The Global Health Cost of Ambient PM2.5 Air Pollution 1 The Global Health Cost of Ambient PM₂.₅ Air Pollution © 2020 International Bank for Reconstruction and Development / The World Bank The Global Health Cost of 1818 H Street NW Washington DC 20433 202-473-1000 Ambient PM₂.₅ Air Pollution www.worldbank.org This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or con- clusions set forth. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved. Rights and Permissions The material in this work is subject to copyright. Because The World Bank encourages dissemina- tion of its knowledge, this work may be reproduced, in whole or in part, for noncommercial pur- poses as long as full attribution to this work is given. Any queries on rights and licenses, including subsidiary rights, should be addressed to World Bank Publications, The World Bank Group, 1818 H Street NW, Washington, DC 20433, USA; fax: 202-522-2625; e-mail: pubrights@worldbank.org. Cover photo Tatiana Grozetskaya / Shutterstock Sample credit for a non-WB image © Ami Vitale / Panos Pictures. Used with the permission of Ami Vitale / Panos Pictures. Further permission required for reuse. Sample credit for a WB image © Curt Carnemark / World Bank. Further permission required for reuse. Report design Spaeth Hill Abbreviations AAP Ambient Air Pollution ALRI Acute Lower Respiratory Infection COPD Chronic Obstructive Pulmonary Disease EAP East Asia and Pacific ECA Europe and Central Asia GBD Global Burden of Disease GDP Gross Domestic Product HAP Household Air Pollution HI High-Income IER Integrated Exposure-Response IHD Ischemic Heart Disease IHME Institute for Health Metrics and Evaluation LAC Latin America and the Caribbean LI Low-Income LMICs Lower-Middle-Income Countries µg/m3 Micrograms Per Cubic Foot MNA Middle East and North Africa NA North America OECD Organisation for Economic Co-Operation and Development PM Particulate Matter—A Mixture Of Solid Particles and Liquid Droplets Found In the Air PM2.5 Particulate Matter Equal to or Less than 2.5 Microns in Diameter PM10 Particulate Matter Equal to or Less Than 10 Microns in Diameter PPP Purchasing Power Parity SA South Asia SSA Sub-Saharan Africa UMI Upper-Middle-Income VSL Value of Statistical Life WHO World Health Organization WTP Willingness to Pay YLD Years Lived with Disability Table of Contents ACKNOWLEDGMENTS 13 EXECUTIVE SUMMARY 15 INTRODUCTION AND OBJECTIVES 25 CONTEXT AND VALUE-ADDED OF THIS REPORT 29 EVOLUTION OF ESTIMATES OF POPULATION EXPOSURE TO AMBIENT PM2.5 33 CURRENT AMBIENT PM2.5 POPULATION EXPOSURE 39 RISKS OF HEALTH DAMAGES FROM AMBIENT PM2.5 EXPOSURE 43 GLOBAL HEALTH DAMAGES OF AMBIENT PM2.5 EXPOSURE 47 GLOBAL COST OF AMBIENT PM2.5 EXPOSURE 57 CONCLUSIONS 61 APPENDIXES 65 REFERENCES 94 Figures & Tables FIGURE 1 FIGURE 10 Annual Cost of Health Damage from Ambient PM2.5 Exposure, % Equivalent Global Number of Deaths from Ambient PM2.5 Air Pollution in 2016 48 of GDP in 2016 by Region 18 FIGURE 11 FIGURE 2 Share of Global Deaths from Ambient PM2.5 Air Pollution in 2016 48 Annual Cost of Health Damage from Ambient PM2.5 Exposure, % Equivalent of GDP in 2016 by Income Group 19 FIGURE 12 Global Deaths from Ambient PM2.5 Air Pollution as a Share of All Global FIGURE 3 Deaths in 2016 49 Ground-Level Monitoring Stations in Absolute Numbers by Country Income Level 34 FIGURE 13 Global Number of Deaths from Ambient PM2.5 Exposure by Region FIGURE 4 in 2016 49 Ground-Level Monitoring Stations in Absolute Numbers by Region 35 FIGURE 14 FIGURE 5 Global Number of Deaths from Ambient PM2.5 Exposure by Income Group Million People per Ground-Level Monitoring Station by Country in 2016 51 Income Level 35 FIGURE 15 FIGURE 6 Number of Deaths from Ambient PM2.5 Exposure per 100,000 Population Million People per Ground-Level Monitoring Station by Region 36 in 2016 by Region 51 FIGURE 7 FIGURE 16 Regional Population-Weighted Ambient PM2.5 Exposure in 2016 39 Number of Deaths from Ambient PM2.5 Exposure per 100,000 Population in 2016 by Income Group 52 FIGURE 8 Population-Weighted Ambient PM2.5 Exposure by Country Income FIGURE 17 Group in 2016 40 Deaths from Ambient PM2.5 Exposure as a Share of All Deaths in 2016 by Region 54 FIGURE 9 Relative Risks of Major Health Outcomes Associated with PM2.5 Exposure, GBD 2016 Study 44 Figures & Tables (cont.) FIGURE 18 TABLE 1 Deaths from Ambient PM2.5 Exposure as a Share of All Deaths in 2016 Global Welfare Cost of Air Pollution, $, Trillions, per Year 26 by Income Group 54 TABLE 2 FIGURE 19 Number of Deaths from Ambient PM2.5 by Region and Country in 2016 50 Annual Cost of Health Damage from Ambient PM2.5 Exposure, % Equivalent of GDP in 2016 by Region 58 TABLE 3 Number of Deaths from Ambient PM2.5 per 100,000 Population in FIGURE 20 2016 by Country 53 Annual Cost of Health Damage from Ambient PM2.5 Exposure, % Equivalent of GDP in 2016 by Income Group 58 TABLE 4 Number of Deaths from Ambient PM2.5 per 100,000 Population in 2016 by Country 55 TABLE 5 Annual Cost of Health Damages from Ambient PM2.5 by Country, % Equivalent of GDP in 2016 59 TABLE 6 Annual Health Damages and Costs of Ambient PM2.5, 2016 65 TABLE 7 Years of Life Lived with Disability (YLD) from Ambient PM2.5 in 2016 81 World Bank Group 12 The Global Health Cost of Ambient PM2.5 Air Pollution 13 Acknowledgments This report was prepared by a team led by Yewande Awe with the core team com- prising Bjorn Larsen, Shafick Hoossein, and Ernesto Sánchez-Triana. The background document for this report was prepared by Bjorn Larsen. The team would like to acknowledge, with thanks, the valuable advice and inputs of the peer reviewers: Stephen Dorey, Fernando Loayza, Jostein Nygard, Helena Naber, Katelijn Van den Berg, and Martin Heger. This report also benefitted from comments provided by the following colleagues: Marcelo Bortman, Carter Brandon, Urvashi Narain, Lek Kadeli, Tamer Rabie, Momoe Kanada, Maria Sarraf, and Hocine Chalal. This report is a product of the Environment, Natural Resources and Blue Economy Global Practice of the World Bank. This work was conducted under the supervision of Juergen Voegele (Vice President, GGSVP); Karin Kemper (Global Director, SENDR); Julia Bucknall (Global Director, SESD2); Benoit Bosquet (Regional Director, SEADR); Benoit Blarel (Lead Environment Specialist, SENDR); Iain Shuker (Acting Practice Manager, SENGL); and Christian Albert Peter (Practice Manager, SENGL). The financial support provided by the Pollution Management and Environmental Health multi-donor trust fund of the World Bank, for the preparation of this report, is gratefully acknowledged. Executive Summary 14 The Global Health Cost of Ambient PM2.5 Air Pollution 15 Executive Summary Air pollution is a major cause of death and disease. Ambient air pollution refers to the contamination of outdoor air; household air pollution refers to the contamination of indoor air. Ambient (or outdoor) air pollution is the world’s leading environmental risk to health and the cause of morbidity and mortality from diseases such as ischemic heart disease (IHD), lung cancer, chronic obstructive pulmonary disease (COPD), stroke, and pneumonia. The majority of deaths related to air pollution are caused by human exposure to fine inhalable particles or fine particulate matter, also known as PM2.5. “Particulate mat- ter” is a mixture of solid particles and liquid droplets found in the air, and “fine particulate matter (PM2.5)” is particulate matter equal to or less than 2.5 microns in diameter. Many people in developing countries live with ambient concentrations of PM2.5 that are multiple times higher than the health-based guideline values for ambient air quality established by the World Health Organization (WHO). About 90 percent of deaths related to air pollution occur in lower-middle-income countries (LMICs) where outdoor air pollution is driven by rapid urbanization, increased motorization and energy use, and the burning of wastes and solid fuels. An estimated 4.1 million people worldwide died prematurely in 2016 because of exposure to ambient PM2.5. About 90 percent of those deaths occurred in LMICs (GBD 2016 Risk Fac- tors Collaborators 2017). The Global Burden of Disease (GBD) studies referred to in this report will be cited as “GBD” followed by the year associated with that particular set of GBD studies. Two-thirds of those deaths occurred in East Asia and Pacific and South Asia. China and India accounted for 52 percent of global deaths from ambient PM2.5. There were 11 countries with 50,000 or more deaths from ambient PM2.5 and five countries with more than 100,000 deaths. Besides being a health problem, ambient air pollution contributes to less-livable condi- tions in cities and hinders economic competitiveness. Poor people are more likely to live in a polluted environment and suffer the adverse impacts of air pollution. In addition, people who are sick as a result of exposure to air pollution are more likely to take days off work and suffer reduced productivity, which in turn undermines their contributions to econom- ic growth. Air pollution could also hinder cities’ ability to attract talented workers, thereby reducing competitiveness. Furthermore, air pollution imposes a heavy economic burden both on the economies of individual LMICs and on the global economy as a result of pre- mature death, illness, lost earnings, and increased health care expenditures—all of which constrain productivity and economic growth. Poor people who have the least means to address the health damage of air pollution often disproportionately carry the economic burden. Executive Summary 16 The Global Health Cost of Ambient PM2.5 Air Pollution 17 Air pollution is also associated with many detrimental but less researched health im- This report estimates the cost of health damages using the estimates of mortality and pacts and conditions (Sánchez-Triana et al. 2015), such as infant mortality (Heft-Neal et morbidity from ambient PM2.5 published in the GBD 2016 study. The GBD assesses mor- al. 2018), low birth weight (Ezziane 2013), preterm delivery, diabetes (Bowe et al. 2018), tality and disability from numerous diseases, injuries, and risk factors, including ambient mental health (Shin et al. 2018), and neurological impairment (Xu et al. 2016; Zhang et al. air pollution. Air pollution has long been recognized as a significant environmental health 2018) including dementia in later life (Carey et al. 2018). As the evidence base for these risk. GBD estimates of the global, regional, and national health burden attributable to air and other conditions becomes stronger, it is envisaged that exposure-response functions pollution, based on nationwide exposures to ambient PM2.5, were published for the first can be developed to obtain global estimates of the health burden of air pollution. time in the GBD 2010 study then followed by the GBD 2013 and annual publications since the GBD 2015. Some air pollutants, notably short-lived climate pollutants such as black carbon, have climate-warming properties (Shindell et al. 2012; World Bank 2020a). In addition, air pol- lution (particularly linked to sulfur dioxide) adversely affects the environment, resulting in Methodology acid rain and associated land and water pollution. Air pollution also has aesthetic impacts such as reduced visibility. However, economic valuation of these impacts can be done This report uses the GBD 2016 estimates of premature mortality and morbidity attribut- only at local and regional levels. Further research is needed to determine how to effec- able to ambient PM2.5 air pollution to value the economic cost in dollar terms. The GBD tively conduct an economic valuation of these impacts at the global level. estimates the major health damages of population exposure to ambient PM2.5 from expo- sure-response relationships that have been established by global research on air pollution Air pollution’s various adverse impacts on multiple facets of the society and economy, and health. These exposure-response relationships provide estimates of the number of particularly of LMICs, squarely place air pollution as a core development challenge. This cases in a country of premature deaths and disease that result from the population’s ex- makes reducing air pollution in developing countries central to achieving poverty reduc- posure to given ambient concentrations of PM2.5. Population exposure levels are estimat- tion and equitable prosperity objectives in those countries. ed based on a combination of ground-level monitoring of ambient PM2.5, satellite imagery, and chemical transport models. Global health crises further highlight the need for continued action in addressing a global and cross-cutting challenge such as air pollution. The current global COVID-19 pandemic, The cost of the health damages from ambient PM2.5 is quantified separately for prema- caused by the novel coronavirus SARS-CoV-2, underscores the importance of reducing ture deaths and morbidity. The cost of premature deaths is estimated from the value of air pollution through preventive and abatement measures. People who contract COVID-19 statistical life (VSL). VSL is a measure of how much individuals are willing to pay for a and have underlying medical problems such as heart disease, lung disease, and cancer reduction in the risk or likelihood of premature death. VSL is influenced by income level are at a higher risk of developing serious illnesses that could lead to death. It is note- and other factors; it is unique for each country. The cost of morbidity is estimated based worthy that air pollution is a cause of the aforementioned diseases. Ongoing research on years lived with disability (YLD) as estimated by the GBD. YLD is a measure of disease is finding relationships between air pollution and the incidence of illness and death due burden that reflects the duration and severity of diseases. YLD from exposure to ambient to COVID-19. Such research suggests that PM2.5 air pollution plays an important role in PM2.5 is converted to days lived with disease, with the cost of a day of disease equated to increased COVID-19 incidence and death rates. One such study reported that PM2.5 is a the average daily wage rate in each country. highly significant