Abstract
Over the past decade, China has undergone an ambitious coal power phaseout that has induced a set of socio-economic rearrangements. To provide new insights into the socio-economic effects of this phaseout in addressing global climate change, we conducted an empirical analysis at both the macro and micro levels in China from 2014 to 2020. We found that the negative impacts of this phaseout led to 3.1% and 1.9% decreases in annual income for rural and urban populations, respectively. Despite facing macro-economic challenges, individuals report overall increased levels of happiness and life satisfaction. Collectively, these findings reveal a phenomenon in which macro socio-economic performance and subjective well-being (SWB) are driven in different directions during the phaseout. Our study uncovers potential socio-economic injustices as well as opportunities in the context of the coal power phaseout, highlighting the importance of flexibility in designing decarbonization strategies.
Similar content being viewed by others
Introduction
Cumulative CO2 emissions budget is rather tight if the global climate goal is getting in line with the 1.5 °C target by the Intergovernmental Panel on Climate Change (IPCC)1. Fossil fuels in general have been considered the major contributor to climate change, with coal combustion in particular accounting for 44% of global carbon dioxide emissions2. Consequently, a critical strategy to achieve climate goals is to shift rapidly away from unabated coal use in the electricity sector3. Currently, a series of global policies on coal power phaseout is expediting one of the most notable transitions ever envisioned. It is expected to yield far-reaching socio-economic consequences, affecting all human beings through changes in physical and mental health, lifestyles, and economic structures4,5,6.
China’s coal industry has undergone unprecedented development in installing substantial coal power capacity over the past decades, thereby facing greater challenges and uncertainties in the coal transition (Supplementary Fig. 1). The overwhelming magnitude of small and medium-sized coal power plants that were built in the last century in China has taken on a great responsibility for household heating, economic activities, and basic electricity services7. Yet, these plants are considerably outdated, less efficient, and highly polluting due to the low penetration of advanced combustion technologies and end-of-pipe control measures8. Recognizing the heavy and disproportionate burdens from those coal power plants, Chinese government has released a series of strengthened regulations to expedite the mandatory coal power renovation and phaseout starting from 2014, represented by the Action plan 2014–2020 for transforming the coal fired power industry and upgrading energy conservation and emission reduction9, and the 13th Five-Year Plan development plan for energy10. All those actions aim to improve energy efficiency and reduce emissions in China’s electric power generation sector, including a mandate to replace and shut down more than 10 million KW of outdated coal power plants before 20209 (Fig. 1 and Supplementary Note 1).
Before 2014, China experienced a modest level of coal power retirement. The retired capacity surge started in 2014 since the strictest policy intervention was implemented. After 2020, the retired speed slowed down due to the hit of COVID-19 and electricity shortages. We set our study period from 2010 to 2020, with the policy intervention beginning in 2014.
While a series of actions have accelerated the pace of shrinking coal-based power systems to address climate change, concerns about unintended socio-economic impacts of the coal power phaseout persist. Growing studies have examined the extent to which coal power phaseout can contribute to climate change mitigation and improve human well-being through health benefits11,12, clean energy investments6, and job creation13. On the other hand, quite a few studies have suggested neutral or even negative impacts of coal power phaseout. A typical example is the disruption to economic development and living standards at the local level. The retirement of coal power plants can bring high welfare costs due to increased social inequality14, asset stranding, and financial losses15. The different nature and skill requirements of jobs in the coal and clean energy sectors will also lead to job structural misalignment and layoffs16.
As the socio-economic impacts of coal power phaseout have been much debated, the economics of happiness is well-suited to complement direct monetary measures in this case17,18. It relies on more expansive notions of utility than does conventional economics, as represented by the Easterlin paradox, which highlights the role of nonpecuniary factors (e.g., health, social environment, and aspirations) in affecting subjective well-being (SWB)19.
Does the coal power phaseout strategy induce broader impacts on socioeconomic trajectories and subjective well-being? Most earlier studies have focused on direct stakeholders, such as coal power workers and surrounding communities14,20,21, through semi-structured interviews or model scenario analyses in specific regions. Less clear, however, are the broader impacts and implications in the context of large-scale phaseout practices. Understanding these impacts is timely and important, as socio-economic consequences can occur in direct, indirect, and induced industries throughout the entire industrial chain13. Specifically, there is limited causal evidence on how the transition away from coal power affects intangible outcomes like SWB, as well as whether it impacts the ultimate reflection of economic consequences such as residential income structure, which is related to vulnerability and adaptive capacity22.
In this study, we aim to evaluate the broader socio-economic impact of retiring small and medium-sized coal power plants in China, focusing on both macroeconomic performance and individual well-being. To achieve this, we utilize the DID (difference-in-differences) method, which is used to estimate the causal effect of the intervention while effectively controlling for confounding factors23, as widely used in previous studies24,25. By identifying the year-to-year changes following the policy intervention within a causal inference framework, we can ascertain the policy effect of the coal power phaseout.
This study has several potential contributions. While there are a few empirical studies and model scenario analyses on coal power transition21,26,27,28, the causal effects of coal power phaseout strategies on socio-economic activities and individual perceptions on a broader scale remains underexplored. In addition, using a quasi-experimental method to observe the treatment effect at the county level in China provides a more nuanced view that distinguishes our study from the existing literature. Second, this study reveals the potential mechanisms behind our findings and how those impacts are distributed across different regions and demographic groups. By complementing the existing literature on designing plant-by-plant phaseout strategies, such as those by ref. 3, our study highlights the necessity of broader socioeconomic and well-being considerations in addition to plant-level heterogeneity (e.g., plant age, stranded assets, combustion technology, and emissions). Last but not least, our study identifies that a decline in macro residential income occurs concurrently with an increase in individuals’ self-reported happiness, and the potential economic value of SWB is nontrivial. The divergent effects emphasize the importance of considering SWB as an essential indicator in policy design alongside standard macroeconomic metrics18,29. Our findings provide unique insights into how coal power phaseout can be achieved concurrently with a just transition.
Results
Effects of coal power phaseout are divergent at different levels
In this study, we construct panel datasets from 2010 to 2020 at two different levels: the county level and the individual level (Supplementary Tables 1 and 2). The period 2010–2013 served as the preceding years since the strengthening policy packages were implemented in 2014. In the DID setup, the treatment group includes the counties that have coal power plant retirement, and counties without plant retirement during the research period are included in the control groups. Individual observations are matched to the sample counties. Parallel trend assumption is tested (Supplementary Tables 3 and 4 and Supplementary Note 2). A detailed empirical strategy is provided in the Methods section.
