Copenhagen Consensus Center
Home Menu

Post-2015 Consensus: Air Pollution Assessment, Larsen

Summary of Targets from the Paper

TARGET Cost $B p.a. Benefit $B p.a. Benefit for Every Dollar Spent
50% of those using unimproved cookstoves switch to improved cookstoves $5 $52 $10
50% of those using unimproved cookstoves switch to LPG cookstoves $68 $126 $2
100% of those using unimproved cookstoves switch to LPG cookstoves $137 $297 $2
Outdoor particulate matter 2.5 does not exceed 35 µg/m3 $50 $13 $0.3
Outdoor particulate matter 2.5 does not exceed 25 µg/m3 $98 $35 $0.4
Outdoor particulate matter 2.5 does not exceed 15 µg/m3 $190 $78 $0.4
Outdoor particulate matter 2.5 does not exceed 10 µg/m3 $304 $130 $0.4

Summary

The last two decades have seen a large body of evidence of substantial health effects for long term exposure to air pollution – especially fine particulate matter – be it in the form of outdoor ambient air pollution (AAP) or household air pollution (HAP) from the use of solid fuels.  There are compelling arguments that air pollution should feature in a new set of post-2015 development goals.

Global health effects and exposure to air pollution

Health effects

Nearly 6 million deaths were attributed to AAP and HAP in 2010.  This is more than from alcohol and drugs, about the same as from active and passive tobacco smoking and four times more than from child and maternal under-nutrition. Of 67 risk factors assessed, it is only surpassed by total dietary risk factors and high blood pressure, of which the latter is influenced by air pollution, tobacco smoking and diet.  In 2012, WHO updated the estimate to 7 million deaths, with AAP associated with 3.7 million and HAP from solid fuel use associated with 4.3 million.

An integrated PM2.5 exposure-response model (IER) was developed to estimate overall health effects. The exposure-response relationships in the IER model are highly non-linear with declining marginal relative risks of health outcomes at higher PM2.5 exposure levels. 

Ambient air pollution exposure

Nearly 90% of the world’s population lived in areas with ambient outdoor PM2.5 concentrations exceeding WHO’s Air Quality Guideline of 10µg/m3 (annual average) in 2005; for many people, exposure is much higher. The highest average concentrations (weighted by population) were in a belt extending from western sub-Saharan Africa through the Middle East, South and East Asia and high income Asia Pacific countries. In South Asia, 99% of the population was exposed to PM2.5 at average concentrations above10µg/m3, while the figure in Western Europe is 92%. Nearly one third of the global population was exposed to fine particulates at a concentration above 35µg/m3, the WHO’s Level 1 Interim Target. Most of the locations where this level was exceeded were in Asia.

Household air pollution exposure

The predominant source of HAP, in terms of global health effects, is the use of solid fuels by households for cooking and other purposes.  About 41% of the world’s population – 2.8 billion - used mainly solid fuels for cooking in 2010. As a proportion of the global population, this has declined, but the absolute number has increased slightly, to nearly 2.9 billion today, mostly in China, India and other Southern Asian countries, sub-Saharan Africa and South East Asia.

Globally, about 15% of the urban population uses solid fuels while 67% of the rural population does so. Wood is the most widely used fuel for cooking, straw and dung are widely used in a few countries and use of coal is quite widespread in China and Mongolia for both cooking and heating. Concentrations of PM2.5 in homes burning wood, dung or straw are often several hundred micrograms per cubic metre; coal gives lower levels of particulates, but these tend to be more carcinogenic.

Use of improved biomass stoves with a chimney or a hood reduces particulate levels from several hundred to 75-125µg/m3, but the health benefit is only perhaps 20-30% because of the highly non-linear exposure-response relationship. Also, vented smoke from improved stoves contributes to outdoor air pollution in the community. For a community-wide improvement, the use of modern energy sources such as LPG is needed. The reduction in health effects for individual households may be of the order of 40-50%, but this can rise to over 65% if all households if the entire community stops using solid fuels.

Targets

Domains of targets

Reductions in health effects

The advantage of targeting a reduction in health effects of air pollution at regional or national levels is the flexibility this provides in how to achieve the targets. However, focusing on locations where reductions can be achieved at a lower cost may result in a socially unacceptable degree of inequity. Such targets are also difficult to monitor, and progress is therefore difficult to verify and subject to disagreements over evidence base and methodologies.