predictor of the number of confirmed cases of COVID-19 and related hospital admissions (Andrée 2020). This report recognizes that PM2.5 comes from both natural (for example, dust) and anthropogenic (for example, vehicle exhaust and emissions from power generation) This report provides an estimate of the global, regional, and national costs of health origins to varying extents. The epidemiologic literature indicates that short- and long- damage—that is, premature mortality and morbidity—from exposure to ambient PM2.5 term exposures to dust have significant health impacts and provides a reasonable basis air pollution in 2016. While recognizing the various costs of air pollution to society, this to assume that the health risk per microgram of natural dust is generally similar to that of report focuses on the cost of premature mortality and morbidity due to ambient air pol- other constituents of PM2.5, with the exception of sulfates and elemental carbon (World lution, the world’s leading environmental health risk. Estimating the health damage of Bank 2020b). Epidemiologic evidence supports the inclusion of the effects of natural dust air pollution in monetary terms provides a suitable metric for policy makers and deci- on mortality and morbidity in the quantification of health impacts of ambient air pollution. sion-makers in developing countries to prioritize the design and implementation of pol- Furthermore, while global studies of the health impacts of PM2.5 have been based on icies and interventions for controlling ambient air pollution amidst competing develop- particle mass, the epidemiologic evidence shows that adverse health damages of PM2.5 ment challenges and budgetary and other resource constraints. An earlier study by the vary according to PM2.5 source and composition. Specifically, trace constituents from World Bank and the IHME (Institute for Health Metrics and Evaluation) (2016) estimated PM2.5 and PM2.5 mass from fossil-fuel combustion are among the greatest contributors to the combined cost of premature mortality from ambient air pollution and household air PM2.5 toxicity (World Bank 2020c). The estimation of health impacts of natural dust, PM2.5 pollution in 2013.1 constituents, and PM2.5 mass from different sources, at a global level, will require strength-  otal air pollution damages in World Bank and IHME (2016) included ambient PM2.5, household PM2.5, and 1 T ambient ozone. Executive Summary 18 The Global Health Cost of Ambient PM2.5 Air Pollution 19 ening the measurement of PM2.5 constituents and source markers and improving the Annual Cost of Health Damage from Ambient PM2.5 Exposure, % Equivalent of GDP in FIGURE 2  understanding of exposure-response relationships. In this report, the valuation of health 2016 by Income Group damage from PM2.5 is based on PM2.5 mass and is not disaggregated by PM2.5 source or constituent (World Bank 2020b). 7% 6% Key Findings % equivalent of GDP 5% The global health cost of mortality and morbidity caused by exposure to ambi- •  4% ent PM2.5 air pollution in 2016 was $5.7 trillion, equivalent to 4.8 percent of global 3% 6.0% gross domestic product (GDP).2 By region, the cost ranged from an equivalent of 5.5% 2.4 percent of GDP in Latin America and the Caribbean, to 5.7 percent in East Asia 2% 3.3% 3.6% and Pacific, and 7.3 percent in South Asia (figure 1). The cost was equivalent to 2.7 2.7% percent of GDP in low-income countries and rose to 6.0 percent in upper-middle-in- 1% come countries (figure 2). The cost was equivalent to 7.5–8 percent of GDP in China 0% and India. LI LMI UMI HI non-OECD HI OECD Of the estimated total global health cost of ambient PM2.5, about 87 percent is due to •  premature mortality and 13 percent to morbidity. Note: LI = low-income countries; LMI = lower-middle-income countries; UMI = upper-middle-income countries; and HI = high-income countries. Assignment of countries to categories based on World Bank income classifications.  nnual Cost of Health Damage from Ambient PM2.5 Exposure, % Equivalent of GDP in FIGURE 1 A 2016 by Region In real terms, the estimated global cost of ambient PM2.5 air pollution in 2016 is 50 •  percent higher than the estimate for 2013 in World Bank and IHME (2016).3 The 8.0% higher cost estimate in this report is related to two key sets of factors: 7.0% 0.7% mproved methodology and availability of data. Specifically, this report uses —I 1.9% % equivalent of GDP 6.0% updated exposure-response functions from the GBD 2016, which quantitative- 0.5% ly relate the ambient levels of PM2.5 to the risk of health damage (for example, 5.0% 0.4% COPD, IHD, ALRI, lung cancer, and stroke). At all exposure levels, the total health 4.0% 0.6% damages of PM2.5 exposure were larger according to the exposure-response 0.4% 0.2% functions used in the GBD 2016 study than according to the exposure-response 3.0% 5.4% 5.0% functions used in the GBD 2013 study.4 Furthermore, ground-level measure- 2.0% 4.0% 3.2% ments utilized by the GBD 2016 study came from an expanded WHO Global 2.6% 2.7% 1.0% 2.2% Ambient Air Quality Database released in 2016 that included data from around 0.0% 6,000 ground monitors in about 3,000 human settlements. SA EAP ECA MNA SSA LAC NA —I nclusion of an estimate of the cost of morbidity, which was not provided in World Bank and IHME (2016). Mortality Morbidity Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa.  lobal health cost and GDP are stated in purchasing power parity (PPP) adjusted US$. GDP in PPP adjusted US$ allows 2 G  orld Bank and IHME (2016) provided an estimate of combined cost of $5.11 trillion for ambient air pollution 3 W for a comparison of the purchasing power of GDP of different countries. The global health cost is expressed as a per- and household air pollution in 2013. The cost of ambient air pollution alone in 2013 was $3.55 trillion (in 2011 centage of GDP only to provide a convenient sense of relative scale. $). The cost in 2016 was $5.7 trillion, or $5.31 trillion in 2011 $. The estimated cost in 2016 is therefore 50 per- cent higher than in 2013. 4  The GBD 2013 study is listed in the Reference section of this report as GBD 2013 Risk Factor Collaborators 2015. Executive Summary 20 The Global Health Cost of Ambient PM2.5 Air Pollution 21 •  Observations about the reasons for variations between GBD mortality estimates for Ensure public access to information on air quality – To reinforce the impact of air •  different years were noted in Ostro et al. (2018), which examined estimates of air quality monitoring networks, air quality management efforts should include a ro- pollution–related mortality provided in GBD 2010, GBD 2013, and GBD 2015. Method- bust system for public dissemination of air quality data in formats that are widely ological and technological improvements and demographic changes were found to understood and easily accessible to members of the public. Public dissemination of account for the observed variations in the mortality estimates. Ostro et al. (2018) also air quality data allows members of the public to take adequate measures to reduce noted the need to strengthen ground-level air quality monitoring and epidemiological their exposure to air pollution and thus provides an important social safety net for the studies to improve estimates of PM2.5 exposure and air pollution-related mortality public, particularly vulnerable groups such as young children, the elderly, and people in LMICs. with health conditions that can be exacerbated by poor air quality. Although the global availability of exposure data in GBD 2016 increased because •  Adopt regional approaches to address air pollution across boundaries – Air pol- •  of increased ground-level monitoring data, there remains a great need to increase lution typically cuts across boundaries of individual cities or countries. As a result, ground-level air quality measurements in LMICs to reduce uncertainties to PM2.5 regional airshed approaches to addressing PM2.5 air pollution may be called for. Such exposure estimates in countries that have limited or no ground-level measurements, approaches require governments to collaborate at the national and international lev- particularly of PM2.5 which is particulate matter equal to or less than 2.5 microns in di- els across multiple administrative jurisdictions and geographical boundaries to ensure ameter. PM10 refers to particulate matter equal to or less than 10 microns in diameter. effective air quality management. It was found that there was only one PM2.5 or PM10 ground-level monitor per 54 million people in low-income countries and one monitor per 16 million people in Sub-Saharan Prioritize key sources of PM2.5 air pollution, notably fossil-fuel combustion, such •  Africa, in contrast to one monitor per 300,000 people in high-income countries.5 as sulfur-emitting coal-fired power plants and diesel-fueled traffic – Air pollution control efforts that prioritize fossil-fuel combustion sources are most likely to return greater health benefits than broad efforts that do not consider the source and com- Recommendations for Policy Action position of PM2.5. Sulfate, a chemical constituent of PM2.5 from coal burning, is one of the greatest contributors to PM2.5 toxicity and has one of the strongest associations The significant health and economic burdens of ambient PM2.5 air pollution call for urgent with cardiovascular disease among the chemical constituents of PM2.5 from fossil-fuel action from policy makers in LMICs to reduce air pollution and the resulting deaths. Some combustion. Reductions in PM2.5 emissions from fossil-fuel combustion, such as sul- key areas for action include the following: fur-emitting coal-fired power plants and diesel vehicles, can be expected to produce the most significant health benefits per unit of PM2.5 reduced. Given that these sourc- Improve ground-level air quality monitoring – Properly operated and maintained •  es are also key contributors to climate warming, air pollution efforts that target ground-level monitoring networks for air quality provide data on the severity of air these sources will also provide climate change mitigation benefits. Notably, reduc- pollution, a fundamental input for effective air quality management. Data for air qual- ing PM2.5 also means reducing black carbon, a component of PM2.5 and a short-lived ity monitoring networks are also useful for identifying the key sources that contribute climate pollutant. to ambient air pollution. Such air quality monitoring networks must be subject to rigorous quality assurance and quality control regimes to ensure that the air quality measurements generated are reliable for informing the design and implementation of interventions to reduce air pollution and protect public health. Thus, high-quality, rou- tine air quality monitoring first and foremost underpins effective air quality manage- ment programs that would also include comprehensive emission inventories; applica- tion of models to understand the transport and fate of air pollutants; assessment of “Global health crises further highlight the need for continued action in costs, health, and other benefits; and public outreach and stakeholder engagement. It is pertinent to note that beyond initial investments in air quality monitoring networks, governments need to ensure effective funding for sustained operation and mainte- nance of air quality monitoring programs in the long term. addressing a global and cross-cutting challenge such as air pollution.”  he numbers of monitors are based on the WHO Global Ambient Air Quality Database released in 2016. Since the prepa- 5 T ration of this report, the WHO has released its 2018 version of the WHO Global Ambient Air Quality Database, which was used by the GBD 2017 study. The 2018 version includes nearly 10,000 ground monitors in nearly 4,400 locations in 108 countries. This represents a substantial improvement in global coverage, although 76 percent of the increase in PM2.5 monitors was in high-income countries. Regarding PM2.5 monitors, there were 64 million people per ground monitor in low-income countries and 29 million per ground monitor in Sub-Saharan Africa, in contrast to about 370,000 people per monitor in high-income countries. These results continue to underscore the need for establishing and strengthening ground-level monitoring networks in LMICs. Executive Summary 22 The Global Health Cost of Ambient PM2.5 Air Pollution 23 Engage a wide range of instruments that are suited to effectively and efficiently •  reduce air pollution and ensure that they are enforced – To reduce air pollution, governments need to apply the instruments and approaches that are most effective for reducing air pollution. Command-and-control instruments such as the establish- ment of ambient air quality standards, emissions standards for vehicles and stationary sources, and vehicle inspection and maintenance programs are well established and applied in many countries. Additional command-and-control instruments include reg- ulations to improve fuel quality, such as by decreasing the sulfur content of fuels. Economic instruments such as air pollution charges and repurposing of fossil-fuel subsidies reduce air and climate pollutants to augment the government revenue that can be allocated to education, health care, renewable energy, and interventions to control air pollution. In addition, policies to promote the conversion of vehicles from diesel to gas or to discourage the use of nitrogen-based fertilizers, which release ammonia—a precursor of secondary PM2.5 formation—may also be used to reduce air pollution. It is important to note that effective application of the various air-quality management instruments requires that governments put in place adequate enforce- ment mechanisms that also include incentives to reduce polluting behaviors. World Bank Group 24 The Global Health Cost of Ambient PM2.5 Air Pollution 25 Introduction and Objectives The detrimental effects of ambient air pollution, notably PM2.5, on health are well known. Ambient air pollution refers to the contamination of outdoor air; household air pollution refers to the contamination of indoor air. Ambient (or outdoor) air pollution is the world’s