We first estimate the effects of small and medium-sized coal power plant phaseout (SMCoalOut) on macro-level residential income (Fig. 2 and Table 1). The negative and statistically significant coefficients (p < 0.05) indicate that SMCoalOut generated about a 3.1% and 1.9% decrease in annual residential income for rural (Table 1, Columns 1–2) and urban (Table 1, Columns 3–4) populations, respectively. We also employ the alternative PSM (Propensity Score Matching) method based on propensity score to match most of the counties, and the estimated effects are 2.9% and 1.9% decreases in rural and urban residential annual income, respectively (Supplementary Note 3).
The dots and color columns are the estimated impacts and direction induced by SMCoalOut, and the error bars indicate 95% confidence intervals. The columns in green color represent the estimation directions of SMCoalOut on annual residential income at the county level, and the column in orange color represents the estimation direction of SMCoalOut on subjective well-being at the individual level. A total of 4323 (393 county samples) and 22,858 (9913 individual samples) observations are included in the county level and individual level analysis, respectively. A two-tailed t-test is performed for each coefficient.
We now turn to the impacts of SMCoalOut on individual subjective well-being (Fig. 2 and Table 2). The results diverge from the macro-level evidence, as the coefficients are positive and statistically significant. The SMCoalOut increases self-reported happiness by 0.398 points for individuals. The estimate would be an increase of 0.397 points if an alternative PSM (Propensity Score Matching) method is employed, as shown in Column (3). Our results are robust to alternative specifications and checks (Supplementary Tables 5 and 6, Supplementary Notes 4 and 5).
Complementing the results of the impacts of coal power phaseout presence (SMCoalOut), our alternative analysis on the scale of retired coal power plants finds the consistent and statistically significant impacts on macroeconomic damages and subjective well-being benefits. We employ two quantitative variables, cumulative retired capacity and the number of retired plants in a county, to replace the key dummy variable. Columns (1–4) in Supplementary Table 7 show that every thousand MW increase in retired capacity decreases rural and urban income by 6.1% and 2.7%, respectively, and each plant retirement decreases them by 1.4% and 0.7%. Similarly, Supplementary Table 7, Columns (5) and (6) report the results at the individual level, showing that every thousand MW and each plant retirement could lead to 0.648 and 0.120 point increases in individual happiness, respectively.
Potential mechanism behind the divergent effects
The central results are clear: overall, individuals report higher levels of happiness and life satisfaction despite facing macroeconomic challenges during the coal power phaseout. Therefore, the next crucial question is what explains the divergent effects on macroeconomic performance and SWB.
We first investigate whether the environmental benefits are among the potential reasons for the increased individual subjective well-being (SWB). The SMCoalOut estimates are negative and statistically significant on PM2.5 (Fig. 3a and Table 3) and are also robust under a wide of checks (Table 3, Columns 3–6). The results confirm that China’s transition away from coal power positively contributes to air quality, with a reduction of 1.54 points. Our results help to explain the potential reasons for individual’s increased happiness, as the positive impacts of air quality improvements on subjective well-being (SWB) have been widely confirmed30,31.
The dots represent the estimated impacts induced by the studied factors, and the error bars indicate 95% confidence intervals. a Estimations of environmental effect using a simple regression and the baseline DID estimator, keeping Control Group I only and changing to city-level clustering; (b) estimation of individual happiness in rural and urban populations; (c) estimations of self-rated health status; (d) estimations of self-rated relative income level in local. A total of 4323 (393 county samples) observations are included in the county level analysis in (a), and a total of 22,858 (9913 individual samples) observations were included in the individual level in (b, c, d). A two-tailed t-test is performed for each coefficient.
Understanding the factors behind individual changes in happiness is just as important as examining the macro-environmental benefits. Consistent with the macro-level analysis, the effects on happiness are estimated separately for rural and urban individuals. The positive effects are 0.545 for rural respondents and 0.357 for urban respondents (Fig. 3b and Table 4, Column 1). To explain these changes, we focus on two key self-reported indicators: employ health status32 and relative income level in local33, both of which are widely discussed in relation to their influence on hedonic happiness.
We first examine the effect of SMCoalOut on health status (Fig. 3c and Table 4, Column 2). The statistically significant and positive estimate for rural respondents indicates that the coal power phaseout intervention leads to a 0.177-point improvement in rural population’s health status. This result is in accord with our expectations, as we have confirmed the environmental benefits in the above section. The relative difficulty in accessing medical service in rural regions may drive such a sensitive increase in health status among rural population34. One possible explanation for the insignificant impact on urban populations is that urban populations have better access to medical services and greater awareness of preventive actions, such as using air purifiers and clean fuels, to mitigate air pollution’s harmful effects34. Consequently, it leads to an insignificant increase in urban individuals’ health status during coal power phaseout.
Self-rated relative income level reflects expectations, unlike absolute income, as it depends on the moving average of past and expected future incomes within a specific social setting, which can be influenced by various factors (e.g., education and health)33. Our result suggests that urban respondents experienced a statistically significant increase of 0.176 points in their relative income level with coal power phaseout (Fig. 3d and Table 4, Column 3). There are two possible explanations for the higher perceived relative income during the coal power phaseout. On the one hand, the special care to urban residents and environmental improvements during the phaseout have changed social settings, increasing urban individuals’ confidence in future33. On the other hand, they may more easily perceive an increase in economic welfare, the “comparison effect”35, if some of their reference groups are those adversely affected by negative externalities of coal power phaseout practice. Our findings contribute to the Easterlin paradox, suggesting that relative income determines hedonic happiness better than absolute income. In contrast, we do not find a statistically significant impact on the relative income levels of rural populations.
While implementing the strictest policies on coal power phaseout, China has seen an incremental construction of new coal power plants in recent years36. By introducing two new models that cover counties with new plant construction (Supplementary Table 8), we suggest that coal power construction is less likely to adequately supplement or compensate for the socio-economic impacts of coal power phaseout. The model in Panel A, Supplementary Table 8, still finds a significant but slightly lower coefficient on both rural and urban residential income, while the impacts of new plant construction on residential income are not significant in Panel B. The results indicate that although newly built plants may mitigate the negative impacts of coal power phaseout on residential income to a certain extent, they do not significantly contribute to its increase.
