Bridging Economic and Environmental Health Principles to Mitigate Air Pollution Impacts

Air pollution remains one of the most severe environmental threats to human health. In 2021, it contributed to nearly 7 million premature deaths globally, ranking among the top risk factors alongside hypertension and poor diet (Bennitt et al., 2025). Despite notable improvements in air quality in several countries due to policy interventions, approximately 99% of the world’s population still lives in areas where pollutant levels exceed the World Health Organization (WHO) global air quality guidelines (AQG). Different countries now face divergent challenges: While countries with moderate or low pollution levels find further abatement increasingly difficult, others remain stuck at persistently high pollution levels. New pollution episodes, such as climate-induced wildfires, further complicate the governance landscape.

Research on the impacts of air pollution has advanced significantly across several key disciplines, most notably environmental epidemiology and economics. Environmental epidemiology has long provided robust evidence linking pollutant exposure to mortality, cardiopulmonary diseases, and other health outcomes. In recent years, the economics literature has emerged, by applying causal inference methods and high-resolution spatiotemporal data to estimate the marginal effects of pollution on health and socio-economic impacts, and to evaluate policy effectiveness. Meanwhile, a parallel line of modeling research has employed integrated assessment frameworks to examine the long-term pathways and outcomes of air pollution control strategies. Together, these research streams have deepened understanding of air pollution– health mechanisms and policy impacts. Yet significant limitations remain in designing effective and adaptive policy responses that align with health and environmental goals.

These limitations are reflected in four critical gaps in the existing literature, many of which stem from the fact that current economic frameworks for air pollution control have not fully incorporated pivotal findings from modern environmental health science. These scientific advances are fundamentally challenging conventional assumptions and revealing the limitations of existing analytical tools, thereby limiting their direct applicability to policy decisions.

First, most existing studies often lack a unified analytical framework that bridges both long-term (chronic) and short-term (acute) perspectives in pollution control and exposure mitigation. This temporal gap hinders the design of targeted policies, making it difficult for policymakers to formulate evidence-based strategies that strategically allocate resources across short-term and long-term interventions for maximizing sustained public health improvements.

Second, the typical application of benefit–cost analysis (BCA) often does not adequately capture phased intervention strategies, which remain relatively underexplored. BCA is conventionally applied to evaluate individual options for achieving a defined policy goal, rather than a phased series of strategies designed to adapt to evolving conditions. This is a critical omission because recent epidemiological evidence reveals that the exposure–response relationship between pollution and health is often nonlinear, with marginal health benefits potentially increasing as a region moves from high to moderate or low pollution levels. A phased approach is therefore essential to tailor policy interventions to the specific and evolving cost–benefit context of different pollution stages. For instance, high-pollution phases require information on rapid emission reductions, moderate phases need data on marginal abatement costs (MAC) and marginal health benefits, and lowpollution phases demand insights into risk management and adaptation measures.

Third, while economic analyses increasingly incorporate spatial dimensions, effectively translating the profound heterogeneity across regions into policy evaluation remains a key challenge. Specialized tools can now map differences in population exposure (e.g., BenMAP; Science Advisory Board, 2024), and research highlights vast spatial variations in abatement costs and health benefits (e.g., Shapiro and Walker, 2020). However, fully capturing this complexity is crucial because health impacts themselves are not uniform; significant adverse effects have been observed even at low pollution concentrations, suggesting no clear “safe threshold”. A spatially explicit economic analysis is thus crucial for the efficient allocation of resources and for identifying geographic areas where interventions can yield the greatest net benefits. This spatial heterogeneity is also central to understanding the distributional consequences and equity implications of pollution control, discussed next.

Fourth, air pollution governance must increasingly navigate multiobjective decision-making, balancing goals such as climate change mitigation, adaptation, and social equity. Standard BCA is often not sufficiently supplemented by systematic distributional analysis and careful discussion of non-quantified effects, making it difficult to evaluate air pollution control in the context of other national priorities. This oversight is critical, as policies can have significant distributional consequences, particularly for vulnerable populations who often bear disproportionate health risks and may also be disproportionately burdened by abatement costs; ideally, an analysis incorporating distributional considerations should consider the distribution of overall net benefits. An integrated framework is required to systematically assess synergies (such as health co-benefits from climate action) and trade-offs, ensuring that pollution control strategies are not only efficient but also equitable and climate-compatible.

To develop such an integrated framework, this paper synthesizes theoretical principles and recent advances from the global environmental health and economics literature. In this process, we find that the recent experience of China offers a particularly critical context for observing these principles in practice. Given China’s rapid, large-scale policy interventions against severe pollution challenges and high population density, its governance setting provides crucial insights into the application, evolution, and complex trade-offs of these economic strategies. This review therefore draws substantially upon the insights emerging from this governance setting, as the lessons learned from China’s complex trade-offs are highly relevant for other regions facing similar development and environmental pressures. India offers a parallel case as a populous, heavily polluted nation, where analyses also underscore immense health costs and the need for context-specific, cost-effective control strategies (Cropper and Park, 2022).

In this context, this paper aims to integrate recent advances in environmental health sciences and explicitly embed core principles from environmental economics into an analytical framework tailored to the evolving realities of air pollution governance. Specifically, this paper seeks to:

1. Develop a dual-track BCA framework that distinguishes between long-term and short-term health impacts and responses, and aligns them with appropriate policy instruments and evaluation logics;

2. Analyze regional marginal benefits (MB)–MAC curves incorporating recent health and cost evidence, and propose a three-phase air pollution control strategy that is differentiated not only by pollution concentration levels but also by spatial context. This involves demonstrating how regional heterogeneity in pollution characteristics, population exposure, and abatement costs fundamentally alters the marginal benefit and cost curves within each phase, thus guiding a more tailored and efficient approach to policy prioritization;

3. Examine how air pollution control can be integrated with climate mitigation and adaptation objectives, as well as environmental equity considerations;

We emphasize that the goal of this paper is to develop and synthesize the underlying analytical framework by bridging economic and health principles. It does not aim to provide an exhaustive survey of all specific evaluation tools (such as established models like BenMAP; Science Advisory Board, 2024) or all regional policy practices. In developing and illustrating the analytical framework, this paper will primarily focus on fine particulate matter (PM_2.5) as the primary example for several key reasons. First, PM_2.5 represents the most significant contributor to the global disease burden from air pollution, making its effective control a top public health priority. Second, the vast and rapidly evolving body of research on PM_2.5 provides the most robust empirical evidence for the complex characteristics central to our analysis, including its nonlinear dose-response relationships, lack of a safe threshold, and heterogeneous impacts across populations. While this paper centers on PM_2.5 to clearly articulate the framework’s mechanics, the underlying principles for evaluating longand short-term trade-offs, spatial disparities, and multi-objective synergies are broadly applicable to other major air pollutants.

In sum, this paper contributes to updating the theoretical basis for evaluating the efficiency of air pollution control pathways and offers an extensible framework to support cross-disciplinary research and policy design. While relevant across contexts, the framework may be especially valuable for countries with high population density, rapid development, and limited policy capacity, as it helps clarify governance strategies suited to those specific conditions.

The remainder of this paper is structured as follows. Section 2begins by reviewing the scientific evidence on the complex health impacts of air pollution and the evolution of policy responses over the past two decades, establishing the foundation for our analysis. Building on these insights, Section 3 develops the core dual-track BCA framework, deconstructing traditional economic assumptions to better reflect real-world dynamics such as nonlinear health benefits and evolving abatement costs. Section 4 then applies this framework to propose a three-phase strategy, outlining how policy priorities should shift as a region moves through different stages of pollution severity. Section 5 extends the analysis by examining how this framework can integrate other critical objectives, particularly climate change and social equity. Finally, Section 6 concludes with a discussion of the framework’s policy implications and offers an outlook on future research.