leading environmental risk to health and the cause of morbidity and mortality from diseases such as IHD, lung cancer, COPD, stroke, and pneumonia. The majority of deaths related to ambient and household air pollution are caused by human exposure to fine inhalable particulate matter, also known as PM2.5. In 2016, about 4.1 million people world- wide died as a result of exposure to ambient air pollution. Understanding the welfare costs associated with ambient air pollution has been a topic of continued attention in several works. Several of these works have applied meth- odologies and estimates of exposure to air pollution used in the Global Burden of Disease Project. These works include Larsen (2014), OECD (2016), WHO (2016), World Bank and IHME (2016), and a broader study by the Lancet Commission on pollution and health (Landrigan et al. 2018). Four of these publications provide global estimates of the wel- fare cost of air pollution, as does a recent World Bank update of the global, regional, and national cost of PM2.5 ambient air pollution in 2015 (Larsen 2017). Multiple studies, includ- ing ones cited in this paragraph, point to the enormous global welfare cost of ambient air pollution in the trillions of dollars, equivalent in magnitude to 2.5–6 percent of global GDP, depending on the valuation of health damages (table 1). Some estimates indicate an upward trend in the global welfare cost of ambient air pollution. For example, the Organ- isation for Economic Co-operation and Development (OECD) estimates that the cost of health damages of ambient air pollution could increase to $20.5–$27.6 trillion (9–12 percent of GDP) by 2060 (OECD 2016).6 It is important to note the following two cost-related findings of these studies: (i) The global cost of ambient air pollution is substantially higher than the cost of household air pollution associated with the burning of solid fuels. (ii) However, the cost of household air pollution is still substantially higher than the cost of ambient air pollution in South Asia and Sub-Saharan Africa and nearly as high as the cost of ambient air pollution in East Asia and Pacific (World Bank and IHME 2016). 6  2010 Purchasing Power Parity (PPP) adjusted US$. Introduction and Objectives 26 The Global Health Cost of Ambient PM2.5 Air Pollution 27 TABLE 1 Global Welfare Cost of Air Pollution, $, Trillions, per Year $ % of Study Domain Year US$ (PPP) GDP 2012 in Larsen (2014) AAP - 1.7 2.5 2012 prices World Bank and 2013 in AAP 3.6 - 3.5 IHME (2016) 2011 prices 2015 in OECD (2016) AAP 3.4 - 6.0 2010 prices Landrigan AAP & 2015 in - 3.8 5.1* et al. (2018) HAP 2015 prices 2015 in Larsen (2017) AAP 5.5 3.3 4.5 2015 prices Note: $ (PPP) = international dollars or purchasing power parity adjusted US$. GDP in PPP adjusted US$ allows a comparison of the purchasing power of GDP of different countries. AAP = ambient air pollution; HAP = household air pollution. * Gross national income. This report provides an updated estimate of the global, regional, and national cost of ambient PM2.5 air pollution in 2016 using the GBD 20167 estimates of mortality and mor- bidity from ambient PM2.5. The estimated global cost in 2016 was $5.7 trillion,8 equivalent to 4.8 percent of global GDP (PPP adjusted).9 In real terms, the estimated cost of am- bient PM2.5 air pollution in 2016 is 50 percent higher than the estimate for 2013 in World Bank and IHME (2016).10 The reasons for the higher cost estimate are mainly changes in exposure-response functions, the substantially higher estimate of global ambient PM2.5 exposure, and the inclusion of an estimate of the cost of morbidity, as discussed below. The higher estimate of global ambient PM2.5 is due more to improved methodology and availability of data than actual worsening of global ambient PM2.5 air quality from 2013 to 2016, although the exact contribution of each of these two factors is difficult to ascertain. This report also provides an overview of global and regional ambient PM2.5 population exposure and the exposure-response functions developed by the GBD 2016 study.  he GBD 2016 study is listed in the References section of this report as GBD 2016 Risk Factor Collaborators 2017. 7 T nternational dollars or purchasing power parity adjusted US$. Expressed in US dollars, the global cost in 2016 was 8 I US$3.3 trillion, equivalent to 4.4 percent of global GDP.  he cost equivalent to percent of GDP is the same whether expressed in GDP or PPP-adjusted GDP for each individual 9 T country, but not when aggregated globally.  he cost of ambient PM2.5 in 2013 was $3.55 trillion (in 2011 $ (PPP)) according to World Bank and IHME 10 T (2016). The cost in 2016 was $5.7 trillion, or $5.31 trillion in 2011 $ (PPP). The estimated cost in 2016 is therefore 50 percent higher than in 2013. World Bank Group 28 The Global Health Cost of Ambient PM2.5 Air Pollution 29 Context and Value- Added of this Report This report provides an estimate of the global, regional, and national costs of health damage—that is, of premature mortality and morbidity—from exposure to ambient PM2.5 air pollution in 2016. While recognizing the various costs of air pollution to society, this report focuses on the cost of premature mortality and morbidity due to ambient air pollution, the world’s leading environmental health risk. Estimating the health dam- age of air pollution in monetary terms provides a suitable metric for policy makers and decision-makers in developing countries to prioritize the design and implementation of policies and interventions for controlling ambient air pollution amidst competing devel- opment challenges and budgetary and other resource constraints. As a development institution, the cost of ambient air pollution underscores the need for the World Bank’s sustained support of governments’ efforts to reduce ambient air pollution. Furthermore, the cost estimate provides a useful metric for informing decision-making and priority setting by governments in tackling the urgent problem of ambient air pollution. The value-added of this report is as follows: The report is based on updated exposure-response functions as used by the GBD •  2016 study. Exposure-response functions quantitatively relate the ambient levels of PM2.5 to the risk of health damage (for example, COPD, IHD, ALRI, lung cancer, and stroke). The exposure-response functions used in this report differ in import- ant aspects from the functions from the GBD 2013 study11 used in World Bank and IHME (2016). The GBD 2016 exposure-response functions reveal a much higher risk of COPD and acute lower respiratory infection (ALRI) from PM2.5 exposure than the functions used in the GBD 2013 study. The GBD 2016 exposure-response functions are somewhat higher for IHD at higher exposure levels, somewhat lower for stroke, and substantially lower for lung cancer. Lung cancer mortality is, however, a very minor share of total mortality from ambient PM2.5. Thus, in aggregate at all exposure levels, the health damages of PM2.5 exposure are larger according to the exposure-re- sponse functions used in the GBD 2016 study than according to the functions used in the GBD 2013 study. 11  The GBD 2013 study is listed in the References section of this report as GBD 2013 Risk Factor Collaborators 2015. Context and Value-Added of this Report 30 The Global Health Cost of Ambient PM2.5 Air Pollution 31 This report is based on global ambient PM2.5 exposure estimates used in the GBD •  2016. These exposure estimates are higher than the estimates used in the GBD 2013 and based on a database of ground-level measurements of air quality that are used for calibrating satellite and chemical transport modeling estimates of PM2.5. The data- base of ground-level measurements used by the GBD 2016 is substantially larger than the database used in the GBD 2013 study. Global population-weighted ambient PM2.5 exposure was 50 µg/m3 in 2016 according to the estimates used in the GBD 2016 study and 32 µg/m3 in 2013 according to the GBD 2013 study. As a result of the changes in exposure-response functions and ambient PM2.5 expo- •  sure estimates from the GBD 2013 study to the GBD 2016 study, this report is based on a global mortality estimate of 4.1 million deaths from ambient PM2.5 in 2016 com- pared to 2.9 million deaths in 2013 used by World Bank and IHME (2016). This report also provides an order-of-magnitude estimate of the cost of •  morbidity of ambient PM2.5 based on the morbidity disease burden reported by the GBD 2016 study, which is found to vary substantially across countries and regions. The remaining sections of this report provide a global and regional overview of PM2.5 ambient air quality monitoring, estimates of population exposure to PM2.5, estimation of health damages from this exposure, and global costs of these health damages. Evolution of Estimates of Population Exposure to Ambient PM2.5 32 The Global Health Cost of Ambient PM2.5 Air Pollution 33 Evolution of Estimates of Population Exposure to Ambient PM2.5 The GBD project estimates health damages from nationwide population exposure to ambient PM2.5. Nationwide exposure is estimated from a combination of satellite imagery, chemical transport modeling, and ground-level PM2.5 and PM10 measurements. The evolution in satellite imagery and chemical transport model estimation techniques, the number of ground-level monitoring locations, and the method of calibrating the satel- lite imagery and chemical transport model estimates with the ground-level measurements has been quite substantial from the GBD 2010 study to the GBD 2016 study. These issues are discussed in some detail in Brauer et al. (2012), Brauer et al. (2016), Shaddick et al. (2018), van Donkelaar et al. (2015), and van Donkelaar et al. (2016). Ground-level measurements of PM2.5 or PM10 employed by the GBD 2010 study cov- ered less than 700 locations (Brauer et al. 2012). Two-thirds of the locations were in the high-income countries of East Asia and Pacific, North America, and Western Europe. There were 222 locations in Central Europe (26), the Middle East and North Africa (9), and LMICs of East Asia and Pacific (133), Latin America and the Caribbean (25), South Asia (21), and Sub-Saharan Africa (8). The majority of the locations in East Asia and Pacific were in China. The ground-level measurements of PM2.5 and PM10 employed by the GBD 2013 study were expanded to 4,073 data points from 3,387 unique locations (Brauer et al. 2016). This included measurement data used by the GBD 2010 study and new data, especially from China and India, including data compiled from a literature survey (van Donkelaar et al. 2015) and the WHO ambient air pollution database. The GBD 2015 study12 utilized the updated and expanded WHO Ambient Air Quality Da- tabase released in 2016. This database contained PM measurements from 6,003 ground monitors in about 3,000 human settlements ranging in size from populations ranging from those in the hundreds to those over 10 million (GBD 2015 study; WHO 2016). The GBD 2016 study utilized the same data as the GBD 2015 study. 12  The GBD 2015 study is listed in the References section of this report as GBD 2015 Risk Factor Collaborators 2016. Evolution of Estimates of Population Exposure to Ambient PM2.5 34 The Global Health Cost of Ambient PM2.5 Air Pollution 35 Analysis of the 2016 online version of the WHO database used by the GBD 2015 study  round-Level Monitoring Stations in Absolute Numbers by Region FIGURE 4 G and the GBD 2016 study reveals the following: 2,500 i. O  nly 530 (9 percent) of the monitors were in low-income and lower-middle-income Number of ground-level countries, although these countries account for nearly 50 percent of the world popula- monitoring stations tion and a little over 50 percent of global deaths from ambient PM2.5 (figure 3). 2,000 ii. O  nly 62 of the monitors were in Sub-Saharan Africa, while 270–370 monitors were 1,500 in each of Middle East and North Africa, Latin America and the Caribbean, and South Asia (figure 4). 2,278 1,000 1,607  here was only one monitor per 54 million people in low-income countries in contrast iii. T 315 370 to one monitor per 300,000 people in high-income countries (figure 5). 500 889 270 62  egionally, there was only one monitor per 16 million in Sub-Saharan Africa compared iv. R - to one monitor per 400,000 in Europe and Central Asia and North America (figure 6). ECA NA EAP MNA LAC SA SSA One limitation of the ground-level measurement data is, however, that over half of the measurement points are of coarse particulate matter pollution PM10, rather than the more Source: Based on data from WHO Global Ambient Air Quality Database 2016. Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; health-damaging fine particulate matter or PM2.5. The measurements of PM10 are convert- NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. ed to PM2.5 using available, albeit imprecise, information about their ratios. This introduces additional uncertainty to the global PM2.5 exposure estimates in countries with relatively few measurement points of PM2.5, which was especially the case, at least up until 2016, in the South Asia, Middle East and North Africa, and Latin America and the Caribbean regions, as well as large parts of the Europe and Central Asia region.  illion People per Ground-Level Monitoring Station by Country Income Level FIGURE 5 M Ground-Level Monitoring Stations in Absolute Numbers by Country Income Level FIGURE 3  60 50 4,000 3,500 40 Number of ground-level (Million) monitoring stations 3,000 30 54.3 2,500 20 2,000 5.6 3,571 10 1.5 0.3 1,500 0 1,000 519 1,698 LI LMI UMI HI 500 11 - Source: Based on data from WHO Global Ambient Air Quality Database 2016. Note: LI = low-income countries; LMI = lower-middle-income countries; UMI = upper-middle-income countries; and HI = high-income countries. Assignment of countries to categories based on World Bank income classifications. LI LMI UMI HI Source: Based on WHO Global Ambient Air Quality Database 2016. Note: LI = low-income countries; LMI = lower-middle-income countries; UMI = upper-middle-income countries; and HI = high-income countries. Assignment of countries to categories based on World Bank income classifications. Evolution of Estimates of Population Exposure to Ambient PM2.5 36 The Global Health Cost of Ambient PM2.5 Air Pollution 37  illion People per Ground-Level Monitoring Station by Region FIGURE 6 M 18 16 14 12 10 (Million) 8 15.9 6 4 1.5 1.9 1.4 2 0.4 0.4 4.7 0 ECA NA EAP MNA LAC SA SSA Source: Based on data from WHO Global Ambient Air Quality Database 2016. Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. World Bank Group 38 The Global Health Cost of Ambient PM2.5 Air Pollution 39 Current Ambient PM2.5 Population Exposure Global population exposure to ambient PM2.5 was 50 µg/m3 in 2016 according to estimates used by the GBD 2016 study. In contrast, the global population exposure for 2013 was 32 µg/m3 according to estimates used by the GBD 2013 study. The difference is due more to changes in estimation methodology and increased availability of ground-level PM monitor- ing data reflected in the WHO database 2016 than to actual worsening of global ambient PM2.5 air quality from 2013 to 2016, although the exact contribution of each of these two factors is difficult to ascertain. The changes in estimation methodology and availability of ground-level PM monitoring data are explained in the supplemental material of the GBD 2016. The global population-exposure estimate for 2016 is five times as high as WHO’s Air Quality Guideline value of 10 µg/m3 for annual average PM2.5. Ambient PM2.5 exposures in 2016 were highest in Middle East and North Africa, South Asia, and Sub-Saharan Africa regions—that is, about seven to nine times as high as in North America. PM2.5 exposure is also high in East Asia and Pacific, dominated by China at 56 µg/m3 (figure 7).  egional Population-Weighted Ambient PM2.5 Exposure in 2016 FIGURE 7 R 90 80 70 60 50 µg/m³ 40 79 77 68 30 43 20 9 10 19 17 0 MNA SA SSA EAP ECA LAC NA Source: Based on the GBD 2016 study. Note: Country-level ambient PM2.5 exposure levels are reported by the World Development Indicators Database (http://datatopics.worldbank.org/ world-development-indicators/). Current Ambient PM2.5 Population Exposure 40 The Global Health Cost of Ambient PM2.5 Air Pollution 41 By country income group, PM2.5 exposure was highest in high-income non-OECD countries, dominated by the countries in the Middle East and North Africa region. Exposure was by far the lowest in high-income OECD countries, followed by upper-middle-income countries, low-income countries, and lower-middle-income countries (figure 8).  opulation-Weighted Ambient PM2.5 Exposure by Country Income Group in 2016 FIGURE 8 P 120 100 80 µg/m³ 60 116 40 67 58 42 13 20 0 LI LMI UMI HI non-OECD HI OECD Source: Based on the GBD 2016 study. Note: Country-level ambient PM2.5 exposure levels are reported by the World Development Indicators Database (http://datatopics.worldbank.org/ world-development-indicators/). LI = low-income countries; LMI = lower-middle-income countries; UMI = upper-middle-income countries; HI = high-income countries; OECD = Organisation for Economic Co-operation and Development member countries. Assignment of countries to cate- gories based on World Bank income classifications. World Bank Group 42 The Global Health Cost of Ambient PM2.5 Air Pollution 43 Risks of Health Damages from Ambient PM2.5 Exposure Exposure-response functions or concentration-response functions are a key input for quantifying the health burden of ambient air pollution. One such function is the integrat- ed exposure-response (IER) function, so called because it integrates exposures to PM2.5 from different sources. The GBD project estimates the health damages of PM2.5 exposure from IER functions for five major health outcomes. The GBD project first developed IER functions for the GBD 2010 study (see appendix B). These IER functions provide the rela- tive risks of health damages of PM2.5 at exposures ranging from less than 10 µg/m3 to several hundred µg/m3. The relative risks from the IER function used by the GBD 2016 study are published in the GBD 2016 study Supplement. They are reproduced in figure 9 for PM2.5 concentrations up to 150 µg/m3. Relative risks of IHD and cerebrovascular disease (stroke) are the smallest for PM2.5 concentrations larger than 30 to 50 µg/m3 and relative risks of ALRI and COPD are the largest at PM2.5 concentrations over 20 µg/m3.13 The relative risks are derived from studies of long-term exposure to ambient air PM2.5, secondhand tobacco smoke, household use of solid cooking fuels, and active tobacco use (Burnett et al. 2014). This provides risk functions that can be applied to a wide range of “Exposure-response functions or concentration-response functions are a key input for quantifying the health burden of ambient air pollution.”  elative risks for IHD and stroke are population age-weighted and vary across countries in relation to the age structure 13 R of IHD and stroke mortality. Risks of Health Damages from Ambient PM2.5 Exposure 44 The Global Health Cost of Ambient PM2.5 Air Pollution 45 ambient PM2.5 concentrations around the world as well as to high household air pollution levels of PM2.5 from the combustion of solid fuels. The risk functions are nonlinear, with declining marginal health damages at higher PM2.5 concentrations. These GBD 2016 exposure-response functions differ in important aspects from the risk functions from the GBD 2013 study used in World Bank and IHME (2016). The GBD 2016 risk functions reveal a much higher risk of COPD and ALRI from PM2.5 exposure than the functions used in the GBD 2013 study. The GBD 2016 risk functions are somewhat higher for IHD at higher exposure levels, somewhat lower for stroke, and substantially lower for lung cancer. Lung cancer mortality is, however, a very minor share of total mortality from ambient PM2.5. Thus, in aggregate at all exposure levels, the health damages of PM2.5 expo- sure are larger according to the exposure-response functions used in the GBD 2016 study than according to the functions used in the GBD 2013 study.  elative Risks of Major Health Outcomes Associated with PM2.5 Exposure, FIGURE 9 R GBD 2016 Study 2.0 1.9 1.8 1.7 Relative risk 1.6 1.5 1.4 1.3 1.2 1.1 1.0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 Ambient PM2.5 (µg/m³ ) LRI COPD LC IHD Stroke Source: Based on the GBD 2016 study Supplement. Note: COPD = chronic obstructive pulmonary disease, IHD = ischemic heart disease, LC = lung cancer; LRI = lower respiratory infections. World Bank Group 46 The Global Health Cost of Ambient PM2.5 Air Pollution 47 Global Health Damages of Ambient PM2.5 Exposure As many as 4.1 million people died from ambient PM2.5 air pollution in 2016 according to the GBD 2016 study. This makes ambient PM2.5 the seventh-largest health risk factor of global deaths among dozens of risk factors assessed by the GBD 2016 study. Globally, IHD accounts for 39 percent of deaths from ambient PM2.5, stroke for 19 percent, COPD for 19 percent, LRI for 16 percent, and lung cancer for 7 percent according to the GBD 2016 study (see figure 10 for global number of deaths associated with ambient PM2.5 and figure 11 for percent share of associated diseases). For perspective, global deaths from ambient PM2.5 air pollution constituted as much as 7.5 percent of all global deaths in 2016. For the five health outcomes of PM2.5 exposure, ambi- ent PM2.5 caused 14–17 percent of global deaths from IHD, stroke, and lung cancer and 27 percent of global deaths from COPD and LRI (figure 12). By region, two-thirds of global deaths from ambient PM2.5 exposure in 2016 occurred in South Asia and East Asia and Pacific (figure 13). India accounted for 78 percent of these deaths in South Asia and China for 77 percent in East Asia and Pacific. Deaths from ambi- ent PM2.5 in these two countries constituted 52 percent of global deaths from ambient PM2.5. Global Health Damages of Ambient PM2.5 Exposure 48 The Global Health Cost of Ambient PM2.5 Air Pollution 49  lobal Number of Deaths from Ambient PM2.5 Air Pollution in 2016 FIGURE 10 G  lobal Deaths from Ambient PM2.5 Air Pollution as a Share of All Global Deaths in 2016 FIGURE 12 G 1,800,000 1,576,105 30% 1,500,000 25% Number of deaths Share of deaths 1,200,000 20% 796,524 786,939 900,000 116 653,406 15% 27% 27% 600,000 10% 17% 16% 116 279,718 14% 300,000 5% 0 0% as ve r as ve r la la y se ti ry se ti e r cu e e r cu e r ce di c s ce di c s to to y tru ea t d ic y tru s ea t d ic s n e va n ra e va ra ar em ar em ca is ca ar bs is ar bs ns pi as ro ns pi as ro he ch ng on c o he ch ng on c o io s se b io s se b ct re ct re di ere Is di ere Is Lu lm ni Lu lm ni fe er fe er pu hro pu hro C C in ow in w Lo C C L Source: Based on data from IHME, GBD 2016 study. Source: Based on data from IHME, GBD 2016 study.  lobal Number of Deaths from Ambient PM2.5 Exposure by Region in 2016 FIGURE 13 G  hare of Global Deaths from Ambient PM2.5 Air Pollution in 2016 FIGURE 11 S 1,600,000 1,382,814 1,322,894 1,400,000 7% Ischemic heart disease 1,200,000 Number of deaths 16% Cerebrovascular disease 1,000,000 39% Chronic obstructive pulmonary disease 800,000 511,900 Lower respiratory infections 600,000 19% 387,995 400,000 Lung cancer 181,015 19% 147,263 100,335 200,000 0 SA EAP ECA MNA SSA LAC NA Source: Based on data from IHME, GBD 2016 study. Source: Based on data from IHME, GBD 2016 study. Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. Global Health Damages of Ambient PM2.5 Exposure 50 The Global Health Cost of Ambient PM2.5 Air Pollution 51 The three (and, in one case, the two) countries with the most deaths from ambient PM2.5 By income group, the highest death rates from ambient PM2.5 exposure are in lower- in each region in 2016 are presented in table 2. There are 11 countries with 50,000 or middle-income countries, and lowest in the high-income countries (figure 16). more deaths from ambient PM2.5 and five countries with more than 100,000 deaths.  lobal Number of Deaths from Ambient PM2.5 Exposure by Income Group in 2016 FIGURE 14 G TABLE 2 Number of Deaths from Ambient PM2.5 by Region and Country in 2016 2,000,000 1,829,176 Deaths Deaths Region Country Region Country 1,800,000 (thousands) (thousands) 1,545,342 1,600,000 China 1,075 United States 93 1,400,000 Number of deaths EAP NA 1,200,000 Indonesia 80 Canada 7 1,000,000 Japan 46 800,000 Russian 125 India 1,034 Federation 600,000 368,895 ECA Ukraine 54 SA Pakistan 125 400,000 263,936 Germany 37 Bangladesh 109 200,000 26,865 0 Brazil 50 Nigeria 69 LI LMI UMI HI non-OECD HI OECD LAC Mexico 24 SSA Ethiopia 37 Source: Based on data from IHME, GBD 2016 study. Argentina 16 Congo, Dem. Rep 33 Note: LI = low-income countries; LMI = lower-middle-income countries; UMI = upper-middle-income countries; and HI = high-income countries. Assignment of countries to categories based on World Bank income classifications. Egypt, Arab Rep. 68 MNA Iran, Islamic Rep. 29  umber of Deaths from Ambient PM2.5 Exposure per 100,000 Population in 2016 by Region FIGURE 15 N Iraq 18 80 Source: Based on data from IHME, GBD 2016 study. 70 Number of deaths per 100,000 Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. 60 In total, 84 percent of global deaths from ambient PM2.5 exposure occurred in middle-in- 50 come countries, and they were nearly evenly split between lower-middle-income and up- 40 75 per-middle-income countries. About 6.5 percent of deaths were in low-income countries 62 and nearly 10 percent in high-income countries (figure 14). 30 56 44 The majority of deaths from ambient PM2.5 exposure in East Asia and Pacific and South 20 38 Asia, and these regions also have the highest rates of deaths from ambient PM2.5, reach- 28 24 10 ing 62 and 75 deaths per 100,000 population, respectively. The lowest death rates are in Latin America and the Caribbean and NA (figure 15). 0 SA EAP ECA MNA SSA LAC NA Source: Based on data from IHME, GBD 2016 study. Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. Global Health Damages of Ambient PM2.5 Exposure 52 The Global Health Cost of Ambient PM2.5 Air Pollution 53  umber of Deaths from Ambient PM2.5 Exposure per 100,000 Population in 2016 FIGURE 16 N TABLE 3 Number of Deaths from Ambient PM2.5 per 100,000 Population in 2016 by Country by Income Group Region Country Death rate Region Country Death rate 70 Number of deaths per 100,000 60 Korea, Dem. 92 United States 29 People’s Rep. 50 EAP NA 40 China 78 Canada 19 62 60 30 Mongolia 51 43 20 36 34 Bulgaria 127 India 78 10 ECA Ukraine 119 SA Afghanistan 78 0 Belarus 101 Nepal 71 LI LMI UMI HI non-OECD HI OECD Central African Cuba 47 90 Source: Based on data from IHME, GBD 2016 study. Republic Note: LI = low-income countries; LMI = lower-middle-income countries; UMI = upper-middle-income countries; and HI = high-income countries. Assignment of countries to categories based on World Bank income classifications. LAC Haiti 46 SSA Niger 70 Guyana 38 Cameroon 68 The three (and, in one case, the two) countries in each region with the highest death Egypt, Arab Rep. 71 rates from ambient PM2.5 (expressed as the number of deaths per 100,000 population) are presented in table 3. The countries with the highest death rates are in Europe and MNA Iraq 49 Central Asia, East Asia and Pacific, Sub-Saharan Africa, and South Asia. There are as Djibouti 46 many as 25 countries in Europe and Central Asia with death rates over 50 per 100,000 population. They are all in Central Asia and the eastern part of Europe. The high death Source: Based on data from IHME, GBD 2016 study. rates in these parts of Europe and Central Asia are primarily associated with high baseline Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; cardiovascular death rates. NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. Deaths from ambient PM2.5 exposure exceed or approach 10 percent of all deaths in South Asia, Middle East and North Africa, and East Asia and Pacific and 8–9 percent in middle- income and high-income non-OECD countries, compared to 3–4 percent in NA and high-income OECD (figure 17 and figure 18). Global Health Damages of Ambient PM2.5 Exposure 54 The Global Health Cost of Ambient PM2.5 Air Pollution 55  eaths from Ambient PM2.5 Exposure as a Share of All Deaths in 2016 by Region FIGURE 17 D The three (and, in one case, the two) countries in each region with the highest death rates—that is, the number of deaths from ambient PM2.5 as a percentage of total deaths— are presented in table 4. The countries with the highest death rates are in Middle East and 12% North Africa, South Asia, Europe and Central Asia, and East Asia and Pacific. There are 10 countries in which deaths from ambient PM2.5 exceed 10 percent of total deaths, including Saudi Arabia in Middle East and North Africa. 