In addition to residential income, the contribution of coal power phaseout on air quality is no longer significant in Panel A. The coefficient for new plant construction on PM2.5 in Panel B is positive at 1.254, although the point estimate is less statistically significant (∗p < 0.1). Although the newly built plants in China are considerably cleaner and more efficient, fossil fuel emissions from these plants still contributed to increased PM2.5 in counties that experienced only new plant construction. Likewise, the contribution of coal power phaseout on emission reduction was more likely trumped, leading the insignificant effect in panel A. We further estimate the effect of new plant construction on individual subjective well-being. Column (4) in Panel B of Supplementary Table 8 shows that new plant construction had no significant effect on individuals’ subjective well-being. Unfortunately, we are unable to confirm this effect in counties experiencing both plant retirements and constructions, as these counties are not covered in the individual-level dataset (Panel A in Supplementary Table 8).
Certain regions and subpopulations responded more to coal power phaseout
We now turn to the remaining question: How can we understand the distribution of these effects among regions and demographic groups? At the county level, we examine the roles of coal power plant retirement scale, clean energy deployment, and coal-based reliance in driving the heterogeneous impacts on residential income. At the individual level, we identify three individual characteristics that are likely to induce heterogeneities in the effects of coal power phaseout on subjective well-being: age, health status, and social status (i.e., position in social hierarchy) (See Supplementary Note 9 for detailed description of the heterogeneous characteristics).
First, we address whether the retirement scale of coal power plants induces heterogeneous impacts. The results in Table 5 Columns (1) and (4) indicate that coal power phaseout may have a larger negative impact on rural residential income in counties with a larger retirement scale, although the point estimate is small. Specifically, retiring one plant is associated with a 1.2% increase in the negative impact on rural residential income. By contrast, we do not find a similar effect on urban residential income. Indeed, past studies have shown very different realities between urban and rural employees, with individuals holding rural Hukou (household registration) facing greater challenges in benefiting from the phaseout14. We, however, interpret the estimates with caution due to the insignificance estimate when clustering the robust standard errors to a higher level (see Supplementary Table 9). This might also suggest that the plant-level burdens are disproportionately distributed in a small fraction of coal power plants, and those do not linearly vary with scale of coal power capacity8.
Our results also confirm significant benefits of deploying clean energy projects37,38. When considering the interaction between clean energy deployment and coal power phaseout, the estimation is significantly and positively correlated with rural residential income but not with urban residential income (Table 5, Columns 2 and 5). Specifically, the deployment of clean energy projects in the county mitigates the negative impact of coal power phaseout on rural residential income by 3.9%. The measurable social benefit of clean energy projects to rural regions is consistent with a great number of past studies24,39. It is also easy to understand its insignificance in urban regions since clean energy projects are typically deployed in rural regions in China40.
Regions highly reliant on coal-based industries appear to face more intractable challenges during this transition, and our estimation confirms such a concern. We introduce the amount of installed coal power capacity in 2013, prior to the intervention, as an indicator of coal power reliance. This is under the assumption that countries with higher capacities have a greater reliance on coal power in employment, electricity production, and economic activities20. The coefficients of CoalReliance for both rural and urban residents are negative and statistically significant, indicating the additional burdens faced by coal-reliant counties (Table 5, Columns 3 and 6). With a 1% increase in installed coal power capacity, the negative impacts on rural and urban residential income increase 4.1% and 3%, respectively. Although ample evidence shows that such a transition creates more economic opportunities28, a decline in the coal industry and related sectors that support coal-based electricity production significantly affect the economic activities of regions heavily reliant on coal, particularly in the absence of effective regulation21. The results further imply that not only the direct stakeholders, such as coal power employees and surrounding communities, but also broader populations experienced great challenges during the transition.
Equally important is understanding the heterogeneous impacts of coal power phaseout on subjective well-being among different demographic groups (Table 6). Additional analysis by subgroups also supports the robustness of our findings (Supplementary Table 10). The coefficient of SMCoalOut * Health in column (1) of Table 6 is negative and statistically significant. The finding aligns with the mechanism analysis, indicating that individuals with poorer health conditions, particularly those suffering from related diseases32, are therefore more likely to experience happiness during the coal power phaseout.
The heterogeneous impacts of age and social status are both positive and statistically significant as shown in columns (2) and (3) of Table 6. The results suggest that elderly individuals respond more positively to coal power phaseout compared to younger adults, which is consistent with many studies demonstrating the vulnerability of older populations to traditional energy pollution41. Furthermore, people with higher social status could be happier about coal power phaseout. Although social status does not always positively correlate with wealth accumulation42, those with higher social status are usually better educated and perform well in economic interactions35. These groups are more likely to adapt innovation, such as the decarbonization strategy.
Since we attributed the increase in happiness partially to environmental improvements, we further investigated how the effect of coal power phaseout on individual perceptions varies with the baseline pollution levels. While we find that the effect of local environmental conditions on moderating the impacts of coal power phaseout on the happiness are statistically significant for both rural and urban populations (Supplementary Table 11), the results of the interaction term between coal power phaseout and PM2.5 indicate that individuals living in counties with lower air pollution are more likely to perceive increased happiness during this phaseout. In addition, we explore the heterogeneous effects of coal power phaseout on individual perceptions by dividing the subsamples into those living in counties with lower and higher initial pollution levels. Consistently, while the effects of coal power phaseout on happiness and relative income level are similar across both groups, individuals residing in counties with lower initial pollution levels experience significant improvements in health outcomes compared to those in counties with higher initial pollution levels (see Supplementary Table 12).
Therefore, our evidence further substantiates the health benefits of coal power phaseout through environmental improvements, demonstrating that the impact on health is more significant in areas with lower initial pollution levels. Our findings are in line with the theory of diminishing marginal utility, which suggests that the utility of further reducing pollution is less significant in counties with initially high pollution compared to counties with initially low pollution43,44.
Economic value of SWB in coal power phaseout is nontrivial
To better understand the extent to which individual subjective well-being could contribute to economic value, we estimated the economic benefits of subjective well-being during the coal power phaseout (Supplementary Table 13 and Supplementary Note 6). According to a back-of-the-envelope calculation using the life satisfaction approach (LSA)31, overall, this coal power phaseout raises an average person’s happiness on monetary amount worth of a total of 20192.62 Chinese yuan (≈US$2777.45).