Improved air quality

Targets of this sort for AAP are easier to monitor and verify if good monitoring equipment is available, although this is currently not the case for the majority of cities in low and middle income countries. Air quality targets may also be economically inefficient if they are nationally or regionally uniform. Monitoring of improvements in households on any scale is costly and impractical.

Reductions in sources of pollution

The advantage of targeting sources of pollution is the relative ease with which many sources can be monitored and costs of achieving the targets be estimated. However, for AAP the exposure and therefore health effects vary greatly between types of pollution source and location. Such targets are much more practical for households, where types of fuel and stoves can be easily monitored.

“Zero” targets

“Zero” targets are targets that would eliminate outdoor and indoor air pollution (PM2.5), or at least bringing anthropogenic PM2.5 concentrations outdoors and indoors below the level known to cause health effects (about 5.8µg/m3).  In practice, this is impossible to achieve everywhere because not all sources are anthropogenic (desert dust, for example). Achieving such a target where possible would mean the complete elimination of solid fuels in the home and other external sources of pollution. Currently, only a small number of locations, mainly in small, pristine areas of Australia, Canada, New Zealand and the United States meet the 5.8µg/m3 standard.

Selected targets

Targets for ambient air pollution

As discussed above, air quality targets for AAP are easier to measure and verify and these will be assessed in more detail in this paper. A reasonable target for most high income countries in the Americas, Europe and Asia/Pacific would be the annual AQG of 10µg/m3.  The interim target of 15-25µg/m3 may be the initial aim for Latin America and the Caribbean and much of Eastern Europe.  The interim target of 25-35 µg/m3 may initially be more realistic for many of the low and middle income countries in Western Africa and Asia.

Targets for household air pollution

The most attractive targets for household air pollution centre on stoves and cooking fuels and these will be assessed in more detail in this paper. To achieve the maximum benefits per dollar spent on household energy and stove interventions, all households would need to participate, and thus achieve a “solid fuel use free” community or, alternatively, an “unimproved stove free” community.

Benefits and costs of household air pollution control

Targets

The first and second interim targets (IT-1 and IT-2), are for a 50% adoption rate of improved cooking stoves and LPG stoves respectively among households currently using biomass or coal. These interim targets can be pursued concurrently. There is also a longer-term final target (FT) of 100% adoption of LPG or other clean cooking (and heating) options.

The interim targets are expected to reduce personal PM2.5 exposure from an average of 250µg/m3 to 100µg/m3 with adoption of improved stoves and to 50µg/m3 with adoption of LPG stoves.  Achievement of the final target is expected to reduce PM2.5 exposure to less than 25µg/m3

For the purposes of this analysis, we can define a range of exposure levels as below:

Level 1 – Biomass largely used on open fire or in unimproved stove; 250µg/m3

Level 2 – Chimney stove or other improved biomass or coal stove with community pollution;          100µg/m3

Level 3 – Mix of gas and biomass or coal in chimney stove or other improved stove with community pollution;            75µg/m3

Level 4 – Gas (e.g., LPG) with community pollution; 50µg/m3

Level 5 – Gas (e.g., LPG) with limited community pollution; 25µg/m3

Level 6 – Gas (e.g., LPG) with very limited community pollution; < 7.3µg/m3

Level 1 is the average current exposure level of the 2.8 billion people who use solid fuels for cooking and heating. The three targets selected for assessment of benefits and costs correspond to exposure levels 2, 4 and 5.

Health effects

Because the response to reduced exposure is highly non-linear, only one-third of the health benefit is realized by a reduction from 250 to 100µg/m3, with two-thirds coming from the further reduction to 25µg/m3. Even at this level, one-third of the baseline effects of household air pollution remain.

An estimated 3.5 million people died and 19.7 billion disease days occurred globally in 2012 from household air pollution.  Almost 900,000 deaths and 4.8 billion disease days could be avoided annually if all households used an improved biomass or coal stove (exposure level 2; 100µg/m3 of PM2.5).  If all households used LPG or other clean fuels, over 2.3 million deaths and 12.8 million disease days could be avoided annually (exposure level 5; 25µg/m3 of PM2.5). 

