Ambient air pollution¹ is a major contributor to human mortality and morbidity [1] [2] [3] . Epidemiological studies have established robust causal relationships between long-term exposure to PM_2.5―particulate matter with aerodynamic diameter of less than 2.5 microns―and premature deaths related to heart disease, stroke, respiratory diseases, and lung cancer, thereby substantially reducing life expectancy [4] . Exposure to PM_2.5 also causes morbidity, resulting in problems such as cases of chronic bronchitis, hospital admissions, work loss days, restricted activity days, and acute lower respiratory infections in children [5] .

In 2015, ambient PM_2.5 was the fifth-ranked mortality risk factor², accounting for 7.6 percent of total global mortality [6] , causing 4.2 million deaths and 103.1 million lost years of healthy life [7] . Ischemic heart disease and cerebro-vascular diseases accounted for 57 percent of the deaths. China and India stood out with the highest air pollution-related mortality, each exceeding 1 million deaths. In addition, globally, household air pollution from the use of solid fuels (e.g. coal, wood, dung) for cooking and heating was the tenth-ranked mortality risk factor in 2015, being responsible for 2.8 million deaths and 85.6 million lost years of healthy life [6] [7] .

Recent efforts estimated that premature mortality cost the global economy about US$225 billion in lost labor income in 2013, or about US$5.1 trillion in welfare losses [8] . In the Middle East and North Africa region, an estimated 125,000 lives were lost in the same year to diseases associated with ambient and household air pollution; this corresponded to welfare losses of about US$154 billion, or 2.2 percent of the regional gross domestic product (GDP).

This paper estimates the economic cost of air pollution in Morocco. It is part of a broader study carried out by the World Bank, which aims to estimate the overall cost of environmental degradation in the country [9] . The paper uses the most updated methodology to value in monetary terms the impact of PM_2.5 on people’s health; puts the results into a broader perspective of the country’s overall cost of environmental degradation; and identifies the most affected areas and groups in Morocco. The analysis has been carried out during 2015-2016, and is based on secondary information collected from Government institutions, national statistics and scientific literature.

Between 2004 and 2014, Morocco experienced strong economic growth, reflected by an overall increase in per capita GDP of 34 percent [10] . During the same period, population grew by an annual 1.25 percent on average; while urban population, which is mostly concentrated on coastal areas, increased even faster (2.1 percent per year) [11] . Coastal cities are also home for most economic activities, such as energy and industry, and have experienced a rapid growth of road traffic [12] . These activities generated a rapid increase in emissions of local and global air pollutants [13] .

Ambient air pollution. The country started to monitor air quality in 1997.The first efforts were carried out by the Ministry of Sustainable Development in the city of Rabat using a mobile laboratory, and were followed by measurements conducted by the Directorate of National Meteorology (Direction de la météorologienationale, DMN) in Greater Casablanca. Currently, the DMN manages the national air quality-monitoring network, which has 29 fixed and 3 mobile stations covering 15 cities [14] . These stations monitor the ambient concentration of several particulates, such as nitrogen dioxide (NO₂), particulate matter with diameter less than 10 microns (PM₁₀), ozone (O₃), and carbon monoxide (CO). Aware of the harmful effects of local pollutants on people’s health, the country has already conducted several studies which valued the impacts of ambient air pollution on people’s health in Casablanca, Mohammedia, and Fès [15] [16] .

Household air pollution resulting from the use of solid fuels for cooking and heating is also associated with substantial health effects [4] . Generally, burning solid fuels (wood, charcoal, agricultural residues) in households causes emissions of PM_2.5 and other pollutants harmful to human health. Other fuels (e.g. liquefied petroleum gas, biogas) are cleaner and generate less PM_2.5. In Morocco, no information is available on PM_2.5 concentrations at the level of rural households. However, per capita energy consumption was estimated at 0.54 tons of oil equivalent (toe) in 2012, which is very low compared to the world average (1.9 toe/capita) and that of Africa (0.67 toe/inhabitant) [17] . Wood and coal accounted for 25 percent of total energy consumption in the same year, according to communications with the Department of Energy and Mines³.

This section estimates the impact of exposure to ambient PM_2.5 on human mortality and morbidity, using 2014 as the year of reference. The valuation is based on four steps, presented below.