Our estimated result is much larger than some similar studies, such as the average benefit of exiting coal quantified at US$370 per capita through human health and environmental benefits6. This is, however, justified, as our analysis encompasses a wider spectrum of SWB beyond mere health-related gains. If measured as the economic value of coal phaseout projects, our result is much in line with the average person’s economic benefit that ranges from ¥13720 (≈US$1934.85) to ¥108430 (≈US$15291.2)45. Although these estimations exhibit variations, the benefits of coal power phaseout to household wealth through increased subjective well-being are considerable compared to our negative results at the county level. Economic benefits from subjective well-being have been confirmed to accumulate through increased productivity46, future expectations47, and working performance48. However, there is no doubt that the economic value of SWB would be various if we measure it by dividing different demographic groups, as certain groups may have different perceptions during this coal power phaseout.
Discussion
Society at large may benefit from the coal power transition if the broad socio-economic impacts of retiring coal power plants are less costly. By analyzing the socio-economic impacts of coal power phaseout strategy aimed at mitigating global climate change, we provide causal evidence of its adverse impact on broader macroeconomic performance at the county level in China. Our estimates show that this phaseout decreases annual income by about 3.1% for rural populations and 1.9% for urban populations. Nevertheless, we conclude that this is also accompanied by increased levels of self-reported happiness and life satisfaction among individuals, revealing a positive socio-economic response to the coal power transition.
Our heterogeneous analysis reveals a varied distribution of those effects among regions and demographic groups. Transitioning away from coal power stands as an important near-term strategy to address climate change, whereas it also creates substantial challenges and disproportionate burdens. Vulnerable populations, including those residing in regions with a heavy reliance on coal power, larger scales of retired capacity and rural areas, and those with worse health status and older age, are more susceptible to both risks and benefits of the coal power phaseout. These findings underscore the potential socio-economic injustices as well as opportunities when implementing decarbonization strategies.
Collectively, our empirical study on coal power phaseout in China reveals a crucial implication at the aggregate level: coal power phaseout negatively impacts macroeconomic performance to some extent while still providing measurable benefits (e.g., environmental improvements and SWB enhancement). That being the case, we should not simply interpret the results as evidence that coal power phaseout necessarily leads to negative outcomes, as other aspects of the transition may yield notably different effects.
Our results highlight that rural and urban populations may perceive happiness through different channels during coal power phaseout. Previous research has shown that individuals, instead of making absolute judgments, often base their perceptions on comparisons with their environment, past experiences, or future expectations49. Relative income level is a key aspect of positional concerns and social comparisons that largely determines happiness, which is also examined in our heterogeneous analysis. Since we did not find any statistically significant difference between the coefficients for rural and urban income reductions, we are cautious in asserting that income reductions due to the coal power phaseout have increased the wealth gap. However, the positive impacts on self-reported relative income levels for urban populations, from the perspective of mental perception, may further support the significant differences between urban and rural realities reported in previous studies, suggesting that individuals with urban Hukou are more likely to benefit from the coal power phaseout14,50. In contrast, rural populations report substantial health benefits from the phaseout, highlighting the positive externalities of coal power policies on public health, especially for those in remote areas with limited access to medical services.
Despite variations in mechanisms across rural and urban populations, the impact of regional heterogeneities in the phaseout on individual happiness, such as environmental conditions, remains similar across both groups. This similarity is attributed to the relatively homogeneous environment shared by all populations within a region. This finding is further supported by the heterogeneous impacts of coal power phaseout on individual perceptions across regions with varying baseline levels of PM2.5. In addition to the discussed channels, the increase in happiness during a declining macro socio-economic performance might also be achieved by switching in household energy use types51, appreciation of land surrounding retired plants52, and improvements in mental health (e.g., depressive symptoms)53.
This study also identifies three potential opportunities to enhance socio-economic benefits as well as residential welfare during the coal power phaseout. First, it highlights the considerable environmental benefits associated with coal power phaseout, particularly improvements in individuals’ health. The health benefits are more pronounced in areas with lower initial pollution levels. By strengthening environmental regulations during the coal power phaseout, localities have the potential to achieve larger marginal health benefits, such as increased economic value from SWB. Prioritized attention should be given to heavily polluted counties to ensure that no one is left behind.
Second, individuals’ subjective well-being is significantly influenced by their socially developed characteristics and attitudes towards themselves, which are crucial for perceiving welfare54. Considering people’s aspirations when improving economic and social environments can help residents build greater confidence in the future49. In the context of the coal power transition, a higher proportion of tax revenue should be allocated to funding public services to build individuals’ adaptive capacity, such as distributed health centers and social theme campaigns54. These efforts are particularly effective for groups that have lost their sense of place or community identity or for those with lower adaptive capabilities during the coal power phaseout, helping them better perceive hedonic happines20,27.
Third, from a broader socio-economic perspective, our results indicate that investing in clean energy programs could help mitigate the negative impacts of the coal power phaseout, particularly for rural populations. Previous studies have widely recognized the contributions of clean energy programs, such as Clean Development Mechanism (CDM) projects and solar photovoltaic (PV) projects, in poverty alleviation25, job creation, and skill retraining13. In this context, access to clean energy projects offers a promising way to generate additional income and economic opportunities. In comparison, urban populations appear to experience fewer burdens during this transition. However, these adverse effects may be exacerbated in coal-reliant regions. Job transformation and vocational training within clean energy programs could assist not only in the re-employment of coal power workers but also in the industrial transformation of local economies20.
Our mechanism analysis also suggests that new plant construction is unlikely to significantly enhance residential income or happiness. Although these newly built plants are considerably cleaner and more efficient, they still contribute to an increase in net carbon emissions. The surge in new plant construction may partially offset the air quality improvements achieved by retiring outdated plants. The risks and uncertainties of newly built capacity in China align with the findings of Yan et al.8 and Zhang et al.36. Yan et al. highlight that new plant construction introduces uncertainties on the net benefits of phaseout decisions, such as health benefits, while Zhang et al. emphasize the increased risks of stranded assets. The negative impact of thermal power plant construction on rural net income is also studied by Du & Takeuchi24. However, we do not advocate for radical decisions on coal power phaseout or construction, as only well-planned and cautious approaches can effectively manage these risks and challenges. This is crucial, as a just transition can only be largely achieved when multiple efforts are made to alleviate the potential burdens of coal power transition.
