russian-Ukrainian conflict, European nations like Germany, the Netherlands, and France, have initiated campaigns aimed at fostering energy-saving actions, including lowering the heating temperature and reducing showers31,46. Furthermore, carbon pricing mechanisms, such as taxes and cap-andtrade systems, have proven effective in altering consumer behaviour by incorporating the environmental costs of carbon-intensive goods48. Subsidies and tax incentives for renewable energy adoption, alongside information-based interventions such as carbon labelling and customised information feedback49–51,have facilitated the transition towards greener consumption patterns. The effectiveness and feasibility of these policies are likely to vary across different countries and income groups. In high-income countries, where infrastructure and resources are already in place, more aggressive measures may be viable52.Similarly, low-income countries should avoid investing in carbon-intensive infrastructures that would lock them into high carbon-intensive expenditures53. For high-income groups, policies could focus on curbing the consumption of luxury goods with high carbon footprints, potentially through progressive taxation or incentives for adopting low-carbon alternatives39,48. Conversely, in lower-income regions, the 5.0 6.0 7.0 8.0 9.0 10.0 0.0 6.0 7.0 8.0 9.0 10.0 0.0 9.8 10.0 10.2 0.0 Expected Reductions Reductions after Rebound SC1: Re-spent on unaffected items Rebound: 46% SC2: Re-spent on the six lowest carbon-intensive items Rebound: 44% SC3: Re-spent on the lowest carbon-intensive item Rebound: 6% Expected Reductions Reductions after Rebound Expected Reductions Reductions after Reboun d -0.0 -0.1 -2.1 -0.0 -1.9 -0.6 -0.3 -3.1 -0.1 -0.2 -0.5 -0.4 -0.0 0.0 -0.0 -0.0 -0.0 -0.7 Global Carbon Reductions (Gt CO2e) Buildings Clothing 5.6 10.4 5.8 10.4 9.7 10.4 Mobility Food Manufactured Products Services Buildings Clothing Mobility Food Manufactured Products Services Buildings Clothing Mobility Food Manufactured Products Services abc Fig. 8 | Global carbon reduction offset triggered by rebound scenarios SC1-SC3. Expected reductions based on lifestyle changes and the reductions after the rebound effect under each scenario are depicted with grey bars in (a–c). The absolute loss reductions by expenditure categories are highlighted with red numbers. The percentages in the arrow show the rebound effect (defined as the ratio of offsetting carbon reductions resulting from re-spending saved money to expected reductions, see more in the Methods section). Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 9 Additional information Supplementary information The online version contains supplementary material available at https://doi.org/10.1038/s41467-025-59269-1. Correspondence and requests for materials should be addressed to Yuli Shan or Klaus Hubacek. Peer review information Nature Communications thanks Hazel Pettifor, David Andersson, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available. Reprints and permissions information is available at http://www.nature.com/reprints Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/ licenses/by/4.0/. © The Author(s) 2025 Article https://doi.org/10.1038/s41467-025-59269-1 Nature Communications | (2025) 16:4599 16