Over the past two decades, fine particulate matter (PM_2.5) and groundlevel ozone have emerged as the dominant air pollutants contributing to global health burdens (Lelieveld et al., 2020; Murray et al., 2020). In countries that have adopted aggressive air quality policies, particularly in China, PM_2.5 concentrations have declined sharply (Greenstone et al., 2021; Jin et al., 2016; Zheng et al., 2018). For example, between 2013 and 2022, China’s PM_2.5 levels fell by 57% under the “Blue Sky Defense War” and related emissions control campaigns.¹ However, these reductions have not translated proportionally into public health gains. PM_2.5-related premature mortality in China remains high, largely due to demographic aging and sustained exposure to low-concentration but still harmful pollution levels (Cao et al., 2024; Xue et al., 2022).

In contrast, ground-level ozone pollution has shown an upward trend in many regions, driven by rising temperatures and complex atmospheric chemistry. For example, in recent years, China’s ozone concentrations increased despite the sharp decrease in concentrations of PM_2.5 (Greenstone et al., 2021; Li et al., 2020), amplifying respiratory and cardiovascular mortality risks (Qiu et al., 2024). Traffic emissions, a major source of nitrogen oxides (NOx) which include nitrogen dioxide (NO₂), are key drivers of both this ozone formation and direct health issues like asthma (Erickson et al., 2020; Guarnieri and Balmes, 2014). Globally, road traffic’s NOx emissions contribute significantly to ozonerelated mortality (Silva et al., 2016).

Climate variability and extreme events now pose additional challenges to air pollution control. Wildfires, exacerbated by climate-linked droughts and heatwaves, exemplify this challenge. Critically, while recent global burned area has declined, human exposure to fires has increased by 40% due to enhanced colocation of fires and settlements (Teymoor Seydi et al., 2025). This intensified exposure to fire-sourced PM_2.5 amplifies health impacts (Zhong et al., 2025), and US projections show climate-driven smoke deaths could exceed 71,000 annually by 2050 (Qiu et al., 2025b). Events like the 2020 California wildfires also demonstrate acute impacts beyond mortality, including spikes in mental health-related emergency visits, disproportionately affecting vulnerable populations (Jung et al., 2025).

These observations point to a growing paradox in air pollution management: even as emissions decline and regulatory targets are met, health burdens remain persistently high and increasingly shaped by non-traditional factors such as climate feedbacks, population aging, and spatial disparities in exposure.

Recent findings from environmental health research provide important insights into how air pollution affects human health across different time scales, population groups, and environmental conditions.

First, air pollution produces both acute and chronic health effects, which differ in onset, severity, and clinical manifestation. Short-term exposure, such as hourly or daily spikes in PM_2.5 or ground-level ozone, can provoke acute events including strokes, heart attacks, and asthma exacerbations, typically peaking within 24 to 72 hours (Chen et al., 2022a; Jiang et al., 2023). In contrast, chronic exposure over longer durations (usually 6 months or more) contributes to the development of long-latency diseases such as lung cancer, neurodegenerative disorders, and respiratory decline, ultimately shortening life expectancy (Lelieveld et al., 2020; Liang et al., 2020; Niu et al., 2022; Yin et al., 2017a).

Second, the relationship between air pollution concentrations and health outcomes is nonlinear. Large epidemiological cohorts have demonstrated that even low levels of PM_2.5 exposure are associated with measurable increases in cardiovascular and respiratory mortality (Di et al., 2017; Yazdi et al., 2021). The slope of the exposure–response curve tends to be steeper at lower concentrations, indicating heightened sensitivity to air pollution in cleaner environments (Burnett et al., 2014; Li et al., 2018). In studies of short-term exposure, the duration of high-concentration episodes has emerged as an important determinant of health outcomes. Evidence shows that longer-lasting pollution events result in more severe acute effects, underscoring the health risks posed by multi-day exposure spikes (Xia et al., 2022; Zhang et al., 2021).

Third, the effects of air pollution are not evenly distributed across the population. Infants, children, older adults, pregnant individuals, and those with chronic illnesses exhibit heightened vulnerability (Chen et al., 2017; He et al., 2022a; Liang et al., 2020; Yin et al., 2017b). For example, prenatal and early-life exposure has been linked to impaired lung development and increased risk of adverse birth outcomes (He et al., 2022b; Yuan et al., 2023), while cumulative lifetime exposure in the elderly correlates with elevated rates of cardiopulmonary mortality (Li et al., 2018; Liang et al., 2020; Yang et al., 2020). Socioeconomic disadvantage further compounds these risks: lower-income communities are more likely to reside near pollution sources and have reduced access to healthcare, mechanisms potentially involving both structural barriers and residential sorting dynamics, leading to persistent disparities in exposure and outcomes (Banzhaf et al., 2019a; Jbaily et al., 2022; Josey et al., 2023).

Finally, meteorological and climatic conditions such as seasonality and weather patterns significantly modify the health effects of air pollution. Ozone concentrations often peak during summer months due to heatwaves and enhanced photochemical activity, leading to increased respiratory morbidity (Coates et al., 2016; Gu et al., 2020; Kalisa et al., 2018; Tian et al., 2020). During winter, the frequent occurrence of temperature inversions, coupled with increased emissions from heating, significantly contributes to higher PM_2.5 concentration levels (Glojek et al., 2022; Ning et al., 2018). Meta-analyses have confirmed consistent seasonal variation in air pollution-related disease burdens (Bell et al., 2005; Orellano et al., 2020). In shorter time windows, compound exposure events, such as concurrent extreme heat and air pollution, can result in amplified health impacts, especially among sensitive populations (Huang et al., 2023c).

In light of growing evidence on the multifaceted health impacts of air pollution, such as shortand long-term effects, population vulnerabilities, and seasonal variation, global and national policy responses have entered a new phase of ambition and precision.

The WHO’s 2021 global AQG marked a turning point by significantly tightening recommended limits for major pollutants, addressing both chronic and acute exposure risks. The annual PM_2.5 guideline was lowered to 5 µg/m³ to protect vulnerable populations from long-term exposure risks, while short-term guidelines (e.g., 24-h or 1-h values) were affirmed as essential tools to communicate health risks and protect susceptible groups from acute pollution episodes (World Health Organization, 2021).

At the national level, regulatory frameworks are evolving in response to the persistence of health burdens at lower pollution levels. The US Environmental Protection Agency revised its PM_2.5 standards to reflect newer health science findings (US EPA, 2023), while there is a growing discourse and expert recommendation in China on updating its air quality standards, reflecting an increasing shift from concentrationbased control to exposure-based risk awareness (Zhu et al., 2022).

The integration of climate and health policy has also gained momentum. At COP28, over 120 countries signed the Declaration on Climate and Health, which formally recognized air pollution as a climateexacerbated health risk, intensified by wildfires, heatwaves, and other extreme weather events (COP28 UAE, 2023). The declaration underscored the need for joint action and co-benefit approaches that align mitigation of climate and air pollution threats.

Additionally, recent global initiatives have called for stronger commitments to reduce air-pollution-related health impacts, including discussions around long-term targets such as halving health burdens by mid-century compared to 2015.² Although formal timelines vary, these calls reflect increasing urgency to go beyond traditional industrial source control and shift toward a risk management paradigm.

The review of the past two decades thus reveals a landscape of increasing scientific complexity, highlighting the nonlinear, thresholdless, and heterogeneous nature of air pollution’s health impacts, as well as the dynamic realities of pollution control. This evolving evidence base underscores the limitations of conventional policy approaches based on simple concentration targets. To effectively translate these scientific insights into robust and efficient policy, a more integrated and dynamic economic framework is essential. The following chapter addresses this need by developing a tailored BCA framework to navigate these real-world complexities and better guide strategies for safeguarding public health.

The traditional economic framework for analyzing air pollution control centers on balancing the MB of pollution reduction against the MAC. Figure 1 illustrates this classical MB–MAC framework, where the intersection of these two curves theoretically defines the socially optimal level of pollution.

However, this classical framework rests on several simplifying assumptions that diverge significantly from recent empirical evidence regarding air pollution’s health impacts and the complexities of abatement. As will be further discussed and illustrated in the rest of this section, these deviations have profound implications for policy.