Monetized values of health effects

Two alternative measures are used. The first uses the value of a statistical life (VSL), equivalent to 50 times GDP per capita, and the second uses a uniform value of $1,000 or $5,000 per life year. Morbidity is valued either as 50% of daily wages or using the uniform values for a year of life.

The global cost of household solid fuel use in 2012 is estimated at $646 billion, applying VSL for mortality and a fraction of wage rates for morbidity. Using the uniform values for life-years, the cost is $111-555 billion.

Non-health benefits

Non-health benefits of interventions included in this paper are fuel and cooking time savings.  Fuel savings are valued as the time that households spend on fuel collection, and time is valued at 50% of wage rates.

Costs of pollution control options

Improved biomass and coal stoves vary widely in price. A price of $30 is applied to most regions where heating is uncommon. In Latin America and the Caribbean, where requirements are different, the cost is taken as $60, and in China a price of $115 is applied, taking account also of heating requirements. The price of an LPG stove is assumed to be $60 in all regions. 30-40 kg of LPG is assumed to be used per person per year, costing $1.3 per kilo.

Benefit-cost ratios

Four cases are considered. Cases 1 and 2 are where Interim Targets 1 and 2 are met; case 3 is for meeting the Final Target. Case 4 is for comparison with case 2, to demonstrate the effect of community pollution.

Global benefits

The global benefits of reaching the initial targets for ICSs and LPG are $120-270 billion per year, depending on the valuation method used.  The global cost of ICS is estimated at nearly $20 billion, with an annualized cost of $5 billion.  The global cost of LPG stoves is also estimated at about $20 billion.  The global cost of LPG fuel is estimated at a little over $60 billion per year for the initial target.

Global benefit-cost ratios

Globally, the benefit-cost ratio of improved biomass or coal stoves (ICS) is in the range of 6-16. The BCRs of LPG adoption (Cases 2 & 3) are much smaller than the BCR for ICS, but the net benefits are greater for LPG, suggesting that LPG should be promoted among those that can afford it.  The BCR of switching to LPG from UCS (initial target) or from ICS (final target) are quite similar and all greater than one.  The BCR for case 4 is 15-25% higher than case 2, reflecting the improvement from reduction of community pollution.

Regional benefit-cost ratios

A similar picture emerges on a regional level. However, valuing health benefits using either VSL or $1,000 per day gives BCRs slightly below 1 for sub-Saharan Africa.

Benefits and costs of outdoor ambient air pollution

Targets

The air quality guideline from WHO is for an annual PM2.5 value of 10µg/m3; there are also interim targets of 35, 25 and 15µg/m3. Reflecting the large differences in regional ambient concentrations of fine particulates, three regional target groups are proposed, with all regions progressed towards the AQG limit in time:

1 – East Asia: IT-1 (35µg/m3)

2 – South Asia, Middle East and North Africa, High Income Asia Pacific, Western sub-Saharan Africa, Central Asia, Central Europe, South East Asia: IT-2 (25µg/m3)

3 – Western Europe, High Income North America, Other sub-Saharan Africa, Eastern Europe, Latin America and Caribbean, Australasia, Oceania: IT-3 (15µg/m3)

Health effects

Health impacts are greatest in percentage terms for regions furthest away from the final target. Reaching the ACQ target would give an 80%, 66% and 39% reduction in health impacts for groups 1, 2 and 3 respectively, but as much as one-third of the global health effects remain after all regions have achieve the 10µg/m3 target. An estimated 3.3 million people died and 9.4 billion disease days occurred globally in 2012 from PM2.5 ambient air pollution (AAP).  Almost 2.2 million deaths and 6.3 billion disease days could be avoided annually if all regions reached the annual PM2.5 AQG of 10µg/m3.

Monetized values of health effects

The global cost of outdoor ambient PM2.5 exposure in 2012 is estimated at $1.7 trillion, applying VSL for mortality and a fraction of wage rates for morbidity.  About $0.9 trillion is in high income regions, although they only account for 11% of global deaths. About $637 billion of the cost is in developing regions in Asia, Africa, Latin America and Oceania, which between them account for 81% of the deaths.

The global cost of outdoor ambient PM2.5 is estimated at $78-388 billion in the same year when applying $1,000 to $5,000 per DALY.

Pollution control options and costs

There are many sources of outdoor air pollution, not all of which can be covered here. However, we look at two policy options and the costs of abatement from four sources of PM2.5.