Overall, this study suggests the following policy implications. First, policymakers are encouraged to strengthen the monitoring of the broader economic environment during the coal power transition to mitigate negative policy externalities, such as decreased residential income. Complementary supporting measures, such as poverty alleviation policies, clean energy investment, and job assistance, should be integrated with coal power phaseout interventions55. Second, the coal power transition creates substantial challenges and disproportionate burdens for certain groups and regions. In addition to accounting for plant-level heterogeneity (e.g., plant age, stranded assets, combustion technology, and emissions), phaseout strategies should be tailored to vulnerable groups and broader socio-economic contexts to avoid one-size-fits-all approaches. Third and finally, our study provides new insights not only into the Easterlin paradox but also into the ongoing debate on how to assess policy effects on welfare. Our results reveal a phenomenon where the effects of macroeconomic performance and subjective well-being (SWB) are driven in different directions, and the economic value of SWB is estimated to be nontrivial in this coal power phaseout. It has been demonstrated that numerous factors, including socio-economic performances and individual perceptions, are influenced in various ways by this transition. These findings emphasize the importance of considering SWB as an essential indicator in decarbonization policy design alongside standard economic metrics29,56.
Although this study makes important contributions to understanding the socio-economic impacts of coal power phaseout strategies, it does have some limitations. First, while we set stringent criteria for selecting sample counties to avoid any potential confounding factors, this approach inevitably results in the loss of certain samples. We acknowledge that data limitations prevented us from capturing the potential adjustments in operation hours of other coal power plants in the treated counties due to the retirements. As a result, the dynamic impacts of replacement and succession of coal power plants on socio-economic performance deserve further investigation. Moreover, although this study focuses on broader socio-economic impacts of coal power phaseout, future studies could explore the causal relationship of these impacts on direct stakeholders (e.g., coal power employees and surrounding communities) nationwide and the economic value of their SWB if relevant data become available. For instance, in addition to widely employed methods such as case studies or semi-structured interviews, incorporating spatial data to capture the distance between coal power plants and populations could provide more insightful and causal results. We also acknowledge that during the studied period, China underwent numerous other socioeconomic changes that could also influence SWB. Future research may build on our findings by exploring additional mechanisms or factors that may contribute to SWB.
Methods
Empirical strategy
The treatment effect in our study is observed at the county level, and the key independent variable is whether the county (including all county-level divisions) has any retirement of coal power plants in the year. We adopt a DID approach to evaluate the impacts of coal power phaseout on socio-economic performances and SWB. The sample period spans from 2010 to 2020, with 2010–2013 serving as the preceding period, as the strictest policy packages for coal power phaseout were implemented starting in 20149. Figure 1 shows that the retired capacity of coal power spiked in 2014, nearly tripling that of 2013, which justifies treating the impact as a shock rather than a continuous retirement process. Counties experiencing coal power plant retirements are designated as the treatment group, while those without any retirements during the study period serve as the control groups.
Although China’s coal power phaseout spiked starting in 2014, the following years still saw a surge in the construction of new, clean, and high-efficiency coal power plants. To avoid the confounding effects of whether newly built power plants would replace retired ones, we applied stringent sample selection criteria: we included only counties that experienced decommissioning of coal power plants with a capacity exceeding 100 MW and ceased any new plant construction since the initial plant closure during the study period. Moreover, to ensure the clean and robust treatment effect, counties with plant retirements between 2010 and 2014 were excluded from our treatment group. Of the total 458 retired plants from 2014 to 2020, over 95% had capacities below 300 MW. Given the disproportionate burdens of small and medium-sized plants relative to their installed capacity8, this phaseout practice allows us to focus on smaller, less efficient, and more polluting plants rather than newer, cleaner ones.
A total of 101 counties that reported coal power retirements from 2014 to 2020 satisfy our selection criteria. After excluding counties with data missing for several variables in the statistical books, such as rural and urban residential income of some counties, we construct a panel data with a total of 62 treatment counties and two groups of control counties. These 62 counties are representative, as their retired capacity accounts for 63% of the total capacity retired across the 101 counties.
Selecting an appropriate control group as a credible benchmark is critical for DID strategy to accurately identify causal effects57. The control counties are bifurcated into: (1) 205 counties that have coal power plants but did not experience any plant retirements after 2010; (2) 126 adjacent counties without any plants in the city where the treatment groups are located. Both groups are reasonably comparable to the treated counties. The former have coal power plants, like the treated group, with the only difference being that they did not experience plant retirement in our study period. The latter does not have any coal power plants but are comparable to the treated counties in terms of geographic proximity. To avoid potential confounding factors from preceding coal power retirements, we also estimated the results by excluding counties with any plant retirements before from our control group I. The findings remain consistent with our reported results.
Overall, we construct a balanced panel dataset consisting of 393 counties from 2010 to 2020 in our county-level analysis, with all counties having same multiple-year records. Figure 4 shows the geographical distribution of the counties in our final sample. The distribution of all counties in full sample pool can be found in Supplementary Fig. 2. Details regarding our sample selection and criteria are provided in Supplementary Note 7.
The yellow point represents treated counties that have plant retirement; the green point represents control counties that have coal power plants but without plant retirement; and blue point represents control adjacent counties. The bottom color reflects the extent of total retired capacity in provinces from 2014 to 2020. The deeper blue color in the base indicates higher retired capacity in the province. Blank areas represent provinces where no data are available. Data information on coal power retirement in some provinces, including Tibet and Taiwan, is unavailable. The distribution of all counties in full sample pools can be found in Supplementary Fig. 2. The base map was obtained from the standard map service of the Ministry of Natural Resources of the People’s Republic of China with the review number GS (2023) 2763.
Data
This study is conducted at both macro and micro levels, thus, the impact evaluation of coal power phaseout is based on two sets of data: county-level data and individual-level data.
Coal power plant data
The detailed information on coal power plants is collected from the Global Energy Monitor58. Their online data service provides one of the most comprehensive sources for global coal power plants. The data are collected at the unit level; however, we refer to each unit as a plant to avoid confusion unless explicitly stated otherwise3. We obtained information on these coal power plants, including their location, retired year, and capacity. We then identified the counties in which they are located and matched the data at both the county level and the individual level.
County level data
Data related to the key dependent variable and other social characteristic variables are compiled from the China County Statistical Yearbook and the statistical yearbooks of each county, city and province24,25. First, we employ the logarithm of per capita disposable residential income as our key dependent variable for our county-level analysis. Specifically, we estimate the impacts by analyzing both rural and urban residential income separately, instead of integrated residential income, to better capture different socio-economic impacts. We also control for some characteristic variables: GDP per capita25, share of primary and secondary industry’ GDP39, share of students in compulsory education59, government revenue24, population density60, and coal power potential20. The detailed variable description can be found in Supplementary Note 9.