On the health benefit side, the classical model often fails to capture critical real-world nuances. The crucial role that the shape of the concentration–response function plays in determining the magnitude and distribution of health benefits from abatement policies has been a central topic in environmental health economics (Pope et al., 2015). For example, extensive epidemiological research indicates that pollutants such as PM_2.5 and ozone lack clear “safe” thresholds (Di et al., 2017). Moreover, these relationships often exhibit steeper slopes at lower pollution concentrations, implying that marginal health benefits from pollution reduction can be greater in relatively cleaner environments (Burnett et al., 2014; Li et al., 2018). Furthermore, a single, averaged marginal benefit curve, as depicted in the classical model, cannot adequately account for the substantial heterogeneity in health impacts across acute versus chronic effects, different population subgroups, and varying temporal scales (Liang et al., 2020; World Health Organization, 2021). These variations are crucial as they influence both the magnitude and timing of health outcomes.

On the cost side, the traditional assumption of a smooth, well-defined MAC curve, representing a readily identifiable least-cost abatement path, is also an oversimplification. As detailed further in Section 3.3, the true position and shape of the MAC frontier are often unknown due to uncertainties in technological advancements, policy learning, and the practical challenges of identifying and implementing the most efficient sequence and combination of abatement interventions (Harrington et al., 2000; Jaffe et al., 2002; Kesicki and Strachan, 2011).

In this section, we focus on two representative categories of health outcomes to characterize the structure of health benefits from air pollution reduction in real-world contexts:

1. Premature mortality risks, often estimated through populationlevel exposure and cause-specific death data, remains the most widely used and policy-relevant indicator of air pollution harm.

2. Morbidity risks such as disease onset and acute symptom aggravation, particularly in respiratory and cardiovascular systems, represent more immediate and clinically observable health effects. These endpoints capture both new cases and short-term health shocks triggered by pollution spikes.

In long-term economic assessments, we can define the per capita annual marginal health benefit of air pollution control as the value of reduced health risks resulting from a small and sustained improvement in air quality. This section focuses on one representative outcome: the benefit associated with a one-microgram-per-cubic-meter (1 µg/m³) annual reduction in PM_2.5 concentration, measured in terms of reduced premature mortality risk.

The per capita marginal health benefit can be expressed as:

Baseline mortality rate × Exposure–response coefficient × Value of reduced mortality risks

The monetary valuation is most commonly derived using the value of reduced mortality risks (traditionally called Value of Statistical Life, VSL), a widely accepted metric in BCA (OECD, 2012; US EPA, 2024; Wang et al., 2024b). Typically, VSL estimates are derived from studies measuring the willingness to pay (WTP) of an average individual to exchange their own income for a small reduction in their own mortality risk within a specific period; this individual WTP is then divided by the risk change to estimate the value per statistical case, making it particularly suitable for evaluating welfare impacts of public health policies. Importantly, this WTP-based VSL theoretically encompasses all aspects of an individual’s valuation of the risk change, including not only non-pecuniary factors like the value of life itself and subjective well-being (Levinson, 2012; Li et al., 2014; Zhang et al., 2017) but also pecuniary considerations such as expected changes in out-of-pocket medical costs and future earnings. Alternative Cost of Illness (COI) approaches are sometimes used; these include using the estimated cost of premature deaths in terms of healthcare expenditures or productivity losses, but their underlying logic differs significantly from the welfare-based WTP framework and they are generally not considered appropriate measures for marginal benefit calculations in BCA.

Not all pollution-induced health damages result in premature death. Many individuals suffer from pollution-related diseases, such as cardiovascular or respiratory illnesses, without immediate mortality. In such cases, per capita marginal benefits from reduced disease incidence can be similarly estimated using:

Baseline morbidity rate × Exposure–response coefficient × Value of reduced morbidity risks

While morbidity-related COI measures based on medical costs (Barwick et al., 2018) or productivity losses (He et al., 2019; Wang et al., 2022) are sometimes used to quantify certain economic burdens, the theoretically preferred approach within BCA relies on estimating individuals’ WTP to avoid illness, with recent estimates provided by studies within the OECD SWACHE project (e.g., Dockins et al., 2024). Monetized Quality-Adjusted Life Years (QALYs), though debated in BCA contexts due to differing theoretical foundations, offer another alternative valuation perspective derived from health economics (e.g., Robinson et al., 2022). Neither COI nor monetized QALYs fully capture the welfare impact as measured by WTP.

It is important to recall that the relationship between pollution levels and health risks, as discussed in detail in Section 2.2, is generally nonlinear. While the shape of the exposure–response curve varies by health outcome, studies focusing on endpoints such as cardiovascular mortality have often found steeper slopes at lower pollution levels, implying substantial per capita health gains from further reduction even in relatively clean environments.

Moreover, pollution control is frequently accompanied by socioeconomic development and rising public awareness of health risks. These dynamics tend to increase the valuation parameters such as value of reduced mortality and morbidity risks over time. Rising income contributes to this trend, reflected in the typically positive income elasticity of WTP for health risk reductions and the standard practice of adjusting valuation estimates accordingly. However, other factors beyond income, such as shifts in risk perception and perceived government efficacy in pollution control, have also been identified as significant drivers, particularly in rapidly developing societies (Wang et al., 2024b). As a result, the per capita marginal health benefit curve may rise during the pollution reduction process, meaning that the cleaner the air, the more valuable each additional unit of improvement becomes from a long-term welfare perspective.

Population size amplifies the total health benefits of air pollution control. At the regional level, the annual marginal health benefit of reducing pollution can be approximated by multiplying the per capita benefit by the total number of residents.

This amplification effect by total population is particularly significant in policy terms within densely populated regions. While high emission intensity from agglomeration economies traditionally makes pollution control in densely populated regions challenging (Wu et al., 2023), such areas also stand to gain disproportionately large health benefits from abatement efforts. The sheer number of people affected (i.e., a large total population concentrated in a specific geography) means that even incremental improvements in air quality can translate into substantial collective health gains. This highlights that, despite their management complexity, high-density areas represent crucial opportunities for impactful public health interventions (Baumgartner et al., 2020).

More refined assessments estimate benefits separately by population subgroups and disease categories, since exposure–response relationships vary by health outcome. These relationships are often nonlinear and differ across illnesses (Burnett et al., 2014; Burnett et al., 2018). In addition, children and older adults generally exhibit stronger physiological responses to pollution exposure (He et al., 2022a; Liang et al., 2020), which increases the relative health impact in areas with larger shares of vulnerable groups. Population aging, combined with population growth, has offset a large portion of the health gains from recent air quality improvements in China (Cao et al., 2024).

On the valuation side, estimates of the value of reduced mortality and morbidity risks associated with children tend to be higher than population averages (Alberini et al., 2010; Robinson et al., 2019). Monetizing health risks for older adults involves several economic considerations. Beyond WTP-based valuations for risk reduction, which can be complex for this group due to factors like life expectancy (Aldy and Viscusi, 2008; Ketcham et al., 2024), a critical consideration for public finance is the substantial healthcare savings achievable by preventing severe illness or premature death. Given their higher prevalence of chronic conditions and greater healthcare utilization, reducing older adults’ exposure to air pollution can yield considerable avoided medical expenditures (Deryugina et al., 2019). While these savings are theoretically distinct from the welfare gains captured by WTP, they represent a significant, tangible outcome relevant to public budgets and are an important factor in policy evaluations. Given continued population aging and the heightened health sensitivity among older individuals, long-term regional health benefits are expected to rise as pollution levels fall.

In sum, contrary to standard textbook assumptions of diminishing returns, this analysis reveals that the regional marginal health benefit curve can paradoxically rise as air quality improves. This phenomenon is primarily driven by local population characteristics, with high-density regions consistently exhibiting a higher benefit curve, underscoring the amplified health gains achievable in populous areas. Furthermore, it is important to consider the dynamic feedback effects: reduced mortality from cleaner air can lead to larger future populations, as more individuals survive and have children, thus altering the baseline population for future benefit calculations (a complexity addressed in some simulation models (Liu et al., 2025)).