Step 1. Measure the PM_2.5 concentration. The Ministry of Sustainable Development monitors only particles with a diameter of less than 10 micrometers (PM₁₀)⁴. Several measurement stations located in Agadir, Benslimane, Casablanca, El Jadida, Fès, Mohammedia, Khouribga, Marrakech, Safi, Salé, and Tangier provided daily data for the period 2012-2015. These data are used to estimate the annual average PM_2.5 concentrations for each city, as follows:

1) The annual average PM₁₀ concentration for each station is quantified based on daily monitoring data.

2) For each station, the annual PM₁₀ concentration is converted to PM_2.5 concentration, using a conversion factor of 0.4 [19] .

3) At the city level, PM_2.5 concentration is estimated as the average of concentration data provided by all measuring stations located in that city. For the largest cities (Casablanca and Marrakech), the annual PM_2.5 concentration is estimated as a population-weighted average of the PM_2.5 concentration at each station.

The results indicate that average PM_2.5 concentrations vary widely, from 3 μg/m³ in Safi to as high as 22 μg/m³ in Tangier (Table 1).

The following steps estimate the health impacts for the cities with PM_2.5 concentrations above the World Health Organization (WHO) air quality standard⁵ of 10 μg/m³. These are Tangier, Marrakech, Casablanca, Mohammedia, Settat, Fès, Benslimane, and Khouribga.

Step 2. Estimate the population exposed to PM_2.5. Data on the percentage of total population exposed to pollution are not available for any monitoring station in Morocco. However, as urban transport generates much of the ambient air pollution, the paper assumes that the entire population of each city is affected by the average PM_2.5 concentration calculated at the previous step (Table 1).

Step 3. Quantify the health impacts of exposure to PM_2.5. Several epidemiological studies revealed strong correlations between long-term exposure to PM_2.5 and premature mortality [20] [21] [22] [23] [24] . In particular, recent research associated PM_2.5 exposure with mortality related to four diseases in adults

Source: [11] for population census, DMN for PM₁₀ concentration per monitoring station (2012-2015). * Estimate based on daily measurements of PM₁₀ concentration.

(ischemic heart disease, stroke, chronic obstructive pulmonary disease, and lung cancer) and acute lower respiratory tract infections in children. These relationships, called integrated exposure-response functions, were used to estimate mortality for each health endpoint, age group, and PM_2.5 concentration in each city [4] . The results show that ambient air pollution was responsible for about 2,200 deaths in 2014. Nearly 50 percent of adult deaths originate in Casablanca, followed by Marrakesh and Tangier, mainly due to ischemic heart disease, stroke and lung cancer (Figure 1). Overall, more than 70 percent of deaths result from ischemic heart disease and stroke.

Figure 2 presents the distribution of premature mortality estimate by age

group and disease. Adults over the age of 55 are at risk of premature mortality due mainly to ischemic heart disease and strokes. Children under five years old are also vulnerable to premature mortality due to acute lower respiratory tract infection.

It should be noted that the Institute of Health Metrics and Evaluation (IHME) has estimated mortality in Morocco at 6,000 deaths for the same year, based on a combination of ground and satellite measurements of PM_2.5 ambient concentration (http://ihmeuw.org/3ts8). Two reasons explain the difference in results: (i) the IHME estimates cover the whole country, while the above estimate is conducted for eight cities only; and (ii) the IHME estimates are partly based on satellite data, which are less accurate than ground-based measurements. Therefore, this paper considers mortality in Morocco to be in the range provided by the two estimates, namely between 2,200 and 6,000 deaths.

Step 4. Estimate the health impacts of exposure to PM_2.5. The cost of mortality is estimated based on the concept of the Value of Statistical Life (VSL). It has been widely used in environmental economics literature to reflect people’s willingness to pay for a reduction in mortality risk [8] [25] [26] . The VSL for Morocco was estimated at about US$191,500 (Box 1). Accordingly, the cost of premature mortality (between 2,200 and 6,000) caused by ambient air pollution ranges between US$420 million and US$1.15 billion (a).