10% TABLE 4 Deaths from Ambient PM2.5 Exposure as a Share of All Deaths by Country in 2016 Share of deaths 8% 6% Death Death Region Country Region Country 10.6% rate (%) rate (%) 8.8% 9.2% 4% China 11.1 United States 3.4 5.5% 5.0% 4.2% 2% 3.3% EAP Korea, Dem. NA 10.0 People’s Rep. 0% Canada 2.6 SA EAP ECA MNA SSA LAC NA Mongolia 7.2 Tajikistan 11.2 Bangladesh 12.8 Source: Based on data from IHME, GBD 2016 study. Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. ECA Uzbekistan 9.9 SA Nepal 11.3 Turkmenistan 9.7 India 10.6 Peru 5.8 Sudan 8.8  eaths from Ambient PM2.5 Exposure as a Share of All Deaths in 2016 by Income Group FIGURE 18 D LAC Honduras 5.6 SSA Mauritania 8.8 10% Cuba 5.5 Cabo Verde 8.4 9% Egypt, Arab Rep. 13.0 8% MNA Kuwait 12.7 7% Share of deaths Saudi Arabia 12.4 6% 5% 8.9% 8.6% Source: Based on data from IHME, GBD 2016 study. 8.2% 4% Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. 3% 5.4% 2% 3.7% The GBD 2016 study also estimates that ambient PM2.5 air pollution caused morbidity in the magnitude of 6 million YLD in 2016, or about 1.5 YLD per death from ambient PM2.5. 1% These YLD are equivalent to nearly 15 billion days of illness, or about 3,600 days of illness 0% per death from ambient PM2.5.14 Morbidity from ambient PM2.5 varied substantially rela- LI LMI UMI HI non-OECD HI OECD tive to mortality across countries, from 0.3–0.5 YLD per death in a dozen countries to as much as 2.0–4.4 in another 11 countries (table D.1, appendix D). YLD per death was lowest Source: Based on data from IHME, GBD 2016 study. in Latin America and the Caribbean (0.73), Sub-Saharan Africa (0.75), Europe and Central Note: LI = low-income countries; LMI = lower-middle-income countries; UMI = upper-middle-income countries; and HI = high-income countries. Asia (0.80), and Middle East and North Africa (1.07) and highest in East Asia and Pacific Classification according to World Bank income taxonomy. (1.60), North America (1.71), and South Asia (1.98). 14  YLD is duration of illness or disability in (fraction of) years multiplied by severity of illness or disability (ranging from 0 to 1). World Bank Group 56 The Global Health Cost of Ambient PM2.5 Air Pollution 57 Global Cost of Ambient PM2.5 Exposure The health damages of ambient PM2.5 exposure can be monetized to provide an estimate of the welfare cost of PM2.5. The valuation of mortality in this report follows the welfare approach or VSL in World Bank and IHME (2016) (see appendix C). Valuation of morbidi- ty, measured as the cost of days of illness, is valued at wage rates (see appendix D). The global cost of health damages from ambient PM2.5 exposure was $5.7 trillion in 2016, equivalent to 4.8 percent of global GDP (PPP adjusted).15 This estimate is 50 percent higher in real terms than the estimate for 2013 in World Bank and IHME (2016). The rea- sons for the higher cost estimate are mainly changes in exposure-response functions, the substantially higher estimate of global ambient PM2.5 exposure, and the inclusion of an estimate of the cost of morbidity. The higher estimate of global ambient PM2.5 exposure is more due to improved methodology and availability of ground-level PM monitoring data than actual worsening of global ambient PM2.5 air quality from 2013 to 2016, although the exact contribution of each of these two factors is difficult to ascertain. About 87 percent of the total global cost of health damages in 2016 is from premature mortality and 13 percent from morbidity. Cost of morbidity as a share of the total cost of health damages by country varies from as low as 3 percent to as high as 38 percent across countries (see table D.1 in appendix D). Regionally, the morbidity cost-share ranges from 7 percent in Latin America and the Caribbean to 18 percent in NA and 26 percent in South Asia. The share is 10–15 percent in the regions of Europe and Central Asia, Middle East and North Africa, East Asia and Pacific, and Sub-Saharan Africa. The overall global cost of morbidity, relative to the cost of mortality, is very similar to the estimate by the OECD in its report on the global economic consequences of outdoor air pollution (OECD 2016) (see appendix D). The total cost of health damages—that is, mortality and morbidity—from ambient PM2.5 was the highest in South Asia, at the equivalent of 7.3 percent of GDP, and in upper- middle-income countries, at the equivalent of 6 percent of GDP. The cost was the lowest in Latin America and the Caribbean and Sub-Saharan Africa and low-income countries (figure 19 and figure 20). 15  Expressed in US dollars, this equates to US$3.3 trillion in 2016, equivalent to 4.4 percent of global GDP. Global Cost of Ambient PM2.5 Exposure 58 The Global Health Cost of Ambient PM2.5 Air Pollution 59  nnual Cost of Health Damage from Ambient PM2.5 Exposure, % Equivalent of GDP FIGURE 19 A The three (and, in one case, the two) countries in each region with the highest welfare in 2016 by Region cost of ambient PM2.5 as a percentage of GDP are presented in table 5. The countries with the highest costs are in Europe and Central Asia, South Asia, East Asia and Pacific, 8.0% and Middle East and North Africa. There are 19 countries in which the welfare cost of ambient PM2.5 exceeds the equivalent of 7 percent of GDP. Seventeen of these coun- 7.0% tries are in Europe and Central Asia, and, specifically, in the eastern part of Europe. This 1.9% 0.7% % equivalent of GDP 6.0% is largely associated with the high baseline death rates in this part of Europe. Cost by 0.5% country is presented in appendix A. 5.0% 0.4% 4.0% 0.4% 0.6% TABLE 5 Annual Cost of Health Damages from Ambient PM2.5 by Country, % Equivalent of GDP in 2016 0.2% 3.0% 5.4% 5.0% 2.0% 4.0% Region Country Cost (%) Region Country Cost (%) 3.2% 2.6% 2.7% 1.0% 2.2% 0.0% China 7.6 United States 3.4 SA EAP ECA MNA SSA LAC NA EAP NA Mongolia 4.5 Canada 2.1 Myanmar 4.3 Mortality Morbidity Bulgaria 12.4 India 7.8 Note: EAP = East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. ECA Ukraine 10.4 SA Pakistan 5.8 Hungary 9.9 Nepal 5.3  nnual Cost of Health Damage from Ambient PM2.5 Exposure, % Equivalent of GDP FIGURE 20 A Cuba 4.3 Cameroon 4.8 in 2016 by Income Group LAC Trinidad and SSA Central African 3.6 4.4 Tobago Republic 7% Barbados 3.5 Chad 4.1 6% Egypt, Arab Rep. 6.4 5% % equivalent of GDP MNA Iraq 4.8 4% Tunisia 4.1 3% 6.0% 5.5% Note: East Asia and Pacific; ECA = Europe and Central Asia; LAC = Latin America and the Caribbean; MNA = Middle East and North Africa; 2% 3.3% 3.6% NA = North America; SA = South Asia; SSA = Sub-Saharan Africa. 2.7% 1% The cost of ambient PM2.5 air pollution estimated in this report for the year 2016, along with cost estimates in previous reports for previous years, cannot be readily compared to 0% infer whether global air quality has worsened or improved. This is mainly because each LI LMI UMI HI non-OECD HI OECD cost estimate is based on (i) exposure-response functions that are evolving over time as new evidence becomes available; (ii) global ambient PM2.5 population-exposure estimates Note: LI = low-income countries; LMI = lower-middle-income countries; UMI = upper-middle-income countries; and HI = high-income countries. that also evolve over time with methodological developments and increased availability of Assignment of countries to categories based on World Bank income classifications. ground-level PM monitoring data; and (iii) modifications in the valuation in health damag- es (that is, the inclusion of the cost of morbidity in this report). Each cost estimate should rather be viewed as a reflection of available evidence and scientific understanding at the time of the estimate. World Bank Group 60 The Global Health Cost of Ambient PM2.5 Air Pollution 61 Conclusions This report provides an estimate of the global cost of ambient PM2.5 air pollution in •  2016 based on the GBD 2016 study. It thus represents an update of the estimated cost in 2013 reported in World Bank and IHME (2016) that was based on the GBD 2013 study. This report distinguishes itself from the 2013 estimate in important aspects. It is •  based on  evised exposure-response functions from the GBD 2016 study that differ from —R the functions from the GBD 2013 study for several health outcomes;  evised global ambient PM2.5 population-exposure estimates from the GBD 2016 —R study that are based on calibration from a substantially larger PM ground-level measurement database than the data used for the GBD 2013 study;  nnual deaths of 4.1 million from ambient PM2.5 according to the GBD 2016 —A study, in contrast to 2.9 million according to the GBD 2013 study—a change which is mainly associated with higher health risks of PM2.5 and higher estimates of global population exposure to ambient PM2.5 in the GBD 2016 study than in the GBD 2013 study, rather than with worsening of air quality; and nclusion of an estimate of the cost of morbidity based on estimates of YLD —I from ambient PM2.5 reported by the GBD 2016 study. Health damages and cost of ambient PM2.5 are staggering, especially in developing •  countries, globally reaching 4.1 million deaths and 15 billion days of illness in 2016, with a welfare cost of $5.7 trillion equivalent to 4.8 percent of global GDP (PPP adjusted). This estimated cost for 2016 is 50 percent higher in real terms than the estimate for •  2013 in World Bank and IHME (2016). The reasons for the higher cost estimate are mainly changes in exposure-response functions, a substantially higher estimate of global ambient PM2.5 exposure due to improved methodology and ground-level PM monitoring data availability, and inclusion of an estimate of the cost of morbidity. The higher estimate of global ambient PM2.5 exposure is more due to improved methodol- ogy and availability of ground-level PM monitoring data than an actual worsening of global ambient PM2.5 air quality from 2013 to 2016, although the exact contribution of each of these two factors is difficult to ascertain. Conclusions 62 The Global Health Cost of Ambient PM2.5 Air Pollution 63 About 87 percent of the total global cost of health damages in 2016 is from prema- •  ture mortality and 13 percent from morbidity. The cost of morbidity as a share of total cost varies from as low as 3 percent to as high as 27 percent across countries. Two-thirds of the health damages occur in the South Asia and East Asia and •  Pacific regions. Costs reach as high as 7.5–8 percent of GDP in China and India, the two countries in •  which over half of global deaths from ambient PM2.5 occur. Although the methodology and ground-level measurements data available for the •  global ambient PM2.5 population-exposure estimates from the GBD 2016 study rep- resent important improvements over the estimates from the GBD 2013 study, based on calibration from a larger PM ground-level measurement database, the database nevertheless contains PM measurements from only half of the countries in the world and is almost entirely lacking large parts of Sub-Saharan Africa. PM measurements are particularly scarce in low-income countries and Sub-Saharan •  Africa, with one monitor per 54 and 15 million people, respectively, in contrast to one monitor per 0.3 million people in high-income countries. World Bank Group 64 The Global Health Cost of Ambient PM2.5 Air Pollution 65 Appendixes Appendix A: Annual Health Damages and Costs of Ambient PM2.5, 2016 TABLE 6 Annual Health Damages and Costs of Ambient PM2.5, 2016 Cost of health damages from PM2.5 Annual PM2.5 Deaths YLD US$, $ (PPP), % of GDP Study from PM2.5 from PM2.5 Millions Millions equivalent (µg/m3) Afganistan 63 26,983 11,838 977 3,265 5.02 Albania 15 1,377 974 521 1,499 4.37 Algeria 37 12,610 15,579 5,165 20,256 3.31 American 4 2 4 - - - Samoa Andorra 11 25 28 - - - Angola 36 7,033 5,941 1,860 3,887 2.08 Antigua and 16 19 14 27 43 1.89 Barbuda Argentina 14 15,611 9,135 18,815 30,127 3.45 Armenia 27 2,075 1,382 640 1,565 6.07 Aruba - - - - - - Australia 6 3,071 3,226 16,852 15,793 1.40 Austria 15 3,270 3,479 15,834 17,949 4.10 Azerbaijan 33 6,780 4,058 2,406 10,707 6.36 Bahamas, 13 94 58 257 258 2.84 The Appendixes 66 The Global Health Cost of Ambient PM2.5 Air Pollution 67 Cost of health damages from PM2.5 Cost of health damages from PM2.5 Annual Annual PM2.5 Deaths YLD US$, $ (PPP), % of GDP PM2.5 Deaths YLD US$, $ (PPP), % of GDP Study Study from PM2.5 from PM2.5 Millions Millions equivalent from PM2.5 from PM2.5 Millions Millions equivalent (µg/m3) (µg/m3) Cayman Bahrain 80 294 706 748 1,584 2.35 - - - - - - Islands Central Bangladesh 101 108,687 109,715 11,289 29,749 5.10 African 66 4,158 1,868 77 140 4.37 Republic Barbados 18 103 70 162 169 3.53 Chad 96 9,108 4,174 398 1,194 4.15 Belarus 20 9,583 5,306 4,605 16,671 9.71 Channel - - - - - - Islands Belgium 16 4,938 5,817 23,324 26,325 5.00 Chile 22 4,918 4,685 6,999 12,158 2.83 Belize 20 76 45 30 53 1.71 China 56 1,075,039 1,644,758 852,964 1,631,202 7.62 Benin 96 5,630 3,875 305 838 3.55 Colombia 17 9,668 8,886 5,251 12,806 1.86 Bermuda 10 13 10 - - - Comoros 19 160 131 8 16 1.34 Bhutan 56 361 697 113 352 5.05 Congo, 56 32,664 21,120 822 1,480 2.35 Dem. Rep. Bolivia 22 3,573 1,831 879 2,048 2.60 Congo, Rep. 56 1,938 1,655 243 909 3.10 Bosnia and 39 3,339 3,308 1,558 3,994 9.41 Herzegovina Costa Rica 19 842 992 1,025 1,440 1.78 Botswana 23 602 839 387 953 2.53 Côte d’Ivoire 58 12,339 7,314 1,373 3,346 3.80 Brazil 13 49,724 35,898 40,484 70,801 2.25 Croatia 20 3,388 2,749 4,340 8,471 8.61 Brunei 6 39 63 107 307 0.94 Darussalam Cuba 17 5,344 3,193 3,784 7,569 4.34 Bulgaria 26 9,087 6,178 6,515 17,015 12.43 Curaçao - - - - - - Burkina Faso 111 10,223 5,130 413 1,094 3.41 Cyprus 18 333 305 897 1,254 4.53 Burundi 46 4,730 2,719 74 202 2.47 Czech 19 6,543 6,226 13,457 25,573 6.98 Republic Cabo Verde 67 238 222 59 130 3.66 Denmark 10 1,806 