Figure 2 does not explicitly include a time dimension. In reality, the path from high to low air pollution may take decades in some regions (Aldy et al., 2022; Parrish et al., 2012), while others may achieve rapid improvements within a few years, like the case of China (Greenstone et al., 2021). The current illustration should be understood as a static, annual-level snapshot. A region may remain at a certain pollution level for an extended period or move quickly toward cleaner air, depending on the pace of intervention.

This distinction carries important implications. All else being equal, regions that act earlier to reduce pollution can spend more years in cleaner conditions, where the annual marginal health benefits of further improvement are higher. In contrast, delayed action means prolonged time in areas of the curve where health benefits from pollution reduction are smaller. Even if the overall transition period is similar, early intervention can lead to greater cumulative health gains over time.

Short-term health effects refer to increased risks of illness, symptom aggravation, or even death that occur within hours or days following exposure to air pollutants. These effects are typically linked to temporary spikes in pollutant concentrations over periods ranging from a few hours to several weeks. A large-scale multi-city time-series study in China found that for every 10 µg/m³ increase in PM_2.5, O₃, and NO₂, daily all-cause mortality risk increased by 0.22%, 0.24%, and 0.90%, respectively (Chen et al., 2017, 2018; Yin et al., 2017b).

These short-term effects are not readily represented in the marginal health benefit diagram discussed earlier. They reflect temporary deviations around a constant long-term concentration. Because the horizontal axis of Figure 2 represents annual average pollution levels, short-term events correspond to a single point on that axis. As such, they do not reveal how marginal health benefits change as annual pollution declines. We will return to this in a dedicated short-term BCA framework in Figure 6 later in the paper.

While it is common to estimate short-term health costs using constant marginal effects, recent studies suggest that the exposure–response relationship is also nonlinear in short-term settings. For instance, evidence from China shows that the risk of mortality associated with short-term PM_2.5 exposure increases linearly at lower concentrations but levels off at higher ones (Chen et al., 2017). In contrast, the short-term effects of NO₂ appear closer to a linear pattern (Chen et al., 2018).

In addition to examining concentration–response relationships, episode-based studies focus on how the combination of pollution intensity and duration shapes short-term health outcomes. Empirical findings consistently show that the longer a high-concentration pollution event lasts, the more severe the health consequences (Xia et al., 2022; Zhang et al., 2021). Temporary but decisive interventions that reduce the frequency and duration of these episodes can therefore produce clear health benefits.

When aggregating short-term health benefits at the regional level, population size remains the primary determinant. Population characteristics also matter: older adults, children, and individuals with chronic illnesses tend to be more sensitive to short-term pollution fluctuations (Chen et al., 2017; Gu et al., 2020; He et al., 2022a). In addition, meteorological conditions and seasonal factors can significantly shape the health risks associated with short-term exposure (Rai et al., 2023; Shi et al., 2020). We discuss how to incorporate short-term effects clearly into benefit–cost analysis of air pollution control policies in Section 3.4.2.

Turning to the cost side of long-term air pollution control, the MAC curve is a central concept. In this paper, and to maintain consistency with the benefit framework, the MAC is specifically defined as the societal cost associated with achieving an additional unit of sustained pollution reduction, typically measured as the annual cost for a 1 µg/m³ decrease in the annual average pollution concentration. Figure 3 illustrates the stylized MAC curve and various potential real-world abatement cost paths.

Theoretically, an idealized MAC curve, often depicted as an “efficient frontier” (as shown by the thick solid line in Figure 3), can be conceptualized for the pollution control pathway. This curve represents the hypothetical lowest possible societal cost to achieve each incremental level of pollution reduction. However, identifying this true MAC frontier in practice faces significant challenges. Crucially, firms’ abatement cost information is often private and not readily observable by regulators. Consequently, estimations typically rely on approximations derived from engineering cost models, industry knowledge, or inferences drawn from market-based mechanisms like emissions trading.

Furthermore, the realization of such an ideal abatement path hinges on a set of stringent and often simultaneously challenging conditions in the real world:

• Sufficient scientific understanding of the entire pollution process, including emissions, atmospheric dispersion, chemical transformation, and human exposure. Note that these processes are often highly context-specific across regions (Seinfeld and Pandis, 2016; Viana et al., 2008).

• Availability and maturity of key abatement technologies for all relevant pollutants at every stage of pollution control.

• Clear policy objectives and well-aligned governance mechanisms, with actions implemented in line with declining exposure levels (Jacobsen et al., 2020); This also often implies coordinating control across pollutants such as PM_2.5, ozone, and NO₂, which requires clear prioritization and resource allocation based on well-defined objectives (Liao et al., 2008; Sillman, 1999).

• Functional institutions and market mechanisms that enable MAC to converge across sectors and actors, ensuring cost efficiency (Stavins, 2003; Wu et al., 2015).

The classical economic framework often presents the ideal MAC frontier as a static curve for a given point in time (represented by the thick solid line in Figure 3). However, this depiction can obscure the inherently dynamic and endogenous nature of how long-term abatement costs truly evolve. The position and shape of this “efficient frontier” are not fixed but are, in reality, continually reshaped by a confluence of real-world processes over time.

Several interconnected mechanisms drive this dynamism and contribute to substantially lower long-term abatement costs. Firstly, learning effects, often termed “learning by doing”, play a significant role, as industries gain experience with pollution control measures, leading to declining marginal costs for established abatement methods (Arrow, 1962; Yeh et al., 2005). This is vividly illustrated by the dramatic cost declines in renewable energy technologies like solar and wind, which serve as a powerful, structural approach to air pollution control (He et al., 2020). More fundamentally, policy can induce major technological innovation. As argued in seminal work by Jaffe et al. (2002), the design and stringency of environmental regulations directly spur research and development, leading to novel, more cost-effective abatement technologies or inherently cleaner production processes. This concept of “induced innovation” is central to understanding how policy can actively drive down future abatement costs, fundamentally lowering or altering the MAC frontier.

Secondly, the evolution of institutional design and policy instruments determines how, and to what extent, these potential cost reductions are realized in practice (Dietz et al., 2003; Goulder and Parry, 2008; Guttman et al., 2018). The shift from rigid command-and-control regulations to more flexible, market-based instruments is notable for its theoretical ability to achieve emission targets at a lower societal cost by allowing firms to find their most cost-effective compliance strategies. Real-world applications, however, reveal significant complexities. The US Acid Rain Program, for example, illustrates how the flexibility of a cap-and-trade system can uncover unforeseen low-cost compliance pathways, such as widespread fuel switching, which played a more significant role than technological innovation in scrubbing alone (Chan et al., 2018).

Furthermore, recent research on US air pollution “offset” markets highlights additional institutional challenges: their price signals reveal vast spatial heterogeneity in abatement costs, and in most cases, these market-revealed costs are far lower than the marginal health benefits, suggesting that even sophisticated regulations can be insufficiently stringent to achieve economically efficient outcomes (Shapiro and Walker, 2020).

Considering this inherently dynamic nature of the MAC frontier, the actual pollution control pathways realized by different regions (as depicted by the illustrative red and blue dashed lines in Figure 3) will invariably reflect their unique contexts and constraints. While some regions, perhaps those with foresight in policy design or early adoption of innovative technologies, might more closely approach or even redefine a lower efficient frontier (akin to the blue dashed line), others may initially face higher abatement costs due to technical limitations, informational gaps, or institutional rigidities (resembling the red dashed line) before eventually benefiting from learning effects and technological diffusion. The specific pathway a region follows is therefore shaped by a complex interplay of factors, including its existing scientific and social knowledge base, the availability and cost of abatement technologies, the efficacy of its policy design and governance structures, and crucially, its political commitment and implementation capacity.