PM2.5 is used to reflect air quality in the county as it is identified as the primary air pollutant of concern57. The data are obtained from Donkelaar et al.61, which estimates global annual and monthly ground-level fine particulate matter (PM2.5) from 1998 to 2021. A set of weather characteristics at the county level, including average wind velocity, total precipitation amounts, and average temperature, are used to control the climate factors in the air quality model53,57. We collected these data from the China National Meteorological Information Center.
Individual level data
Another key dependent variable in our study is the measure of self-rated happiness62. The measures of SWB and other mental self-rated variables are based on the China Family Panel Studies (CFPS), a nationally representative survey of Chinese communities, families, and individuals. The CFPS is funded by Peking University and involves a wide and comprehensive range of China’s development in society, economics, demographics, education, and health. It is a follow-up and structured questionnaire survey that is conducted through face-to-face or telephone interviews every two years since 2010, with a small-scale sample maintenance survey also conducted in the year following each full-sample survey63. Initially covering 25 provinces and 162 counties, the CFPS expanded over time, making the covered regions representative of China as a whole. Five major questionnaires were designed in the CFPS, and we adopted the adult questionnaire in our study. The detailed sampling method about CFPS can be found in Supplementary Note 8.
A total of 49 counties are covered in the database and have complete panel data, including 19 treatment counties and 30 control counties. These counties are representative, as they are distributed across 18 provinces in China, and the total retired capacity accounts for more than 40% of treated counties in the county-level analysis (Supplementary Fig. 3). A total of 22,858 observations are then matched to the sample counties (Supplementary Tables 2 and 14). The overall happiness and life satisfaction levels of respondents are relatively high, with an average value of 7.387 and 3.816, respectively. The gender ratio is close to one, and the urban-rural ratio is ~0.71. The average age of respondents is around 48 years, with an average of 7.5 years of education. 84.4% of respondents in our samples are married. The proportion of the CPC members is about 5%. In addition, the surveys were conducted throughout the year, covering all months and seasons. We provide a comparison of key demographic characteristics of all respondents who effectively answered the SWB-related questions in the CFPS survey with official statistics from the National Bureau of Statistics of China (NBS) (Supplementary Table 15). Overall, the sample distribution is reasonable and representative.
We adopt the key question in the survey from 2010 to 2020: “How happy are you with your life?” rated on a scale from 0 (extremely unhappy) to 10 (extremely happy). The indicators of SWB in CFPS have been applied to numerous studies, including those on air pollution31,53,64, energy inequality65, and environmental quality66. As the question in the first wave (year 2010) was on scale from 1 to 5 that differs from other survey years, we proportionately multiply the responses to match the scale from 0 to 10. Rich and prolonged measures of self-reported perceptions in the CFPS survey not only enable us to track individual mental changes over time during this phaseout but allow us to control a set of individual characteristic variables: marriage, gender, urbanization, age, average education year, relative income level, employment status, social status, health status and political party54. Supplementary Note 9 provides a literature check and description of control variables. We have unified the measures and units of the variables in the same categories to make the results interpretable.
Econometric models
To examine the causal effects of coal power phaseout on socio-economic outcomes, we employ the difference-in-differences (DID) estimation method with a series of fixed effects. Identification of the impacts of coal power phaseout in the DID method relies on annual changes in socio-economic performance and subjective well-being in the treated counties following the intervention, compared to the contemporaneous changes in the control counties. The fixed effects estimation allows us to control for time-invariant and time-varying unobservable county and individual characteristics that may be correlated with coal power plant retirement. We also include a seasonal fixed effect in the individual-level model to control the seasonal fluctuations67. An event study is conducted to validate DID estimator, and the insignificant estimates prior to the intervention support the parallel trend assumption (Supplementary Tables 3 and 4 and Supplementary Note 2).
County level model
The county-level analysis uses balanced panel data on 393 counties in China from 2010 to 2020. The general form of the model can be written as follows:
where \({y}_{{it}}\) represents the regional socio-economic variables, specifically the rural and urban residential annual income, in the county i in year t. \({D}_{{it}}\) is the treatment indicator, which equals one starting from the year when the coal power plants have been retired in county i, and zero otherwise. Xit is a vector of control variables. ui and λt are county and year fixed effects, respectively. εit is the error term.
Individual level model
At the individual level, we use a different panel dataset on individual subjective well-being, consisting of 22,858 observations from 49 counties, which are also covered in the county-level analysis from 2010 to 2020. The general form of the model adopted can be written as follows:
where \({g}_{{jin}}\) indicates the individual self-reported variables, basically the subjective well-being, of respondent j in county i in wave n. \({D}_{{it}}\) is the same treatment indicator as county-level model, which equals one starting from the year when the coal power plants have been retired in county i, and zero otherwise. Zin and Wjin are vectors of social and individual level controls, respectively. ui, δj, λn and θs are county, individual, year and seasonal fixed effects, respectively. εit is the error term. Alternative model specifications in our study are described in the Supplementary Note 10.
Matching strategy
Although adopting the DID estimator enables us to better control confounding factors, the sample selection may be subject to self-sample selection bias if coal power plant sites are not randomly assigned. We, therefore, adopt the Propensity Score Matching (PSM) method to mitigate potential bias by pairing treatment counties with counties that have similar observed attributes from the control pool. A set of socioeconomic and individual characteristics is used to select control groups. We have tested different matching ratios, all of which yielded similar results to our full sample analyses. The kernel density distribution of treatment and control groups before and after matching can be found in Supplementary Fig. 4. All results in our study were tested using the PSM-DID model developed by ref. 68, and the results are consistent with our basic models (See Column (3) in Table 2, Supplementary Tables 16–18, and Supplementary Note 3).
Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Data availability
The socioeconomic data at the county level supporting the findings of this study are publicly available at https://github.com/Gao2025/Coal-power-phaseout-impacts. Coal power plants dataset can be found in Global Coal Plant Tracker (https://globalenergymonitor.org/). The data at the individual level are available only by application from the Institute of Social Science Survey at Peking University at www.isss.pku.edu.cn/cfps due to privacy and ethical considerations. The dataset contains individual location, basic sociodemographic information, and economic conditions, all of which are derived and collected by the authors. Data analysis at the individual level supporting the findings of this study was conducted at the restricted data lab of the Institute of Social Science Survey at Peking University. The source data underlying all the figures in the main article and supplementary information are provided as a Source Data file. Source data are provided with this paper.