2,042 10,839 10,084 3.54 Cambodia 26 5,915 7,830 582 1,712 2.91 Djibouti 72 433 359 63 113 3.66 Cameroon 140 15,986 9,779 1,166 3,708 4.82 Dominica 17 21 12 13 21 2.54 Canada 8 6,958 7,410 32,553 33,994 2.13 Appendixes 68 The Global Health Cost of Ambient PM2.5 Air Pollution 69 Cost of health damages from PM2.5 Cost of health damages from PM2.5 Annual Annual PM2.5 Deaths YLD US$, $ (PPP), % of GDP PM2.5 Deaths YLD US$, $ (PPP), % of GDP Study Study from PM2.5 from PM2.5 Millions Millions equivalent from PM2.5 from PM2.5 Millions Millions equivalent (µg/m3) (µg/m3) Dominican 24 2,964 1,798 1,825 4,128 2.55 Greece 11 5,818 6,977 11,661 17,251 5.99 Republic Ecuador 13 2,240 1,720 1,144 2,163 1.17 Greenland 4 3 5 - - - Egypt, 126 67,555 62,069 21,604 68,429 6.42 Grenada 18 39 18 33 49 3.27 Arab Rep. El Salvador 33 1,969 1,365 706 1,441 2.64 Guam 9 58 71 - - - Equatorial 71 240 389 217 666 2.14 Guatemala 29 4,105 2,420 1,400 2,683 2.04 Guinea Eritrea 51 1,832 1,264 - - - Guinea 46 6,585 3,422 196 506 3.11 Guinea- Estonia 6 388 271 728 1,216 3.14 59 1,153 619 44 112 3.90 Bissau Eswatini 24 425 402 111 335 2.99 Guyana 20 295 116 106 186 3.07 Ethiopia 50 37,342 27,642 1,683 4,131 2.33 Haiti 24 4,941 1,613 220 530 2.74 Faroe Islands - - - - - - Honduras 29 2,516 1,544 454 911 2.11 Hong Kong Fiji 8 191 183 90 167 1.94 - - - - - - SAR, China Finland 6 1,099 1,286 5,425 5,420 2.29 Hungary 25 8,999 7,971 12,355 26,028 9.94 France 12 16,444 12,083 67,383 75,814 2.73 Iceland 7 58 78 393 337 1.96 French - - - - - - India 76 1,034,420 2,274,778 177,298 681,683 7.83 Polynesia Gabon 47 684 690 482 1,215 3.39 Indonesia 17 79,739 141,136 28,062 91,270 3.01 Iran, Gambia, The 95 754 579 23 83 2.41 49 28,716 34,603 13,890 47,760 3.53 Islamic Rep. Georgia 21 3,533 1,997 1,142 2,963 7.97 Iraq 73 18,191 14,795 8,245 31,040 4.81 Germany 13 36,938 40,479 170,910 198,597 4.93 Ireland 9 1,060 1,258 6,827 7,633 2.32 Ghana 54 11,803 8,768 1,344 3,814 3.15 Isle of Man - - - - - - Gibraltar - - - - - - Israel 19 1,815 2,399 8,032 8,163 2.52 Appendixes 70 The Global Health Cost of Ambient PM2.5 Air Pollution 71 Cost of health damages from PM2.5 Cost of health damages from PM2.5 Annual Annual PM2.5 Deaths YLD US$, $ (PPP), % of GDP PM2.5 Deaths YLD US$, $ (PPP), % of GDP Study Study from PM2.5 from PM2.5 Millions Millions equivalent from PM2.5 from PM2.5 Millions Millions equivalent (µg/m3) (µg/m3) Italy 15 24,555 22,019 85,764 107,210 4.64 Luxembourg 16 175 239 1,758 1,810 2.93 Macao Jamaica 15 802 504 330 599 2.35 - - - - - - SAR, China Japan 13 45,783 71,794 207,804 221,564 4.21 Madagascar 21 8,500 5,226 207 777 2.07 Jordan 37 1,621 2,398 696 1,540 1.80 Malawi 28 5,126 3,231 91 354 1.67 Kazakhstan 20 9,360 6,341 7,105 23,901 5.32 Malaysia 18 10,461 18,160 11,693 34,060 3.95 Kenya 16 6,509 6,142 748 1,621 1.06 Maldives 27 84 221 80 123 2.23 Kiribati 4 8 7 1 1 0.48 Mali 92 6,385 5,165 350 948 2.49 Korea, Dem. Malta 12 150 146 429 649 3.92 People’s 36 23,360 22,642 - - - Rep. Marshall 11 15 17 4 5 2.22 Korea, Rep. 29 16,803 40,527 59,523 77,273 4.22 Islands Mauritania 124 1,762 1,650 153 546 3.29 Kosovo - - - - - - Mauritius 14 420 833 441 966 3.62 Kuwait 111 860 2,115 2,498 6,363 2.19 Kyrgyz Mexico 19 24,390 22,256 19,333 42,106 1.85 18 2,480 1,347 191 628 2.91 Republic Micronesia, 9 25 22 6 7 1.79 Lao PDR 28 3,183 3,786 631 1,658 3.97 Fed. Sts. Moldova 20 3,175 1,639 477 1,340 7.07 Latvia 15 1,670 1,166 2,457 4,530 8.88 Monaco - - - - - - Lebanon 33 1,796 2,697 1,453 2,570 3.06 Mongolia 30 1,536 879 503 1,669 4.51 Lesotho 27 1,021 759 74 226 3.38 Montenegro 20 421 322 277 696 6.63 Liberia 17 968 635 25 45 1.21 Morocco 25 12,406 12,305 3,138 8,684 3.09 Libya 64 2,421 3,204 - - - Mozambique 21 6,689 4,596 150 477 1.36 Liechtenstein - - - - - - Myanmar 49 25,280 55,512 2,932 13,273 4.35 Lithuania 17 2,496 1,687 3,945 7,945 9.23 Appendixes 72 The Global Health Cost of Ambient PM2.5 Air Pollution 73 Cost of health damages from PM2.5 Cost of health damages from PM2.5 Annual Annual PM2.5 Deaths YLD US$, $ (PPP), % of GDP PM2.5 Deaths YLD US$, $ (PPP), % of GDP Study Study from PM2.5 from PM2.5 Millions Millions equivalent from PM2.5 from PM2.5 Millions Millions equivalent (µg/m3) (µg/m3) Namibia 26 657 816 266 680 2.59 Poland 26 26,382 23,847 35,256 79,248 7.51 Nauru - - - - - - Portugal 9 3,576 2,918 7,807 12,067 3.82 Puerto Nepal 78 20,453 33,528 1,117 3,780 5.28 20 1,358 1,036 - - - Rico (US) Netherlands 15 6,337 9,031 32,458 36,474 4.21 Qatar 148 290 1,263 1,570 3,374 1.03 New - - - - - - Romania 19 16,933 12,553 16,474 41,082 8.82 Caledonia Russian New Zealand 6 552 511 2,413 2,391 1.30 16 125,455 78,840 111,381 294,897 8.68 Federation Nicaragua 23 1,017 892 176 454 1.33 Rwanda 53 3,765 3,088 179 488 2.14 Niger 204 14,390 7,582 302 814 4.02 Samoa 4 10 10 4 6 0.46 Saint Martin Nigeria 122 68,887 59,119 12,313 33,170 3.04 - - - - - - (French) North 32 1,599 1,447 814 2,351 7.47 San Marino - - - - - - Macedonia Northern São Tomé 15 41 33 6 11 1.64 Mariana 11 11 37 - - - and Principe Islands Saudi Arabia 188 11,210 18,537 24,473 66,510 3.79 Norway 8 1,097 1,411 8,299 6,950 2.24 Senegal 57 6,219 4,431 412 1,105 2.79 Oman 78 1,193 1,910 2,059 5,873 3.11 Serbia 19 5,968 5,472 3,158 8,570 8.37 Pakistan 76 124,577 174,172 16,486 58,943 5.81 Seychelles 15 30 58 53 101 3.74 Palau - - - - - - Sierra Leone 42 3,049 1,669 95 283 2.60 Panama 14 653 626 908 1,528 1.65 Papua Singapore 25 1,373 2,143 7,187 11,922 2.42 14 3,312 2,904 517 654 3.06 New Guinea Sint Maarten - - - - - - Paraguay 24 1,881 1,245 662 1,553 2.41 (Dutch) Slovak 20 3,614 2,970 6,456 11,989 7.21 Peru 26 7,526 5,549 4,004 8,624 2.08 Republic Slovenia 18 905 1,065 2,356 3,636 5.36 Philippines 23 44,389 71,263 11,837 31,312 3.88 Appendixes 74 The Global Health Cost of Ambient PM2.5 Air Pollution 75 Cost of health damages from PM2.5 Cost of health damages from PM2.5 Annual Annual PM2.5 Deaths YLD US$, $ (PPP), % of GDP PM2.5 Deaths YLD US$, $ (PPP), % of GDP Study Study from PM2.5 from PM2.5 Millions Millions equivalent from PM2.5 from PM2.5 Millions Millions equivalent (µg/m3) (µg/m3) Solomon Trinidad 9 144 115 20 22 1.66 17 473 299 761 1,580 3.63 Islands and Tobago Somalia 24 4,024 1,749 - - - Tunisia 36 4,932 5,483 1,736 5,460 4.13 South Africa 36 21,061 32,619 11,669 29,264 3.96 Turkey 37 27,103 45,783 32,340 72,681 3.77 South Sudan 49 6,926 3,620 348 882 3.85 Turkmenistan 38 3,185 1,417 1,903 5,029 5.26 Turks and Spain 10 12,289 9,448 37,072 50,740 3.01 - - - - - - Caicos Islands Sri Lanka 26 7,330 16,965 3,470 11,144 4.27 Tuvalu - - - - - - St. Kitts - - - - - - Uganda 74 14,566 10,568 584 1,754 2.29 and Nevis St. Lucia 17 44 34 30 45 2.21 Ukraine 19 53,665 26,015 9,715 36,765 10.42 St. Vincent United Arab 105 3,093 6,732 11,794 22,702 3.38 and the 17 38 18 23 38 2.98 Emirates Grenadines United 12 24,231 22,706 108,127 115,470 4.13 Sudan 78 19,046 14,805 3,784 7,411 3.96 Kingdom United 9 93,376 164,451 622,383 622,383 3.35 Suriname 22 191 109 114 248 3.14 States Uruguay 12 1,188 887 1,799 2,555 3.43 Sweden 5 1,384 1,796 8,061 7,682 1.58 Uzbekistan 47 20,297 9,605 3,365 10,385 5.01 Switzerland 11 2,041 3,177 17,900 14,282 2.71 Syrian Arab Vanuatu 9 83 63 17 18 2.19 44 6,789 5,486 - - - Republic Venezuela, RB 26 8,559 7,319 - - - Tajikistan 61 4,651 2,679 245 919 3.53 Vietnam 26 40,170 78,104 7,835 23,030 3.87 Tanzania 22 14,314 9,080 850 2,693 1.79 Virgin Islands - - - - - - Thailand 23 25,432 65,373 15,832 45,334 3.89 (British) Virgin Islands 18 61 31 - - - Timor-Leste 17 256 433 28 56 1.97 (US) West Bank 19 1,289 876 286 286 2.14 Togo 84 3,603 2,437 138 357 3.15 and Gaza Yemen, Rep. 73 12,562 9,531 911 2,308 3.34 Tonga 4 5 6 2 3 0.42 Appendixes 76 The Global Health Cost of Ambient PM2.5 Air Pollution 77 Cost of health damages from PM2.5 for x < xcf (A2.1a) Annual PM2.5 Deaths YLD US$, $ (PPP), % of GDP for x ≥ xcf (A2.1b) Study from PM2.5 from PM2.5 Millions Millions equivalent (µg/m3) Zambia 31 5,713 3,431 495 1,647 2.53 where x is the ambient concentration of PM2.5 in µg/m3 and xcf is a counterfactual concen- tration below which it is assumed that no association exists. The function allows for the prediction of relative risks over a very large range of PM2.5 concentrations, with RR(xcf + Zimbabwe 25 4,552 3,772 308 613 1.89 1) ~ 1 + αβ and RR(∞) = 1 + α being the maximum risk (Burnett et al. 2014; Shin et al. 2013). The parameter values of the risk function are derived based on studies of health out- Source: Ambient PM2.5 is from the World Development Indicators (database), http://datatopics.worldbank.org/world-development-indicators/. Annual deaths from PM2.5 are from the GBD 2016 study. The annual cost is estimated using the methodology applied in World Bank and IHME (2016). comes associated with long-term exposure to ambient PM pollution, secondhand tobac- Note: Cost of health damages of PM2.5 is not estimated for some countries and territories due to lack of estimate of deaths from ambient PM2.5 in the GBD 2016 study or absence of GDP per capita (PPP) in the World Development Indicators Database. co smoke, household air pollution from solid cooking fuels, and active tobacco smoking (Burnett et al. 2014). This provides a risk function that can be applied to a wide range of ambient PM2.5 concentrations around the world as well as to high household air pollution Appendix B: levels of PM2.5 from the combustion of solid fuels. Health Damages of Ambient PM2.5 The disease outcomes assessed are IHD, cerebrovascular disease (stroke), lung cancer, COPD, and ALRI. The risk functions for IHD and cerebrovascular disease are age-specific Particulate matter, and especially PM2.5, is the ambient air pollutant that is associated with five-year age intervals from 25 years of age, while singular age group risk functions globally with the largest health damages. Health damages of PM2.5 exposure include both are applied for lung cancer, COPD, and ALRI. premature mortality and morbidity. The most substantial health damages of PM2.5 are car- diovascular disease, COPD, lung cancer, and ALRI (Lim et al. 2012; Mehta et al. 2013; GBD The attributable fraction of disease from PM2.5 exposure is calculated by the 2013 Risk Factors Collaborators 2015; GBD 2015 Risk Factors Collaborators 2016; GBD following expression: 2016 Risk Factors Collaborators 2017; Pope et al. 2009; Pope et al. 2011). The methodol- ogies to estimate these health damages have evolved as the body of research evidence has increased. (A2.2) Ambient PM air pollution where Pi is the share of the population exposed to PM2.5 concentrations in the range xi-1 to Over a decade ago, Pope et al. (2002) found an elevated risk of cardiopulmonary and xi. This attributable fraction is calculated for each disease outcome k and age group l. The lung cancer mortality from long-term exposure to outdoor PM2.5 in a study of a large pop- disease burden (B) in terms of annual cases of disease outcomes due to PM2.5 exposure is ulation of adults 30 or more years of age in the United States. Cardiopulmonary mortality then estimated by: includes mortality from respiratory infections, cardiovascular disease, and chronic respi- ratory disease. The WHO used the findings of research by Pope and his colleagues—see Pope et al. (2009) and Pope et al. (2011)—when estimating global mortality from outdoor (A2.3) air pollution (Ezzati et al. 2004; WHO 2009). Since then, recent research suggests that the marginal increase in the relative risk of mortality from PM2.5 declines with increasing con- centrations of PM2.5 (Pope et al. 2009; Pope et al. 2011). Pope et al. (2009) and Pope where Dkl is the total annual number of cases of disease k in age group l, and AFkl is the et al. (2011) derive a shape of the PM2.5 exposure-response curve based on studies of attributable fraction of these cases of disease k in age group l due to PM2.5 exposure. mortality from actively smoking cigarettes, secondhand smoke, and outdoor PM2.5 air pollution. An integrated exposure-response function The GBD 2010–2016 studies take Pope et al. (2009) and Pope et al. (2011) some steps further by deriving an IER relative risk function, RR, for disease outcome, k, in age group, l, associated with exposure to fine particulate matter pollution (PM2.5) both in the ambient and household environments: Appendixes 78 The Global Health Cost of Ambient PM2.5 Air Pollution 79 Appendix C: Most of the meta-analyses of VSL are entirely or predominantly based on hedonic wage studies. A meta-analysis prepared for the OECD is, however, exclusively based on stat- Valuation of Premature Mortality ed-preference studies, which are arguably of greater relevance for the valuation of mortality risk from environmental factors than hedonic wage studies (Lindhjem et al. The predominant measure of the welfare cost of premature death used by economists 2011; Navrud and Lindhjem 2010; OECD 2012). These stated-preference studies are from is VSL. VSL is based on the valuation of mortality risk. Everyone in society is constantly a database of more than 1,000 VSL estimates from multiple studies in over 30 countries, facing a certain risk of dying. Examples of such risks are occupational fatality risk, risk of including in developing countries (www.oecd.org/env/policies/VSL). traffic accident fatality, and environmental mortality risks. It has been observed that indi- viduals adjust their behavior and decisions in relation to such risks. For instance, individu- Narain and Sall (2016) present a benefit-transfer methodology for valuing mortality from als demand a higher wage (a wage premium) for a job that involves a higher occupational environmental health risks, drawing on the empirical literature of VSL, especially OECD risk of fatal accidents than other jobs, individuals may purchase safety equipment to (2012). The methodology is applied in the recent publication by the World Bank and IHME reduce the risk of death, and/or individuals and families may be willing to pay a premium (2016) on the global cost of air pollution. The proposed benefit transfer function is: or higher rent for properties (land and buildings) in a cleaner and less polluted neighbor- hood or city. (A3.2) Through the observation of individuals’ choices and willingness to pay (WTP) for re- ducing mortality risk (or minimum amounts that individuals require to accept a higher where VSLc,n is the estimated VSL for country c in year n, VSLOECD is the average base VSL mortality risk), it is possible to estimate the value to society of reducing mortality risk, or, in the sample of OECD countries with VSL studies ($3.83 million), Yc,n is GDP per capita equivalently, measure the welfare cost of a particular mortality risk. For instance, it may in country c in year n, and YOECD is the average GDP per capita for the sample of OECD be observed that a certain health hazard has a mortality risk of 2.5/10,000. This means countries ($37,000), and ɛ an income elasticity of 1.2 for LMICs and 0.8 for high-income that 2.5 individuals die from this hazard for every 10,000 individuals exposed. If each countries. All values are in PPP prices. For VSL in US dollars, VSLc,n is therefore multiplied individual on average is willing to pay US$40 for eliminating this mortality risk, then every by the ratio of PPP conversion factor to nominal exchange rates, available in the World 10,000 individuals are collectively willing to pay US$400,000. Dividing this amount by Development Indicators Database from the World Bank. the risk gives the VSL of US$160,000. Mathematically it can be expressed as follows: VSL = WTPAve x 1/ R (A3.1) Appendix D: where WTPAve is the average WTP per individual for a mortality risk reduction of magni- tude R. In the illustration above, R = 2.5/10,000 (or R = 0.00025) and WTPAve = US$40. Valuation of Morbidity Thus, if 10 individuals die from the health risk illustrated above, the cost to society is Two valuation techniques are commonly used to estimate the cost of morbidity or illness. 