A critical strategic consideration arising from these varying pathways relates to the timing of abatement efforts, especially for regions starting from high pollution levels. Although ambitious early-stage abatement might entail higher initial marginal costs for some, exiting high-pollution zones sooner can unlock a longer period of substantial accumulated health benefits. Consequently, from a net present value (NPV) perspective, the societal gains from proactive and early action can often outweigh those of more delayed strategies, even if the initial abatement investments appear costlier in the short term.

Beyond the long-term abatement strategies aimed at reducing annual average pollution concentrations, a distinct category of short-term measures also plays a role in mitigating immediate health risks. These can include actions such as implementing temporary emergency control measures during acute pollution episodes, individual or societal protective expenditures (e.g., on masks or air purifiers), and the operation of early warning and public advisory systems.

While such measures incur tangible societal costs and can be effective in reducing acute health impacts, their primary effect is on short-term exposure and immediate health outcomes, rather than on lowering the underlying long-term average pollution concentrations. Consequently, the costs associated with these short-term interventions are conceptually different from the MAC aimed at sustained, long-term pollution reduction, as depicted in Figure 3. A more detailed examination of the costs and benefits associated with these short-term risk mitigation and adaptation strategies will be presented in Section 3.4.2.

Our critique of the traditional static model, leads to a more constructive perspective: air pollution control is inherently a dynamic optimization problem. The central challenge is managing the pollution stock (e.g., annual average concentrations) over time. This requires choosing an optimal abatement path based on initial conditions (the current pollution level) and considering complex factors like discount rates and technological change.

While a fully integrated dynamic model (e.g., similar to the model used in Phaneuf and Requate, 2016) is theoretically comprehensive, it can be difficult to operationalize for real-world policymaking. Therefore, this section introduces the “Dual-Track BCA framework” as a pragmatic and actionable approach to structure and operationalize this dynamic challenge. This framework disaggregates the problem into two distinct, yet interconnected, tracks:

• A long-term stock management framework (Track 1): This track (discussed in Section 3.4.1) evaluates the optimal path for reducing the pollution stock (i.e., annual average concentrations) over time. It compares the MB and MAC of different long-term strategies, to guide the choice of reduction pathway.

• A short-term exposure management framework (Track 2): This track (discussed in Section 3.4.2) focuses on managing the acute health risks from short-term exposure flows (i.e., pollution episodes) that occur along that long-term path.

The key advantage of this dual-track framework is its policy flexibility, particularly in contexts with significant institutional or resource constraints. For example, if a region finds that its long-term stock management (Track 1) is progressing slowly (e.g., due to high abatement costs or institutional gridlock), the framework provides a clear rationale for strategically shifting resources to short-term exposure management (Track 2). By enhancing early warnings, protective measures, and other high-benefit short-term interventions, it is possible to harvest significant net health benefits (i.e., reduced acute illnesses) even while the long-term pollution stock remains high. This offers a pragmatic, health-centric pathway for governance in high-pollution regions.

This approach, applied through the three-phase strategy detailed in Section 4, provides a structured way to navigate the trade-offs and complementarities between long-term structural interventions and shortterm adaptive measures.

Effective long-term (chronic) air pollution management necessitates that each region accurately assesses its initial position concerning its marginal health benefit (MB, thick solid line in Figures 4 and 5) and MAC (thin solid/dotted/dashed lines in Figures 4 and 5) curves. Given that both the health burden from air pollution and the feasibility of policy options vary significantly across locations, understanding these relative positions is crucial for diagnosing the starting point and formulating an appropriate abatement strategy. Indeed, the interplay between a region’s MB and MAC curves fundamentally reflects its structural advantages or disadvantages in achieving cost-effective pollution control.

As pollution levels decline, two typical scenarios may arise:

• Regions with relatively low initial MB values (Figure 4): For these areas, early abatement efforts must prioritize low-cost interventions. Overly aggressive or high-cost measures, if implemented without careful consideration of timing and alternatives, risk incurring net societal losses.

• Densely populated regions with higher initial MB values (Figure 5): These regions, while often facing severe pollution, also stand to realize substantially larger marginal health benefits from each unit of pollution reduction. This provides a wider margin for policy action and potentially justifies more rapid and intensive interventions (conceptually, their MB curve in Figure 5 lies higher). Conversely, prolonged inaction or delayed intervention in such high-benefit settings can lead to substantial and compounding health and economic losses over time.

The strategic considerations for pollution abatement are further nuanced by a region’s current pollution level and its unique biophysical and developmental context. For instance, regions that have already achieved substantial pollution reductions but still exceed optimal healthbased targets face a different set of challenges. While the economic principle of pursuing reductions as long as marginal health benefits exceed MAC remains valid for maximizing net social welfare, the actual MAC for these further reductions can be significantly influenced by factors previously discussed, such as accumulated learning and technological capabilities, as well as by inherent regional characteristics.

From the perspective of cost structure, a region’s atmospheric environmental capacity, primarily determined by meteorological patterns and dispersion conditions, plays a crucial role. In areas with inherently poor dispersion characteristics, such as the Beijing–Tianjin–Hebei region (Cai et al., 2017; Wu et al., 2017) or the Sichuan Basin in China (Shu et al., 2021), achieving equivalent air quality improvements often demands significantly higher abatement efforts and consequently, higher costs, compared to regions with more favorable conditions.

This interplay of historical development and environmental endowment leads to distinct long-term abatement cost trajectories. In many highly urbanized countries or regions, early opportunities to optimize industrial layout in alignment with local atmospheric capacity may have been foregone, resulting in spatial path dependencies that are now economically and socially costly to reverse. By contrast, less urbanized and still-developing areas, particularly in parts of Africa and South Asia, often find their spatial configurations less fixed. These regions are still within a crucial “planning window,” where strategic decisions regarding industrial and urban location today will profoundly influence their future pollution levels, the associated health burdens, and ultimately, the long-term costs of abatement.

While the preceding discussion focused on the long-term management framework aimed at influencing annual average pollution concentrations, effective air pollution control also necessitates a framework for evaluating short-term response strategies, particularly during acute pollution episodes. Building upon the long-term context set by a region’s baseline annual average pollution level, we can conceptualize a distinct, short-term BCA space.

For any given long-term average pollution level, Figure 6 illustrates this short-term evaluation space. This framework allows for the assessment of interventions designed to mitigate immediate health risks. The vertical axis typically represents the monetized short-term health benefits achievable from temporary exposure reductions or protective actions, while the horizontal axis represents the costs of implementing these short-term measures. This analytical structure is crucial for designing and prioritizing rapid response strategies tailored to specific, often transient, pollution contexts.

Within the short-term BCA evaluation space, all policy measures share a common feature: they do not significantly alter long-term pollution levels, but may still incur financial costs and help reduce acute health risks to varying degrees.

This framework offers a practical way to assess short-term measures by plotting them in a two-dimensional space. Measures that fall above the 45-degree diagonal line yield greater health benefits than costs and can be considered desirable short-term strategies.

Based on existing literature and real-world experience, short-term measures can be broadly grouped into three categories:

• Low-cost, high-benefit strategies: These include risk communication (Barwick et al., 2024), public guidance to modify behavior, such as changing medical appointment times (Huang et al., 2023a), and proactive chronic disease management. Air pollution alerts to prompt behavioral changes (Li and Folmer, 2023; Neidell, 2009) are also a key strategy, although caution is warranted: the common advice to stay indoors is not always fully beneficial due to infiltration and indoor sources (Adamkiewicz et al., 2011; Buonocore et al., 2021; Laumbach et al., 2021).

• High-cost, high-benefit strategies: Examples include production halts, traffic restrictions (Chen et al., 2013; He et al., 2016), suspension of public events, or vehicle rationing (Viard and Fu, 2015). While expensive, such measures can produce substantial short-term health benefits under certain conditions.