Code availability
Computer code used to process county-level data and generate the figures, tables, and results in this study are publicly available on https://github.com/Gao2025/Coal-power-phaseout-impacts.
References
Rogelj, J. et al. Mitigation Pathways Compatible with 1.5 °C in the Context of Sustainable Development. In Global warming of 1.5 °C. An IPCC Special Report on the impacts of global warming of 1.5 °C above pre-industrial levels and related global greenhouse gas emission pathways, in the context of strengthening the global response to the threat of climate change, sustainable development, and efforts to eradicate poverty (eds Masson-Delmotte, et al.) (IPCC, 2018).
Bogmans, C. & Li, C. M. International Monetary Fund (IMF). A greener future begins with a shift to coal alternatives. https://www.imf.org/zh/Blogs/Articles/2020/12/08/blog-a-greener-future-begins-with-a-shift-to-coal-alternatives (2020).
Cui, R. Y. et al. A plant-by-plant strategy for high-ambition coal power phaseout in China. Nat. Commun. 12, 1468 (2021).
Komisarow, S. & Pakhtigian, E. L. Are power plant closures a breath of fresh air? Local air quality and school absences. J. Environ. Econ. Manag. 112, 102569 (2022).
Zhang, X. et al. Immediate actions on coal phaseout enable a just low-carbon transition in China’s power sector. Appl. Energy 308, 118401 (2022).
Rauner, S. et al. Coal-exit health and environmental damage reductions outweigh economic impacts. Nat. Clim. Change 10, 308–312 (2020).
Wang, Z. et al. Unequal residential heating burden caused by combined heat and power phase-out under climate goals. Nat. Energy 8, 881–890 (2023).
Yan, X. et al. Cost-effectiveness uncertainty may bias the decision of coal power transitions in China. Nat. Commun. 15, 2272 (2024).
National Development and Reform Commission. Action plan 2014 –2020 for transforming the coal-fired power industry and upgrading energy conservation and emission reduction (in Chinese). https://chinaenergyportal.org/en/action-plan-2014-2020-for-transforming-the-coal-fired-power-industry-and-upgrading-energy-conservation-and-emission-reduction/ (2014).
National Energy Administration. 13th Five-Year Plan development plan for energy (in Chinese). https://chinaenergyportal.org/13th-fyp-development-plan-energy/ (2016).
Goforth, T. & Nock, D. Air pollution disparities and equality assessments of US national decarbonization strategies. Nat. Commun. 13, 7488 (2022).
Mayfield, E. N. Phasing out coal power plants based on cumulative air pollution impact and equity objectives in net zero energy system transitions. Environ. Res. Infrastruct. Sustain. 2, 021004 (2022).
Vanatta, M. et al. The costs of replacing coal plant jobs with local instead of distant wind and solar jobs across the United States. iScience. 25, 104817 (2022).
Wang, X. & Lo, K. Political economy of just transition: disparate impact of coal mine closure on state-owned and private coal workers in Inner Mongolia, China. Energy Res. Soc. Sci. 90, 102585 (2022).
Zhang, W., Ren, M., Kang, J., Zhou, Y. & Yuan, J. Estimating stranded coal assets in China’s power sector. Uti. Policy 75, 101352 (2022).
Cai, W., Mu, Y., Wang, C. & Chen, J. Distributional employment impacts of renewable and new energy–A case study of China. Renew. Sustain. Energy Rev. 39, 1155–1163 (2014).
Hák, T., Janoušková, S. & Moldan, B. Sustainable development goals: a need for relevant indicators. Ecol. Indic. 60, 565–573 (2016).
Du, J. et al. Global effects of progress towards Sustainable Development Goals on subjective well-being. Nature Sustain. 7, 360–367 (2024).
Easterlin, R. A. Does economic growth improve the human lot? Some empirical evidence. In Nations and Households in Economic Growth (Academic Press, 1974).
Carley, S. & Konisky, D. M. The justice and equity implications of the clean energy transition. Nat. Energy 5, 569–577 (2020).
Rud, J. P., Simmons, M., Toews, G. & Aragon, F. Job displacement costs of phasing out coal. J. Public Econ. 236, 105167 (2024).
Lutz, W. & Muttarak, R. Forecasting societies’ adaptive capacities through a demographic metabolism model. Nat. Clim. Change 7, 177–184 (2017).
Heckman, J. J. & Robb, R. Jr Alternative methods for evaluating the impact of interventions: an overview. J. Econ. 30, 239–267 (1985).
Du, Y. & Takeuchi, K. Can climate mitigation help the poor? Measuring impacts of the CDM in rural China. J. Environ. Econ. Manag. 95, 178–197 (2019).
Zhang, H. et al. Solar photovoltaic interventions have reduced rural poverty in China. Nat. Commun. 11, 1969 (2020).
Burke, P. J., Best, R. & Jotzo, F. Closures of coal‐fired power stations in Australia: local unemployment effects. Aust. J. Agric. Resour. Econ. 63, 142–165 (2019).
Mayer, A. A just transition for coal miners? Community identity and support from local policy actors. Environ. Innov. Soc. Transit. 28, 1–13 (2018).
Feng, K., Song, K., Viteri, A., Liu, Y. & Vogt-Schilb, A. National and local labor impacts of coal phase-out scenarios in Chile. J. Clean. Prod. 414, 137399 (2023).
Stiglitz, J. E. Measuring What Counts: The Global Movement For Well-being (The New Press 2019).
Li, Y. et al. A psychophysical measurement on subjective well-being and air pollution. Nat. Commun. 10, 5473 (2019).
Zhang, X., Zhang, X. & Chen, X. Valuing air quality using happiness data: the case of China. Ecol. Econ. 137, 29–36 (2017).
Steptoe, A., Deaton, A. & Stone, A. A. Subjective wellbeing, health, and ageing. Lancet 385, 640–648 (2015).
Ravallion, M. & Lokshin, M. Self-rated economic welfare in Russia. Eur. Econ. Rev. 46, 1453–1473 (2002).
Snider, G. et al. Impacts of stove use patterns and outdoor air quality on household air pollution and cardiovascular mortality in southwestern China. Environ. Int. 117, 116–124 (2018).
Weiss, Y. & Fershtman, C. Social status and economic performance: a survey. Eur. Econ. Rev. 42, 801–820 (1998).