Energy subsidies

World energy subsidies contribute to energy waste and pollution.  Energy consumption subsidies averaged over $400 billion per year during 2007-2010 and $520-540 billion per year during 2011-2012 according to the International Energy Agency (IEA).  Eliminating these subsidies would provide economic efficiency gains and reduce pollution while actually saving money.

Taxation policies

While direct tax instruments for PM2.5 abatement are often difficult to design, indirect instruments can provide PM2.5 emission reductions at lower cost to society than regulatory, command-and-control options. These instruments include fuel taxes, vehicle taxes, and tax rebates on PM2.5 control technology.

Household use of solid fuels

Household use of solid fuels does not only cause serious air pollution in the immediate household environment, but also contributes to outdoor pollution. Abatement could be via use of improved cooking stoves or switching to LPG. LPG is more expensive but also far more effective in reducing pollution. Taking into account fuel savings, moving to improved biomass stoves may actually save money, while the net cost of avoiding the emission of one ton of PM2.5 by switching to LPG can be up to $50,000 in East Asia.

Solid waste management

Improved municipal solid waste management can reduce waste burning and hence air pollution. The costs translate to $10-12,000 per ton of PM2.5  abatement from avoided burning of waste in the lowest income regions of the world, rising to $24-28,000 in East Asia and Latin America.

Fuel quality

The majority of primary PM2.5 emissions from vehicle fuel combustion come from diesel vehicles, which accounted for 44% of global road transport fuel use in 2012. Fuel quality makes a big difference, and In recognition of the road transport sector’s contribution to air pollution, there is globally a major push for ultra-low sulfur (<50ppm) diesel for road vehicles.  Depending on the starting level, reducing sulfur levels to 50ppm gives abatement costs of $14-29,000 per ton of PM2.5.

Road vehicle technologies

There are various ways to reduce road vehicle emissions. The option assessed here is retrofitting of in-use diesel vehicles with diesel particulate filters (DPFs).  The cost per ton of PM2.5 abated is in the range $30-100,000 for relatively high usage vehicles used mainly in cities. The cost is relatively high, but the technology is also highly effective.

Benefit-cost ratios

Global benefits

Global benefits of reaching the final target, i.e., annual AQG of 10µg/m3 of PM2.5, are in the range $52 – 971 billion per year and about 87% of the avoided deaths are in low- and middle-income countries.

Benefits per ton of PM2.5 emissions

Benefits per ton of PM2.5 emissions reductions are very location-specific and will depend on PM2.5 ambient concentrations and intake fractions. Using VSL as a cost basis, the values range from as low as $5,800 in Oceania to nearly one hundred times that in North America. Using a DALY value of $5,000, the range is from $3,950 to $29,000, with the poorest regions – sub-Saharan Africa, South Asia and South East Asia – having generally high figures, along with Eastern Europe.

Regional benefit-cost ratios

As an estimated 81% of global deaths from outdoor ambient PM2.5 occur in low- and middle-income countries, the benefit-cost analysis in this paper concentrates on these regions.

Benefit-cost ratios of controlling PM2.5 emissions to the outdoor environment from household use of solid fuels are larger than the BCRs of other abatement options assessed. Regional BCRs of improved biomass stoves range from 1.3 to 23.3. For conversion to LPG in East Asia, BCRs are generally greater than one for health benefits valued on the basis of VSL or a DALY of $5,000.

Regional BCRs of improved municipal solid waste management to minimize uncontrolled burning of waste range from 0.13 to 2.88.  Regional BCRs of ultra-low sulfur diesel (ULSD) fuel for road vehicles range from 0.1 to 5.0.  For retrofitting in-use diesel vehicles with diesel particulate filters (DPFs), benefit-cost ratios range from 0.02 to 1.47. 

Conclusions

PM2.5 air pollution is a major cause of premature death and disease. The benefits of controlling indoor air pollution from solid fuel cooking are much greater than the costs, and reducing solid fuel use also improves outdoor air quality, particularly in Asia. The often low to moderate BCRs suggest that outdoor PM2.5 abatement in especially low-income countries should be selective and well-targeted.  They also suggest that a high priority is to control PM2.5 emissions from household use of solid fuel, for both indoor and outdoor exposure reduction.