• High-cost, lowor no-benefit strategies: These involve poorly designed or ineffective interventions, such as traffic control measures that fail to coordinately control NOx and VOCs (Lv et al., 2020); and truck bans that shift freight movement (Deng et al., 2020; Yang et al., 2019), leading to pollution spikes especially during night due to temperature inversions (Liu and Kim, 2019; Zhang et al., 2019b). These approaches often deliver little benefit or even exacerbate air quality problems.

While short-term BCA analysis is relevant for all regions, it is especially important in two types of settings:

1. Areas with already low annual average pollution levels, where further long-term reductions face technical or economic limits.

2. Regions with extremely high pollution levels, such as those in parts of South Asia and Southeast Asia (Health Effects Institute, 2025), represent settings in which short-term interventions may be the only viable way to mitigate health harms in the absence of immediate structural change.

This framework also offers a useful perspective for understanding adaptation in air pollution governance (Smit and Wandel, 2006). Adaptive strategies do not directly reduce pollution concentrations, but aim to strengthen people’s ability to cope with pollution in the near term. They create additional health benefits and reduce healthcare costs during the gradual process of air quality improvement.

Finally, the effectiveness and necessity of these short-term measures are often influenced by meteorological conditions, such as temperature and humidity, which can modify pollution-related health impacts, often through seasonal variation (Duan et al., 2019; Stafoggia et al., 2023). This further highlights the importance of integrating seasonality into short-term policy and adaptation planning.

Crucially, while the dual-track BCA framework separates long-term and short-term analyses for clarity, these two tracks are interconnected. Short-term interventions can influence long-term cost pathways; for example, repeated emergency controls may accelerate the learning effects for certain abatement technologies, thus altering the long-term MAC. Conversely, effective short-term investments, such as robust early warning systems, reduce acute health losses that occur during the gradual process of air quality improvement. By mitigating these immediate damages, such adaptive measures can indirectly increase the overall NPV of long-term structural policies. Therefore, a truly optimal approach requires a coordinated strategy that considers these important trade-offs and complementarities.

Building on the long-term and short-term BCA frameworks, it becomes evident that the priorities for policy intervention shift significantly across different pollution levels. Based on this insight, we propose a three-phase strategy, structured by pollution concentration, to guide air pollution control pathways that center on managing health risks.

This phase-based approach is intended to help policymakers make more informed decisions by aligning the emphasis on long-term structural measures and short-term adaptive interventions with the pollution level a region currently faces. In the sections that follow, we define each phase, outline its key characteristics, and present the corresponding strategic framework.

In this phase, pollution levels are extremely high, and the population faces widespread health risks. However, the MB of pollution reduction tends to be relatively low, meaning that each unit of abatement yields limited immediate health benefits. At the same time, MACs are often low, since the most prominent and controllable emission sources can be targeted first.

The primary objective in this phase is to initiate abatement efforts by focusing on major emission sources, recognizing that the source structure will evolve as control progresses. Conventionally, policy implementation often begins with large industrial point sources, as they are typically easier to regulate and more politically manageable (Geng et al., 2024; Zhang et al., 2019a; Zheng et al., 2018). For instance, during the initial phase of China’s air pollution control (2013–2017), characterized by very high pollution levels, strategies targeting industrial point sources were found to be the largest contributor to the reduction in populationweighted PM_2.5 concentration (Geng et al., 2024).

At the same time, a more nuanced strategic consideration involves identifying and prioritizing smaller, dispersed, but highly polluting sources. Emerging evidence suggests these sources can contribute disproportionately to the overall pollution load in this phase (Liu et al.,2016b; Tong et al., 2018). Targeting such high-intensity emitters can therefore offer a more favorable benefit–cost ratio. Research on China’s “coal-to-electricity” policy shows that while the private benefits from improved indoor air alone may not justify the costs, the policy’s full value is realized through the substantial spillover public health benefits from improved ambient air quality, which can make it more favorable than some large-scale industrial or power sector interventions (Jin et al., 2017; Zhang et al., 2019c). Similarly, within the power sector, ultra-low emission retrofits on smaller, less efficient plants have proven more costeffective than applying the same standards to larger, cleaner facilities (Jin et al., 2020).

This phase may also offer stronger political feasibility for action. When health risks become highly visible and public pressure increases, political systems are more likely to respond decisively (Greenstone et al., 2021; Jin et al., 2016). The timing of policy activation plays a major role in determining how long a society remains in a high-pollution state.

Even in severely polluted conditions, meaningful action can still be taken through short-term interventions. Drawing on the short-term BCA framework, it is possible to identify temporary measures that offer high health benefits at relatively low cost (Section 3.4.2). Adaptive strategies, in particular, provide an important channel for mitigating acute health damage when annual average concentrations cannot be quickly reduced.

These include the use of pollution forecasting and alert systems to prompt behavioral responses; encouraging protective expenditures such as wearing effective masks or improving indoor air filtration; and leveraging forecast-based management to implement temporary shutdowns or traffic restrictions when a pollution episode is expected to persist. Such actions help flatten and shorten the intensity of severe episodes and reduce short-term health damage, even if they do not affect long-term pollution levels.

This phase is characterized by pollution levels roughly around Interim Target 1 (annual mean PM_2.5 of 35 µg/m³) set in the WHO’s 2021 AQG (World Health Organization, 2021). A key shift at this stage is the following: marginal health benefits increase significantly, meaning that each additional 1 µg/m³ reduction in pollution yields a larger health gain (Xue et al., 2022). At the same time, MAC begin to diverge, with different regions or periods facing one of two cost paths.

Path 1: Accelerate toward cleaner air

When MAC remain relatively low and the society retains sufficient capacity to reallocate resources without undermining other development goals, it is preferable to advance pollution control aggressively. In this case, marginal benefits far outweigh the costs.

Path 2: Slow further abatement and focus on other priorities

If the cost of continued pollution control is already high and exceeds what the society can feasibly bear, it may be reasonable to rely on longerterm structural and economic progress to deliver future improvements in air quality.

This phase often involves difficult trade-offs between health objectives and other social goals. There is no single optimal strategy, but maintaining strong attention to health impacts remains critical regardless of the path chosen.

In this phase, where further reductions in long-term pollution levels may be difficult or slow, policy efforts should shift toward managing health outcomes more directly. Rather than focusing solely on concentration targets, the emphasis should be placed on reducing annual health losses and mitigating short-term risks. This requires the development of healthbased performance indicators and stronger systems for monitoring and managing population-level health impacts (WHO, 2025).

During this phase, the not so severe but still considerable pollution levels imply the frequent occurrence of high-exposure episodes, which often exhibit seasonal patterns. Interventions should therefore aim to lower their frequency and intensity. For vulnerable groups such as children, the elderly, and people with chronic diseases, enhancing early warning systems, behavioral guidance, and targeted support remains essential. Public awareness also forms the foundation for individual behavior changes and the organized allocation of public resources.

In this final phase, air quality has reached a relatively high level; this state conceptually resembles the “steady state” often discussed in dynamic optimization models, where pollution concentrations stabilize near WHO AQG 2021 levels. In this phase, marginal health benefits continue to rise, meaning that even small reductions in pollution may yield measurable health improvements; however, the marginal cost of abatement becomes extremely high, as further reductions demand significantly greater resources and effort.

At this stage, the transition from concentration-based control to riskoriented management has largely been completed. As adverse health effects still occur even at low concentrations, the central policy task shifts from further reduction to the mitigation of residual risks through exposure prevention, health protection, and warning mechanisms.

In this phase, assuming successful transitions, most conventional pollution sources, particularly those from fossil fuel combustion, are typically under better control or have been significantly reduced. With this sustained improvement in air quality, where cleaner air is experienced more consistently, public expectations and institutional responses tend to relax. This, in turn, increases vulnerability to unpredictable and episodic exposures. When such events occur, such as those triggered by wildfires (Burke et al., 2023; Liu et al., 2016a; Qiu et al., 2025a,b; Reid et al., 2016), dust storms (Goudie, 2014), they can lead to broader and more severe health impacts. Effective management now requires not only technical capacity but also a mindset focused on emerging threats: events that are infrequent, fast-developing, and increasingly influenced by a changing climate.