Zhang, W. et al. Quantifying stranded assets of the coal-fired power in China under the Paris agreement target. Clim. Policy 23, 11–24 (2023).
Carley, S., Brown, M. & Gillen, M. Extending clean energy technologies to underserved communities and households. In Proc. 2023 APPAM Fall Research Conference (APPAM, 2023).
Meckling, J., Sterner, T. & Wagner, G. Policy sequencing toward decarbonization. Nat. Energy 2, 918–922 (2017).
Gao, S., Zhou, P. & Zhang, H. Does energy transition help narrow the urban-rural income gap? Evidence from China. Energy Policy 182, 113759 (2023).
Geall, S. & Shen, W. Solar energy for poverty alleviation in China: state ambitions, bureaucratic interests, and local realities. Energy Res. Soc. Sci. 41, 238–248 (2018).
Deiana, C. & Giua, L. This site is closed! the effect of decommissioning mining waste facilities on mortality in the long run. J. Environ. Econ. Manag. 119, 102797 (2023).
Smith, A. The Wealth of Nations. (The Modern Library, Random House, 1937).
Li, Z. & Jin, B. A breath of fresh air: coal power plant closures and health in China. Energy Econ. 129, 107235 (2024).
Gong, Y., Li, S., Sanders, N. J. & Shi, G. The mortality impact of fine particulate matter in China: evidence from trade shocks. J. Environ. Econ. Manag. 117, 102759 (2023).
Xie, L. et al. Who benefits from household energy transition? A cost-benefit analysis based on household survey data in China. China Econ. Rev. 77, 101878 (2023).
Oswald, A. J., Proto, E. & Sgroi, D. Happiness and productivity. J. Labor Econ. 33, 789–822 (2015).
Bartolini, S. & Sarracino, F. Do people care about future generations? Derived preferences from happiness data. Ecol. Econ. 143, 253–275 (2018).
Adnan Bataineh, K. Impact of work-life balance, happiness at work, on employee performance. Int. Bus. Res. 12, 99–112 (2019).
Stutzer, A. The role of income aspirations in individual happiness. J. Econ. Behav. Organ. 54, 89–109 (2024).
Raimi, D., Carley, S. & Konisky, D. Mapping county-level vulnerability to the energy transition in US fossil fuel communities. Sci. Rep. 12, 15748 (2022).
Zhao, J., Wang, M. & Zhu, J. Household energy transition and social status: evidence from large-scale heating renovation in China. Environ. Res. Lett. 17, 115011 (2022).
Tan, H., Thurbon, E., Kim, S. Y. & Mathews, J. A. Overcoming incumbent resistance to the clean energy shift: How local governments act as change agents in coal power station closures in China. Energy Policy 149, 112058 (2021).
Zhang, X., Zhang, X. & Chen, X. Happiness in the air: How does a dirty sky affect mental health and subjective well-being? J. Environ. Econ. Manag. 85, 81–94 (2017).
Dolan, P., Peasgood, T. & White, M. Do we really know what makes us happy? A review of the economic literature on the factors associated with subjective well-being. J. Econ. Psychol. 29, 94–122 (2008).
Liu, Feng et al. Influencing factors of public support for modern coal-fired power plant projects: an empirical study from China. Energy Policy 105, 398–406 (2017).
Krueger, A. B. & Stone, A. A. Progress in measuring subjective well-being. Science 346, 42–43 (2014).
Li, K., Yuan, W., Li, J. & Ai, H. Effects of time-dependent environmental regulations on air pollution: evidence from the Changsha-Zhuzhou-Xiangtan region, China. World Dev. 138, 105267 (2021).
Global Energy Monitor. Global Coal Plant Tracker. https://globalenergymonitor.org/ (2024).
Ge, J. & Lei, Y. Mining development, income growth and poverty alleviation: a multiplier decomposition technique applied to China. Resour. Policy 38, 278–287 (2013).
McLean, E. V. et al. Country-level analysis of household fuel transitions. World Dev. 114, 267–280 (2019).
Van Donkelaar, A. et al. Monthly global estimates of fine particulate matter and their uncertainty. Environ. Sci. Technol. 55, 15287–15300 (2021).
Levinson, A. Happiness, Behavioral Economics, and Public Policy Working Paper No. 19329 (National Bureau of Economic Research, 2013).
Xie, Y. & Lu, P. The sampling design of the China Family Panel Studies (CFPS). Chin. J. Sociol. 1, 471–484 (2015).
Xue, T., Zhu, T., Zheng, Y. & Zhang, Q. Declines in mental health associated with air pollution and temperature variability in China. Nat. Commun. 10, 2165 (2019).
Wang, Q. et al. Examining energy inequality under the rapid residential energy transition in China through household surveys. Nat. Energy 8, 251–263 (2023).
Zhang, J., Cheng, M. & Yu, N. Internet use and lower life satisfaction: the mediating effect of environmental quality perception. Ecol. Econ. 176, 106725 (2020).
Nisbet, E. K., Zelenski, J. M. & Murphy, S. A. Happiness is in our nature: exploring nature relatedness as a contributor to subjective well-being. J. Happiness Stud. 12, 303–322 (2011).
Heckman, J. J., Ichimura, H. & Todd, P. E. Matching as an econometric evaluation estimator: evidence from evaluating a job training programme. Rev. Econ. Stud. 64, 605–654 (1997).
Acknowledgements
The authors acknowledge the financial support from the National Natural Science Foundation of China, received by P.Z. (No. 71934007 & No. 72243012).
Author information
Authors and Affiliations
Contributions
S.G., P.Z., H.Z., and S.Y. conceived the initial framework. P.Z. and H.Z. gave important guidance on the framework development and finalization. S.G. and S.Y. collected, processed, and analyzed the data. S.G. wrote the first draft of the manuscript. S.G., P.Z., and H.Z. reviewed and revised the manuscript. P.Z. supervised the analysis and acquired funding. S.G., P.Z., H.Z., and S.Y. contributed to the interpretation of the results and final version of the paper.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks Kaihui Song, Zhaohua Wang, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Source data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, 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 you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. 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-nc-nd/4.0/.
About this article
Cite this article
Gao, S., Zhou, P., Zhang, H. et al. Evaluating socio-economic and subjective well-being impacts of coal power phaseout in China. Nat Commun 16, 2320 (2025). https://doi.org/10.1038/s41467-025-57561-8
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-025-57561-8