10 x VSL = 10 x US$0.16 million = US$1.6 million. The cost-of-illness approach includes the cost of medical treatment, the value of income, The main approaches to estimating VSL are through the revealed preferences and the and the time lost to illness. The second approach equates the cost of illness to individuals’ stated preferences of people’s WTP for a reduction in mortality risk or their willingness WTP for avoiding an episode of illness. The latter therefore includes the welfare cost of to accept an increase in mortality risk. Most of the studies of revealed preferences are pain and suffering from illness. hedonic wage studies, which estimate labor market wage differentials associated with Studies in many countries have found that individuals’ WTP to avoid an episode of an differences in occupational mortality risk. Most of the stated preference studies rely on acute illness is generally much higher than the cost of treatment and value of income and contingent valuation methods, which in various forms ask individuals about their WTP time losses (Alberini and Krupnick 2000; Cropper and Oates 1992; Dickie and Gerking for mortality risk reduction. 2002; Wilson 2003). Studies of WTP for a reduction in the risk of mortality have been carried out in numerous The OECD, in its report on the global economic consequences of outdoor air pollution, countries. A commonly used approach to estimate VSL in a specific country without such includes both the cost of mortality and morbidity (OECD 2016). Mortality is valued using studies is therefore to use a benefit transfer based on meta-analyses of WTP studies from VSL, and the cost of morbidity is estimated both in terms of other countries. Several meta-analyses have been conducted in the last two decades. Me- ta-analyses assess characteristics that determine VSL, such as household income, size of risk reduction, other individual and household characteristics, and, often, characteristics of the methodologies used in the original WTP studies. Appendixes 80 The Global Health Cost of Ambient PM2.5 Air Pollution 81 i. Market impacts or cost-of-illness (reduced labor productivity and increased health ex- Cost of morbidity (C) in country k is calculated as follows: penditures associated with bronchitis, asthma, hospital admissions, and restricted-activity days from illness), and Ck = Dk * wk (A4.2) ii. Nonmarket impacts (welfare cost of pain and suffering from illness). or Globally, the OECD estimated the cost of market impacts or cost-of-illness to about 0.2 Ck = YLDk * 365 * (wk /d) (A4.3) percent of GDP or equivalent to 4 percent of the cost of mortality. Expressed in terms of where w is the average daily wage rate, or the cost per day of illness.16 The average daily welfare, using the equivalent variation of income, the cost was 0.4 percent of GDP or 8 wage rate is estimated as follows: percent of the cost of mortality. The nonmarket impacts or welfare cost was equivalent to 0.5 percent of GDP or 9 percent of mortality cost. wk = GDPk / Lk / 250 * sk (A4.4) Thus, the total cost of morbidity was estimated at 0.7–0.9 percent of GDP or 13–17 per- where GDP is the country’s total GDP, L is total labor force, s is labor compensation share cent of the cost of mortality according to OCED report. of GDP, and annual working days is averaging 250. GDP and L are from the World Devel- opment Indicators Database (http://datatopics.worldbank.org/world-development-indica- Estimating the cost of morbidity requires much more, and less accessible, data (including tors/) and s is from PENN World Table, version 8. baseline health data) than estimating mortality. A simplified approach is therefore applied in this report using the following steps: This provides an estimated global cost of morbidity from ambient PM2.5 equivalent to 0.6 percent of GDP in 2016 or 16 percent of the cost of mortality. This is very close to the esti- i. Estimates of YLD from ambient PM2.5 from the GBD 2016 study are converted to days of mates by OECD presented above. illness by applying the disability weights in the GBD studies. The cost of morbidity, as a share of the total cost of health damages of ambient PM2.5, ii. The cost of a day of illness is then approximated as the average daily wage rates to varies from 3 percent to 38 percent across countries. The share is mainly determined by reflect income losses from illness, health expenditure, time losses, and the welfare cost of YLD per death from ambient PM2.5 as well as wage rate relative to GDP per capita. YLD pain and suffering. per death varies from 0.33 to 4.36 across countries (table D.1). iii. The cost of a day of illness is also applied to individuals without income because illness prevents most of these individuals from undertaking household work and other activities with a social value, as well as involves all the non-income impacts of illness. TABLE 7 Years of Life Lived with Disability (YLD) from Ambient PM2.5 in 2016 The cost of morbidity is thus estimated as follows. First, annual disease days (D) in coun- try k are calculated as: YLD Cost of Cost of morbidity YLD Study per death morbidity (% of total cost of Dk = YLDk * 365 / d from PM2.5 from PM2.5 (US$, millions) health damages) where YLD is years lost to disease from exposure to ambient PM2.5, taken from the GBD Afghanistan 11,838 0.44 113 12 2016 study, and d is the disability weight of disease, here applied a weight of 0.15. The disability weight is a measure used in the GBD studies to calculate YLD from days Albania 974 0.71 50 10 of illness, disease, or injury. The weight for diseases associated with exposure to ambient PM2.5 ranges from 0.006 to 0.133 for infectious diseases (for example, ALRI); from 0.036 Algeria 15,579 1.24 972 19 to 0.569 for cancers (for example, lung cancer); from 0.033 to 0.224 for heart disease; from 0.019 to 0.588 for stroke; and from 0.019 to 0.408 for COPD, with the ranges reflect- American Samoa 4 1.60 - - ing various degrees of severity and stages of illness and treatments (Salomon et al. 2015). A disability weight of 0.15, as applied here, is the weighted average for mild, moderate, Andorra 28 1.13 - - and severe illness (with frequency distributions of 45 percent, 45 percent, and 10 percent respectively) weighted by each illness’s share of global YLD from ambient PM2.5 exposure. 16  Note that the cost of morbidity, Ck, is independent of the numerical value of d if the cost per day of illness, wk, varies in proportion to the disability of illness, d (that is, wk = wk (d) = βk d where βk is a constant). This can be seen by substituting βk d for wk in equation A4.3. Appendixes 82 The Global Health Cost of Ambient PM2.5 Air Pollution 83 YLD Cost of Cost of morbidity YLD Cost of Cost of morbidity YLD YLD Study per death morbidity (% of total cost of Study per death morbidity (% of total cost of from PM2.5 from PM2.5 from PM2.5 (US$, millions) health damages) from PM2.5 (US$, millions) health damages) Angola 5,941 0.84 261 14 Brazil 35,898 0.72 3,253 8 Antigua and Virgin Islands 14 0.74 2 8 - - - - Barbuda (British) Argentina 9,135 0.591 1,039 6 Brunei Darussalam 63 1.62 13 13 Armenia 1,382 0.67 59 9 Bulgaria 6,178 0.68 451 7 Aruba - - - - Burkina Faso 5,130 0.50 41 10 Australia 3,226 1.05 1,709 10 Burundi 2,719 0.57 10 13 Austria 3,479 1.06 1,757 11 Cabo Verde 222 0.93 7 12 Azerbaijan 4,058 0.60 70 3 Cambodia 7,830 1.32 91 16 Bahamas, The 58 0.62 9 4 Cameroon 9,779 0.61 113 10 Bahrain 706 2.41 102 14 Canada 7,410 1.06 3,224 10 Bangladesh 109,715 1.01 1,707 15 Cayman Islands - - - - Central African Barbados 70 0.68 15 10 1,868 0.45 2 3 Republic Belarus 5,306 0.55 317 7 Chad 4,174 0.46 49 12 Belgium 5,817 1.18 3,246 14 Channel Islands - - - - Belize 45 0.59 2 8 Chile 4,685 0.95 562 8 Benin 3,875 0.69 44 14 China 1,644,758 1.53 93,045 11 Bermuda 10 0.77 - - Colombia 8,886 0.92 457 9 Bhutan 697 1.93 34 30 Comoros 131 0.82 1 18 Bolivia 1,831 0.51 50 6 Congo, Dem. Rep. 21,120 0.65 123 15 Bosnia and 3,308 0.99 257 16 Congo, Rep. 1,655 0.85 32 13 Herzegovina Botswana 839 1.39 26 7 Costa Rica 992 1.18 147 14 Appendixes 84 The Global Health Cost of Ambient PM2.5 Air Pollution 85 YLD Cost of Cost of morbidity YLD Cost of Cost of morbidity YLD YLD Study per death morbidity (% of total cost of Study per death morbidity (% of total cost of from PM2.5 from PM2.5 from PM2.5 (US$, millions) health damages) from PM2.5 (US$, millions) health damages) Côte d’Ivoire 7,314 0.59 142 10 France 12,083 0.73 6,064 9 Croatia 2,749 0.81 465 11 French Polynesia - - - - Cuba 3,193 0.60 264 7 Gabon 690 1.01 41 8 Curaçao - - - Gambia, The 579 0.77 3 14 Cyprus 305 0.91 72 8 Georgia 1,997 0.57 57 5 Czech Republic 6,226 0.95 1,243 9 Germany 40,479 1.10 19,157 11 Denmark 2,042 1.13 1,423 13 Ghana 8,768 0.74 142 11 Djibouti 359 0.83 12 18 Gibraltar - - - - Dominica 12 0.59 1 8 Greece 6,977 1.20 1,439 12 Dominican 1,798 0.61 98 5 Greenland 5 1.40 - - Republic Ecuador 1,720 0.77 52 5 Grenada 18 0.47 2 6 Egypt, Arab Rep. 62,069 0.92 2,276 11 Guam 71 1.23 - - El Salvador 1,365 0.69 63 9 Guatemala 2,420 0.59 105 7 Equatorial Guinea 389 1.62 25 12 Guinea 3,422 0.52 19 9 Eritrea 1,264 0.69 - - Guinea-Bissau 619 0.54 5 10 Estonia 271 0.70 54 7 Guyana 116 0.39 6 6 Eswatini 402 0.95 21 19 Haiti 1,613 0.33 14 6 Ethiopia 27,642 0.74 202 12 Honduras 1,544 0.61 47 10 Hong Kong Faroe Islands - - - - - - - - SAR, China Fiji 183 0.95 12 13 Hungary 7,971 0.89 1,304 11 Finland 1,286 1.17 670 12 Iceland 78 1.35 58 15 Appendixes 86 The Global Health Cost of Ambient PM2.5 Air Pollution 87 YLD Cost of Cost of morbidity YLD Cost of Cost of morbidity YLD YLD Study per death morbidity (% of total cost of Study per death morbidity (% of total cost of from PM2.5 from PM2.5 from PM2.5 (US$, millions) health damages) from PM2.5 (US$, millions) health damages) India 2,274,778 2.20 47,747 27 Lebanon 2,697 1.50 242 17 Indonesia 141,136 1.77 4,689 17 Lesotho 759 0.74 10 14 Iran, Islamic Rep. 34,603 1.21 1,270 9 Liberia 635 0.66 4 16 Iraq 14,795 0.81 785 10 Libya 3,204 1.32 - - Ireland 1,258 1.19 855 13 Liechtenstein - - - - Isle of Man - - - - Lithuania 1,687 0.68 260 7 Israel 2,399 1.32 1,059 13 Luxembourg 239 1.36 244 14 Italy 22,019 0.90 8,647 10 Macao SAR, China - - - - Jamaica 504 0.63 27 8 Madagascar 5,226 0.61 21 10 Japan 71,794 1.57 27,633 13 Malawi 3,231 0.63 11 12 Jordan 2,398 1.48 178 26 Malaysia 18,160 1.74 1,983 17 Kazakhstan 6,341 0.68 363 5 Maldives 221 2.64 19 24 Kenya 6,142 0.94 148 20 Mali 5,165 0.81 59 17 Kiribati 7 0.84 0 14 Malta 146 0.97 43 10 Korea, Dem. 22,642 0.97 - - Marshall Islands 17 1.16 1 16 People’s Rep. Korea, Rep. 40,527 2.41 11,294 19 Mauritania 1,650 0.94 28 18 Kosovo - - - - Mauritius 833 1.98 67 15 Kuwait 2,115 2.46 251 10 Mexico 22,256 0.91 1,434 7 Micronesia, Kyrgyz Republic 1,347 0.54 18 9 22 0.86 1 13 Fed. Sts. Lao PDR 3,786 1.19 89 14 Moldova 1,639 0.52 53 11 Latvia 1,166 0.70 182 7 Monaco - - - - Appendixes 88 The Global Health Cost of Ambient PM2.5 Air Pollution 89 YLD Cost of Cost of morbidity YLD Cost of Cost of morbidity YLD YLD Study per death morbidity (% of total cost of Study per death morbidity (% of total cost of from PM2.5 from PM2.5 from PM2.5 (US$, millions) health damages) from PM2.5 (US$, millions) health damages) Papua Mongolia 879 0.57 34 7 2,904 0.88 72 14 New Guinea Montenegro 322 0.76 27 10 Paraguay 1,245 0.66 56 8 Morocco 12,305 0.99 471 15 Peru 5,549 0.74 182 5 Mozambique 4,596 0.69 16 11 Philippines 71,263 1.61 1,904 16 Myanmar 55,512 2.20 630 22 Poland 23,847 0.90 3,344 9 Namibia 816 1.24 47 18 Portugal 2,918 0.82 746 10 Nauru - - - - Puerto Rico (US) 1,036 0.76 - - Nepal 33,528 1.64 219 20 Qatar 1,263 4.36 180 11 Netherlands 9,031 1.43 4,584 14 Romania 12,553 0.74 1,293 8 Russian New Caledonia - - - - 78,840 0.63 7,931 7 Federation New Zealand 511 0.93 185 8 Rwanda 3,088 0.82 28 16 Nicaragua 892 0.88 21 12 Samoa 10 0.99 1 21 Niger 7,582 0.53 41 13 San Marino - - - - São Tomé Nigeria 59,119 0.86 1,568 13 33 0.82 1 21 and Principe North Macedonia 1,447 0.91 90 11 Saudi Arabia 18,537 1.65 2,958 12 Northern 37 3.28 - - Senegal 4,431 0.71 50 12 Mariana Islands Norway 1,411 1.29 981 12 Serbia 5,472 0.92 419 13 Oman 1,910 1.60 154 7 Seychelles 58 1.95 10 18 Pakistan 174,172 1.40 3,675 22 Sierra Leone 1,669 0.55 13 14 Palau - - - - Singapore 2,143 1.56 856 12 Sint Maarten Panama 626 0.96 67 - - - - (Dutch) Appendixes 90 The Global Health Cost of Ambient PM2.5 Air Pollution 91 YLD Cost of Cost of morbidity YLD Cost of Cost of morbidity YLD YLD Study per death morbidity (% of total cost