Ultimately, the dual-track and phased framework proposed here points toward the need for a dynamic, intertemporal evaluation to provide a robust economic basis for choosing between proactive and more gradualist approaches. A comprehensive NPV assessment offers a promising path forward, by systematically comparing the discounted streams of long-term costs and benefits with the aggregated net benefits from shortterm adaptive measures, which can be aggregated (e.g., annually) and discounted alongside the long-term streams. However, operationalizing such a framework presents significant methodological challenges. These include the selection of an appropriate social discount rate for long-term health outcomes and clarifying the interactions between shortand longterm cost–benefit streams. Furthermore, evaluating short-term adaptive measures introduces additional complexity, as these are often targeted at vulnerable populations. This raises difficult questions about how to value health benefits for different subgroups and how to aggregate these benefits, rather than simply summing them up. Addressing these challenges is a critical next step for developing economic tools that can robustly guide optimal, health-centric, and equitable pollution control strategies over the long run.

No society operates with a single development goal. In the process of air pollution control, objectives related to health protection, climate change mitigation and adaptation, and social equity often coexist and interact. These goals are reflected not only in overarching strategies but also in the choice of policy tools, implementation pathways, and public narratives. Therefore, when applying the longand short-term BCA frameworks proposed in this paper, these broader goals should be deliberately embedded in the analysis rather treating air pollution control in isolation.

More specifically, climate objectives frequently influence emission structures through interventions such as energy transitions and transportation reforms, creating substantial overlaps with air pollution strategies. At the same time, equity-related goals like protecting vulnerable groups and reducing health disparities affect the prioritization of interventions and the design of adaptive policies.

In a multi-objective setting, the attribution of costs and benefits becomes increasingly blurred. Many pollution control measures simultaneously contribute to health, climate, and equity objectives, making it difficult to disaggregate their impacts. Indeed, while economic theory offers methods to formally integrate distributional concerns, such as using equity weights based on the marginal utility of income, the practical application of such approaches remains challenging and often controversial. For instance, recent attempts to mandate distributional weighting in regulatory analysis in the United States (e.g., revisions to Circular A4; Fraas et al., 2025) have faced significant debate and political hurdles, highlighting the complexities beyond theoretical frameworks. This reinforces the need for goal clarity and structural transparency within the analytical framework. Explicitly listing multiple objectives can help identify potential Pareto improvements and reveal synergistic policy pathways beyond health gains alone.

When different goals align, they can reinforce one another and generate sustained policy momentum. Public health and equity objectives often attract greater societal support. As pollution levels decline and perceived urgency weakens, invoking broader narratives, such as climate justice, child protection, or resilience in aging societies, can help maintain long-term political and resource commitments to air quality improvement and prevent policy fatigue.

The goals of air pollution control and climate change mitigation are deeply intertwined, particularly through their common reliance on transitioning energy and transport sectors away from fossil fuels. This shared origin creates significant opportunities for health co-benefits; measures to reduce greenhouse gas emissions simultaneously curtail air pollutants, leading to substantial and more immediate improvements in public health (Cheng et al., 2021; Huang et al., 2023b; Peng and Ou, 2022). Recognizing the substantial health co-benefits of synergistic climate action is therefore critical for efficient policymaking. However, despite broad scientific consensus, a recent perspective by Shen et al. (2025) highlights a persistent gap: these health co-benefits are rarely incorporated effectively into real-world climate policies, often due to timescale and agent mismatches between costs and benefits, and a lack of tangible economic interpretations of the health gains.

This underscores the urgent need for analytical frameworks, such as the one developed in this paper, that can systematically address these complexities. By explicitly distinguishing between longand short-term BCA, proposing a phased approach, and considering spatial and equity dimensions, our Dual Track BCA framework aims to provide a more operational tool to help bridge this gap.

From a long-term perspective, explicitly incorporating monetized health co-benefits into the BCA can significantly alter the MAC of air pollution control, potentially making more ambitious abatement pathways economically favorable when the full spectrum of health gains from synergistic climate action is accounted for. Conversely, a failure to pursue such integrated planning risks locking into high-carbon, highpollution development paths, making future attainment of both climate and air quality goals far more costly.

From a short-term perspective, climate change primarily reshapes the BCA by amplifying the health risks associated with air pollution exposure and thus modifying the marginal health benefits of adaptive interventions. The exacerbation of acute pollution episodes by rising temperatures and altered weather patterns, which leads to events like heatwaves compounded with ozone or PM_2.5 from wildfires and dust storms (as noted previously), heightens health risks, especially for vulnerable populations. Consequently, short-term adaptive strategies, ranging from early warning systems to enhancing indoor air quality and targeted medical support, become even more critical. The BCA of these measures must therefore consider their efficacy in mitigating these climate-amplified, compounded health risks, highlighting the need to integrate climate adaptation and air quality management in short-term planning.

Ultimately, optimizing policy in the face of interconnected air pollution and climate challenges necessitates a unified analytical framework that monetizes and integrates the health impacts of both pollution control and climate actions, including mitigation co-benefits and adaptation benefits. Developing a “climate-air-health co-benefit/cost matrix” could be a practical step, translating diverse emission reductions into comparable, monetized health outcomes. This typically involves applying established valuation methods like the VSL for health impacts (Section 3.2.1) and the Social Cost of Carbon (SCC) (Revesz et al., 2017) for climate damages, thereby enabling consistent policy assessment. It is critical to clarify, however, that standard SCC estimates face significant limitations in capturing the full spectrum of health impacts. They typically omit the significant, near-term air quality co-benefits of mitigation actions (Markandya et al., 2018) and may also underquantify the full extent of direct health damages from climate change itself, such as heat stress (Carleton et al., 2022). This limitation underscores the need for the integrated “climate-air-health co-benefit/cost matrix” proposed here.

While this integrated monetization faces uncertainties (e.g., in exposure–response functions, valuation, and climate projections), these can be increasingly managed. For instance, approaches like probabilistic modeling and sensitivity analysis are commonly employed to quantify such uncertainties in environmental assessments. The application of these robust methods (see, for example, work by Wang et al. (2024a) and Xiao et al. (2025) in related contexts) can foster more confident decision-making for identifying synergies, managing trade-offs, and aligning pollution control with health-centric sustainable development goals.

Beyond optimizing for overall efficiency, air quality management must also address critical questions of distributive justice. While the environmental justice literature is vast, this section specifically applies its core economic principles, particularly the causal mechanisms of siting, sorting, and institutional failures as reviewed by Banzhaf et al. (2019b), to the unique context of air pollution and its health impacts. We argue that the distinct physical properties of air pollution (e.g., its transboundary nature) and the heterogeneous nature of its health effects (e.g., disproportionate impacts on vulnerable groups) create specific challenges and opportunities for achieving environmental equity that require tailored analysis.

The concept of environmental justice, broadly defined as the fair treatment and meaningful involvement of all people in environmental decisionmaking, has become central to understanding the societal dimensions of air pollution. A vast body of research, particularly in the United States, has consistently documented that low-income populations and people of color are disproportionately exposed to air pollutants such as PM_2.5 and NO₂ (see, e.g., Clark et al., 2017; Collins and Grineski, 2022; Spiller et al., 2021; Tessum et al., 2021). This finding of unequal exposure is one of the most robust and recurrent conclusions in the environmental justice literature.

However, identifying these correlations is insufficient for designing effective remedies. More recent economic analyses, therefore, have shifted focus from simply documenting unequal outcomes to investigating their underlying causal mechanisms, such as discriminatory siting, residential sorting, and governance failures (Banzhaf et al., 2019a). Economic analysis seeks to identify the causal mechanisms that generate and sustain these disparities in air pollution exposure and health outcomes.

One long-standing explanation is the discriminatory siting of polluting facilities, which may be disproportionately located in communities with less political and economic power. This challenge is amplified in the context of air pollution, as sources like major roadways and industrial clusters create diffuse, widespread pollution plumes that require sophisticated atmospheric dispersion modeling, rather than simple proximity analysis, to assess their health impacts (Ash and Fetter, 2004; Holland et al., 2016; Tessum et al., 2017).

An alternative mechanism is residential sorting. This pathway suggests that even if initial facility siting is neutral, disparities can arise as households “vote with their feet” (Tiebout, 1956). Higherincome households are more willing and able to pay a premium for cleaner air, while lower-income households may be constrained to live in more polluted, lower-cost areas (Chen et al., 2022b). This sorting process can exacerbate health inequities by concentrating vulnerable groups in high-pollution zones and can even lead to “green gentrification,” where environmental improvements raise housing costs and displace the original residents, preventing them from reaping the health benefits.

Furthermore, disparities can stem from institutional failures. For example, market-based instruments designed for cost-efficiency can inadvertently worsen health inequities if they do not account for spatial differences in health damages. Seminal analyses of both the U.S. Acid Rain Program and more recent air pollution “offset” markets have shown that “one-for-one” trading can allow pollution to shift towards more densely populated areas, leading to increased aggregate health damages or situations where the marginal health benefits of abatement far exceed the market-driven abatement costs (Chan et al., 2018; Shapiro and Walker, 2020). An additional dimension of institutional failure is the government’s control over public information regarding pollution risks. This can lead to a situation where households relying on official media are less likely to take protective actions, thereby undermining public health and weakening the demand for more stringent pollution control policies (Ravetti et al., 2019).

Finally, air pollution exposure can create a vicious intergenerational cycle. Early-life exposure to air pollution is linked to adverse birth outcomes (He et al., 2022b; Šrám et al., 2005) and poorer long-term health (Yuan et al., 2023), which can in turn limit educational attainment and future earning potential (Almond and Currie, 2011; Voorheis, 2017). This can trap families in a cycle of poverty and high pollution exposure, as health-related disadvantages in one generation constrain the next generation’s ability to move to cleaner environments.

A common issue in air pollution policy debates is the conflation of inequality (unequal outcomes) with injustice (unfair processes or a violation of rights). From a purely economic perspective based on residential sorting, observed inequalities in air pollution exposure may not necessarily signify injustice. The theory posits that lower-income households, facing greater budget constraints, may rationally choose to live in more air polluted areas with lower housing costs, effectively trading environmental quality for other necessities. In this view, the resulting exposure disparity reflects differences in income and preferences, rather than a failure of policy or institutions (Banzhaf et al., 2019b).

However, relying solely on this “revealed preference” argument for air pollution policy is problematic for several reasons. First, it assumes perfect information and rational choice, a strong assumption when dealing with the complex, often invisible, and long-latency health risks of air pollution. It is questionable whether households, particularly those with lower educational attainment, can fully comprehend and rationally trade off the risk of their children suffering long-term cognitive development issues for a small reduction in housing costs.

Second, this perspective often downplays the role of constrained choice and structural barriers. Factors like housing market discrimination, lack of transportation options, and limited mobility can severely restrict the “choices” available to disadvantaged groups, making their exposure in polluted environment less a matter of preference and more a consequence of systemic constraints (Lin et al., 2024).

Therefore, while unequal outcomes do not automatically prove injustice, they serve as a critical signal that warrants a deeper investigation into the underlying causal mechanisms. Effective and equitable policy must move beyond simplistic assumptions of choice and instead focus on dismantling institutional barriers, mitigating unintended consequences, and addressing the root causes of vulnerability.

The preceding analysis reveals that navigating the distinction between environmental inequality and injustice is a central challenge for policy. While not all exposure disparities warrant intervention, an effective and equitable approach to air quality management should be guided by a set of core principles that address the root causes of vulnerability and systemic disadvantage.

For long-term exposure, priority should be given to reducing emissions from fixed pollution sources located near disadvantaged populations, such as residential areas adjacent to industrial zones or major transport corridors. Urban planning should incorporate housing– environment equity principles, to prevent heavily polluted areas from becoming low-cost residential clusters. Housing rights protection and rent stabilization measures can also help prevent displacement following environmental improvements. Breaking intergenerational cycles of exposure requires early interventions, such as targeted health programs for pregnant women and young children.

For short-term exposure, it is essential to improve pollution warning systems and risk communication tools, particularly for populations with limited literacy. Vulnerable groups such as the elderly, pregnant women, and individuals with chronic health conditions should receive dedicated protection resources and emergency health services. Public infrastructure should be designed to ensure equitable access for disadvantaged groups to green spaces, indoor air filtration, and cooling centers, especially during pollution episodes. In responding to pollution–climate compound events (such as heatwaves combined with ozone spikes or dust storms with PM_2.5), governments should establish rapid-response public health coordination systems that specifically prioritize at-risk populations.

The three-phase strategy and dual-track BCA framework proposed in this paper emphasize the context-dependent relationship between marginal health benefits and MAC. Future research may further advance this framework by empirically identifying and modeling the MB and MAC curves across a range of pollution levels, demographic compositions, and policy pathways. This includes exploring structural estimation methods to capture the nonlinear exposure–response relationships between pollution concentration and diverse health outcomes. A significant frontier for improving MB estimation is the comprehensive valuation of morbidity impacts. Unlike the well-established monetization of mortality risks, the substantial welfare losses from non-fatal illnesses, such as chronic diseases, impaired cognitive function, and lost productivity, are often poorly quantified in BCA. Developing rigorous methods to monetize these diverse morbidity outcomes, despite significant data and methodological challenges, thus remains a critical area for future research.

On the cost side, there is a need to construct abatement cost data at the urban or regional scale, reflecting realistic policy portfolios and implementation sequences. In early developing countries and high-pollution regions, external research capacity may be needed to approximate the actual boundaries of cost–benefit performance, particularly during the early stages of pollution control. This research is essential for establishing direct policy tools such as regional-scale MB–MAC databases to guide the context-specific, long-term stock management framework proposed in this paper.

As air quality improves and the marginal costs of conventional control strategies rise, redirecting pollution control towards health risk management becomes increasingly necessary. Future studies should examine how health-based performance metrics can be embedded into the policy goal-setting process. This also requires rethinking how monitoring and response systems are designed, which involves moving beyond average concentration levels to account for short-term pollution episodes, seasonal variation, and time-sensitive health impacts. Another important area is the evaluation of interventions such as early warning systems, public information campaigns, and adaptive investments. These measures are often overlooked in regulatory analysis but may provide high returns in terms of health protection, especially under the short-term BCA framework. Advancing this line of inquiry will support a shift from managing ambient concentrations to managing exposure and its health consequences. This line of inquiry directly supports a key policy recommendation: the development of practical short-term risk management toolkits that bundle effective warning systems with evidence-based, low-cost protective options for vulnerable populations.

The paper highlights that the effectiveness of air pollution control is not only shaped by the scientific understanding of pollution and health but also by whether the design of policies aligns with the evolving structure of marginal benefits and marginal costs. In practice, however, the formulation of pollution control policy is still largely led by environmental and public health fields, while the influence of economic reasoning remains limited. Future work should explore how to establish effective interdisciplinary communication channels, enabling economic analysis — through marginal logic, resource allocation principles, and institutional design — to connect more directly with the empirical insights and operational needs of the environmental health community. Strengthening collaboration across disciplines in identifying core problems, interpreting evidence, and selecting instruments will be essential for making economic theory more relevant and actionable in the design and implementation of air pollution policy. A crucial outcome of such strengthened collaboration should be the systematic assessment of joint climate and health co-benefits, alongside rigorous analysis of equity impacts, in all major abatement decisions.

Yana Jin acknowledges funding from Jing-Jin-Ji Regional Integrated Environmental Improvement-National Science and Technology Major Project (2025ZD1208501), National Natural Science Foundation of China (72304015), and The Fundamental Research Funds for the Central Universities, Peking University. Shiqiu Zhang acknowledges funding from Energy Foundation (grant no. G-2407-35641), and funding from the National Social Science Fund of China (grant no. 21AZD060).