of Study per death morbidity (% of total cost of from PM2.5 from PM2.5 from PM2.5 (US$, millions) health damages) from PM2.5 (US$, millions) health damages) Slovak Republic 2,970 0.82 513 8 Togo 2,437 0.68 25 18 Slovenia 1,065 1.18 311 13 Tonga 6 1.13 0 17 Trinidad Solomon Islands 115 0.80 3 14 299 0.63 30 4 and Tobago Somalia 1,749 0.43 - - Tunisia 5,483 1.11 243 14 South Africa 32,619 1.55 2,310 20 Turkey 45,783 1.69 4,529 14 South Sudan 3,620 0.52 46 13 Turkmenistan 1,417 0.44 103 5 Turks and Spain 9,448 0.77 3,199 9 - - - - Caicos Islands Sri Lanka 16,965 2.31 1,135 33 Tuvalu - - - - St. Kitts and Nevis - - - - Uganda 10,568 0.73 74 13 St. Lucia 34 0.77 2 8 Ukraine 26,015 0.48 716 7 Saint Martin United - - - - 6,732 2.18 1,263 11 (French) Arab Emirates St. Vincent and 18 0.48 1 6 United Kingdom 22,706 0.94 10,834 10 the Grenadines Sudan 14,805 0.78 628 17 United States 164,451 1.76 113,746 18 Suriname 109 0.57 8 7 Uruguay 887 0.75 117 6 Sweden 1,796 1.30 1,078 13 Uzbekistan 9,605 0.47 232 7 Switzerland 3,177 1.56 2,922 16 Vanuatu 63 0.77 2 12 Syrian 5,486 0.81 - - Venezuela, RB 7,319 0.86 - - Arab Republic Tajikistan 2,679 0.58 14 6 Vietnam 78,104 1.94 1,432 18 Tanzania 9,080 0.63 73 9 Virgin Islands (US) 31 0.51 - - West Bank Thailand 65,373 2.57 2,533 16 876 0.68 49 17 and Gaza Timor-Leste 433 1.69 11 38 Appendixes 92 The Global Health Cost of Ambient PM2.5 Air Pollution 93 YLD Cost of Cost of morbidity YLD Study per death morbidity (% of total cost of from PM2.5 from PM2.5 (US$, millions) health damages) Yemen, Rep. 9,531 0.76 160 18 Zambia 3,431 0.60 50 10 Zimbabwe 3,772 0.83 43 14 Source: Annual YLD from PM2.5 is from the GBD 2016 study. The annual cost is estimated using the methodology presented in this annex. Note: Cost of morbidity of PM2.5 is not estimated for some countries and territories due to lack of estimate of YLD from ambient PM2.5 in the GBD 2016 study or absence of GDP per capita in the World Development Indicators Database (http://datatopics.worldbank.org/world-develop- ment-indicators/). World Bank Group 94 The Global Health Cost of Ambient PM2.5 Air Pollution 95 References Dickie, M., and S. Gerking. 2002. “Willingness to Pay for Reduced Morbidity.” Paper presented at the Workshop “Economic Valuation of Health for Environmental Policy: Assessing Alternative Approaches,” Orlando, Florida, March 18–19, 2002. Ezzati, M., A. D. Lopez, A. A. Rodgers, and C. J. Murray. ‎2004‎. Comparative Quanti- fication of Health Risks: Global and Regional Burden of Disease Attributable to Selected Major Risk Factors. Geneva: WHO. Ezziane, Z. 2013. “The Impact of Air Pollution on Low Birth Weight and Infant Mortality.” Alberini, A., and A. Krupnick. 2000. “Cost-of-Illness and Willingness-to-Pay Estimates of Review of Environmental Health 28(2-3): 107–15. the Benefits of Improved Air Quality: Evidence from Taiwan.” Land Economics 76: 37–53. GBD 2010 Risk Factors Collaborators. 2012. “GBD 2010: A Multi-Investigator Collabora- Andrée, P. J. 2020. “Incidence of COVID-19 and Connections with Air Pollution Expo- tion for Global Comparative Descriptive Epidemiology.” Lancet 380 (9859): 2055–58. sure: Evidence from the Netherlands.” Policy Research Working Paper 9221. World Bank, Washington, DC. GBD 2013 Risk Factors Collaborators. 2015. “Global, Regional, and National Compara- tive Risk Assessment of 79 Behavioural, Environmental and Occupational, and Metabolic Bowe, B., Y. Xie, T. Li, Y. Yan, H. Xian, and Z. Al-Aly. 2018. “The 2016 Global and Na- Risks or Clusters of Risks in 188 Countries, 1990–2013: A Systematic Analysis for the tional Burden of Diabetes Mellitus Attributable to PM2.5 Air Pollution.” Lancet Planetary Global Burden of Disease Study 2013.” Lancet 386: 2287–323. Health 2 (7): e301-e312. GBD 2015 Risk Factors Collaborators. 2016. “Global, Regional, and National Compara- Brauer, M., M. Amann, R. T. Burnett, A. Cohen, F. Dentener, M. Ezzati, S. B. Henderson, tive Risk Assessment of 79 Behavioural, Environmental and Occupational, and Metabolic M. Krzyzanowski, R. V. Martin, R. Van Dingenen, A. van Donkelaar, and G. D. Thurston. Risks or Clusters of Risks, 1990–2015: A Systematic Analysis for the Global Burden of 2012. “Exposure Assessment for Estimation of the Global Burden of Disease Attributable Disease Study 2015.” Lancet 388: 1659–724. to Outdoor Air Pollution.” Environ Sci Technol 46 (2): 652–60. GBD 2016 Risk Factors Collaborators. 2017. “Global, Regional, and National Compara- Brauer, M., G. Freedman, J. Frostad, A. van Donkelaar, R. V. Martin, F. Dentener, R. van tive Risk Assessment of 84 Behavioural, Environmental and Occupational, and Metabolic Dingenen, K. Estep, H. Amini, J. S. Apte, K. Balakrishnan, L. Barregard, D. Broday, V. Risks or Clusters of Risks, 1990–2016: A Systematic Analysis for the Global Burden of Feigin, S. Ghosh, P. K. Hopke, L.D. Knibbs, Y. Kokubo, Y. Liu, S. Ma, L. Morawska, J. L. Disease Study 2016.” Lancet 390: 1345–422. Sangrador, G. Shaddick, H. R. Anderson, T. Vos, M. H. Forouzanfar, R. T. Burnett, and A. Cohen. 2016. “Ambient Air Pollution Exposure Estimation for the Global Burden of Heft-Neal, S., J. Burney, E. Bendavid, and M. Burk. 2018. “Robust Relationship Between Disease 2013.” Environ Sci Technol 50 (1): 79–88. Air Quality and Infant Mortality in Africa.” Nature 559(7713): 254–8. Burnett, Richard T., C. Arden Pope III, Majid Ezzati, Casey Olives, Stephen S. Lim, Sumi Landrigan, P. J., R. Fuller, N. J. R. Acosta, O. Adeyi, R. Arnold, N. N. Basu, A. B. Baldé, Mehta, Hwashin H. Shin, Gitanjali Singh, Bryan Hubbell, Michael Brauer, H. Ross An- R. Bertollini, S. Bose-O’Reilly, J. I. Boufford, P. N. Breysse, T. Chiles, C. Mahidol, A. M. derson, Kirk R. Smith, John R. Balmes, Nigel G. Bruce, Haidong Kan, Francine Laden, Coll-Seck, M. L. Cropper, J. Fobil, V. Fuster, M. Greenstone, A. Haines, D. Hanrahan, D. Annette Prüss-Ustün, Michelle C. Turner, Susan M. Gapstur, W. Ryan Diver, and Aaron Hunter, M. Khare, A. Krupnick, B. Lanphear, B. Lohani, K. Martin, K. V. Mathiasen, M. Cohen. 2014. “An Integrated Risk Function for Estimating the Global Burden of Disease A. McTeer, C. J. L. Murray, J. D. Ndahimananjara, F. Perera, J. Potočnik, A. S. Preker, J. Attributable to Ambient Fine Particulate Matter Exposure.” Environmental Health Ramesh, J. Rockström, C. Salinas, L. D. Samson, K. Sandilya, P. D. Sly, K. R. Smith, A. Perspectives: 122 (4): 397–403. Steiner, R. B. Stewart, W. A. Suk, O. C .P. van Schayck, G. N. Yadama, K. Yumkella, and M. Zhong. 2018. “The Lancet Commission on Pollution and Health.” Lancet 391 Carey, I. M., H. R. Anderson, R. W. Atkinson RW, S. D. Beevers, D. G. Cook, D. P. Stra- (10119): 462–512. chan, D. Dajnak, J. Gulliver, and F. J. Kelly. 2018. “Are Noise and Air Pollution Related to the Incidence of Dementia? A Cohort Study in London, England.” BMJ Open 8: e022404. Larsen, B. 2014. “Benefits and Costs of the Air Pollution Targets for the Post-2015 Devel- opment Agenda.” Air Pollution Assessment Paper, Post-2015 Consensus Project, Copen- Cropper, M., and W. Oates. 1992. “Environmental Economics: A Survey.” Journal of hagen Consensus Center, Copenhagen. https://www.copenhagenconsensus.com/sites/ Economic Literature 30: 675–740. default/files/air_pollution_assessment_-_larsen.pdf. World Bank Group 96 The Global Health Cost of Ambient PM2.5 Air Pollution 97 Larsen, B. 2017. “Global Ambient Air Pollution: New Perspectives on Exposure, Health Salomon, J. A., J. A. Haagsma, A. Davis, C. M. de Noordhout, S. Polinder, A. H. Have- Effects and Costs – Global, Regional and National Estimates for 2015.” Report prepared laar, A. Cassini, B. Devleesschauwer, M. Kretzschmar, N. Speybroeck, C. J. Murray, and for the World Bank, Washington, DC. T. Vos. 2015. “Disability Weights for the Global Burden of Disease 2013 Study.” Lancet Global Health 3: e712–23. Lim, S. S., T. Vos, A. D. Flaxman, et al. 2012. “A Comparative Risk Assessment of Burden of Disease and Injury Attributable to 67 Risk Factors and Risk Factor Clusters in 21 Re- Sánchez-Triana, E., S. Enriquez, B. Larsen, P. Webster, and J. Afzal. 2015. Sustainability gions, 1990–2010: A Systematic Analysis for the Global Burden of Disease Study 2010.” and Poverty Alleviation: Confronting Environmental Threats in Sindh, Pakistan. Lancet 380: 2224–60. Directions in Development. Washington, DC: World Bank. Lindhjem, H., S. Narvud, N. A. Braathen, and V. Biausque. 2011. “Valuing Mortality Risk Shaddick, G., M. L. Thomas, A. Green, M. Brauer, A. van Donkelaar, R. Burnett, H. Chang, Reductions from Environmental, Transport, and Health Policies: A Global Meta-Analysis A. Cohen, R. Van Dingenen, C. Dora, S. Gumy, Y. Liu, R. Martin, L. A. Waller, J. West, J. of Stated Preference Studies.” Risk Analysis 31 (9): 1381–407. V. Zidek, and A. Prüss-Ustün. 2018. “Data Integration Model for Air Quality: A Hierarchi- cal Approach to the Global Estimation of Exposures to Ambient Air Pollution.” Journal Mehta, S., H. Shin, R. Burnett, T. North, and A. Cohen. 2013. “Ambient Particulate Air of the Royal Statistical Society. Series C, Applied Statistics 2018 67: 231–53. Pollution and Acute Lower Respiratory Infections: A Systematic Review and Implications for Estimating the Global Burden of Disease.” Air Qual Atmos Health 6: 69–83. Shin, H., A. Cohen, C.A. Pope III, M. Ezzati, S. S. Lim, B. Hubbel, and R. T. Burnett. 2013. “Critical Issues in Combining Disparate Sources of Information to Estimate the Global Narain, U., and C. Sall. 2016. Methodology for Valuing the Health Impacts of Air Pollu- Burden of Disease Attributable to Ambient Fine Particulate Matter Exposure.” Working tion: Discussion of Challenges and Proposed Solutions. Washington, DC: World Bank. Paper prepared for Harvard Center for Risk Analysis “Methods for Research Synthesis: A Cross-Disciplinary Workshop,” October 3, 2013. Navrud, S., and H. Lindhjem. 2010. Meta-Analysis of Stated Preference VSL Studies: Fur- ther Model Sensitivity and Benefit Transfer Issues. (ENV/EPOC/WPNEP(2010)10/FINAL). Shin J., J. Y. Park, and J. Choi. 2018. “Long-Term Exposure to Ambient Air Pollutants Environment Directorate. Paris: OECD. and Mental Health Status. A Nationwide Population-Based Cross-Sectional Study.” PLOS One 13 (4): e0195607. OECD (Organisation for Economic Co-operation and Development). 2012. Mortality Risk Valuation in Environment, Health, and Transport Policies. Paris: OECD. Shindell, D., J. C. I. Kuylenstierna, E. Vignati, R. van Dingenen, M. Amann, Z. Klimont, S. C. Anenberg, N. Muller, G. Janssens-Maenhout, F. Raes, J. Schwartz, G. Faluvegi, L. OECD (Organisation for Economic Co-operation and Development). 2016. Pozzoli, K. Kupiainen, L. Höglund-Isaksson, L. Emberson, D. Streets, V. Ramanathan, K. The Economic Consequences of Outdoor Air Pollution. Paris: OECD. Hicks, N. T. K. Oanh, G. Milly, M. Williams, V. Demkine, and D. Fowler. 2012. “Simulta- Ostro, B., J. V. Spadaro, S. Gumy, P. Mudu, Y. Awe, F. Forastiere, and A. Peters. 2018. neously Mitigating Near-Term Climate Change and Improving Human Health and Food “Assessing the Recent Estimates of the Global Burden of Disease for Ambient Air Pollu- Security.” Science 335: 183–89. tion: Methodological Changes and Implications for Low- and Middle-Income Countries.” Van Donkelaar, A., R. Martin, M. Brauer, and B. Boys. 2015. “Use of Satellite Observa- Environmental Research 166: 713–25. tions for Long-Term Exposure Assessment of Global Concentrations of Fine Particulate Pope, C. A. III, R. T. Burnett, D. Krewski, M. Jerrett, Y. Shi, E. E. Calle, and M. J. Thun. Matter.” Environmental Health Perspectives 123: 135–43. 2009. “Cardiovascular Mortality and Exposure to Airborne Fine Particulate Matter and Van Donkelaar, A., R. V. Martin, M. Brauer, N. C. Hsu, R. A. Kahn, R. C. Levy, A. Lyapus- Cigarette Smoke: Shape of the Exposure-Response Relationship.” Circulation tin, A. M. Sayer, and D. M. Winker. 2016. “Global Estimates of Fine Particulate Matter 120: 941–48. Using a Combined Geophysical-Statistical Method with Information from Satellites, Pope, C. A. III, R. T. Burnett, M. J. Thun, E. E. Calle, D. Krewski, K. Ito, and G. Thurston. Models, and Monitors.” Environmental Science and Technology 50: 3762–72. 2002. “Lung Cancer, Cardiopulmonary Mortality, and Long-Term Exposure to Fine Par- WHO (World Health Organization). 2009. Estimated Deaths and DALYs Attributable to ticulate Air Pollution.” Journal of the American Medical Association 287: 1132–41. Selected Environmental Risk Factors, by WHO Member States, 2004. http://www.who. Pope, C. A. III, R. T. Burnett, M. Turner, A. Cohen, D. Krewski, M. Jerrett, S. M. Gapstur, int/quantifying_ehimpacts/national/countryprofile/intro/en/index.htm. and M. J. Thun. 2011. “Lung Cancer and Cardiovascular Disease Mortality Associated WHO (World Health Organization). 2016. Ambient Air Pollution: A Global Assessment with Ambient Air Pollution and Cigarette Smoke: Shape of the Exposure-Response of Exposure and Burden of Disease. Geneva: WHO. Relationships.“ Environmental Health Perspectives 119 (11): 1616–21. World Bank Group 98 The Global Health Cost of Ambient PM2.5 Air Pollution 99 Wilson, C. 2003. “Empirical Evidence Showing the Relationships Between Three Approaches for Pollution Control.” Environmental and Resource Economics 24: 97–101. World Bank and IHME (Institute for Health Metrics and Evaluation). 2016. The Cost of Air Pollution: Strengthening the Economic Case for Action. Washington, DC: World Bank. World Bank. 2020a. Air Pollution and Climate. How Many Stones to Hit Both Targets? Washington, DC: World Bank. World Bank. 2020b. Implications of Exposure to Dust on the Burden of Disease from Air Pollution. Washington, DC: World Bank. World Bank. 2020c. Are All Air Pollution Particles Equal? – How Constituents and Sourc- es of Fine Air Pollution Particles (PM2.5) Affect Health. Washington, DC: World Bank. Xu, X., S. U. Ha, and R. Basnet. 2016. “A Review of Epidemiological Research on Adverse Neurological Effects of Exposure to Ambient Air Pollution.” Frontiers in Public Health, 4: 157. Zhang, X., X. Chen, and X. Zhang. 2018. “The Impact of Exposure to Air Pollution on Cognitive Performance.” Proceedings of the National Academy of Sciences 115 (37): 9193–97. Appendixes 100 WWW.WORLDBANK.ORG