The cost of morbidity includes resource costs (i.e. financial costs for avoiding, protecting, or treating pollution-associated illnesses), opportunity costs (i.e. indirect costs from the loss of time for work and leisure), and disutility costs (i.e. cost of pain, suffering, or discomfort). The literature assessing causal relationships between exposure to PM_2.5 and morbidity is much more limited than that for mortality⁶ [5] . So far, no commonly accepted method has been developed to value the overall cost of morbidity due to air pollution [32] ⁷.

However, results of studies conducted in several OECD countries indicate that morbidity costs can be roughly approximated to 10 percent of mortality costs [5] [30] [32] [33] . In the absence of surveys on the willingness to pay to avoid pollution-related illnesses in Morocco, morbidity costs are estimated at 10 percent of mortality costs, i.e. between US$42 million and US$115 million (b).

Adding up the costs of mortality and morbidity (a + b), the total loss due to ambient air pollution ranges between US$462 million and US$1.26 billion, with an average of US$863 million (Table 2).

This section estimates the cost of household air pollution, through the impact of using solid fuel for cooking and heating in rural households. A similar step-by-step approach is used:

Step 1. Measure PM_2.5 concentration. PM_2.5 concentrations in rural households using solid fuel for cooking vary considerably, depending on the location of the kitchen (e.g. indoors or outdoors), method and duration of cooking, ventilation practices, etc. Concentrations of PM_2.5 often reach several hundred μg/m³

Source: authors’ calculation. Note: * except for the last column. The totals may not add up exactly due to rounding.

in the kitchen and more than 100 μg/m³ in the rest of the household⁸.

No measure of PM_2.5 concentration was found for Morocco’s rural households; however, the WHO has compiled a global database of 154 studies on measures of household air pollution. Although the database does not provide PM_2.5 concentrations in any North African country, it contains a few relevant results. For example, [34] report indoor PM_2.5 concentrations in some areas of India ranging from about 160 μg/m³ in living areas to about 600 μg/m³ in kitchens; and [35] measured concentrations of approximately 100 μg/m³ in outdoor terraces with open fire cooking in rural Mexico following adoption of improved wood stoves. As wood stoves or ovens are widely used in Morocco for cooking [36] , this paper conservatively assumes an annual average PM_2.5 concentration of 100 μg/m³ in rural households that use solid fuel for cooking.

Step 2. Estimate the population exposed to PM_2.5. Household exposure to PM_2.5 emitted by combustion of solid fuels depends on the activity patterns inside the household. A recent study [17] indicated that 20 percent of rural households used wood for cooking in 2010 and projected a reduction to 5 percent by 2040. Applying this trend over time resulted in about 18 percent of rural households using wood for cooking in 2014. Applying this proportion to the total rural population of 13.4 million [11] , the total population exposed to household PM_2.5 is estimated at 2.4 million people in 2014.

Step 3. Estimate the health impacts of exposure to PM_2.5. The same integrated exposure-response functions are applied as in Section 3. The results show that household air pollution was responsible for about 1,350 deaths in 2014, of which nearly 90 percent are caused by ischemic heart disease, stroke, or acute lower respiratory tract infections. Figure 3 shows that the oldest (over 55 years) and the youngest (less than 5 years old) are the groups most affected by household air pollution.

Step 4. Estimate the health impacts of exposure to PM_2.5. Using a VSL of US$191,500 (Box 1), the cost of premature mortality (1,350 deaths) caused by household air pollution is estimated to be between US$248 million and US$271 million. In addition, morbidity costs are valued at 10 percent of mortality cost, i.e. between US$25 million and US$27 million. Adding these values, the total health cost due to household air pollution ranges between US$273 million and US$298 million, with an average of US$285 million (Table 2).

The total health cost of air pollution is estimated between US$734 million and US$1.6 billion. This corresponds to an average of US$1.14 billion, or 1.05 percent of the country’s GDP in 2014 (Table 2). Ambient air pollution dominates the total cost (75 percent of the total), primarily as a result of high exposure to

ambient PM_2.5 in urban areas such as Casablanca, Tangier, and Marrakesh. Household air pollution is a significant problem for the 18 percent of the rural households that use solid fuel for cooking.

It should be noted that the above estimates are subject to several limitations. From a methodological point of view, mortality estimates refer only to the impacts of PM_2.5 on five respiratory diseases, for which robust causal relationships have been established in the literature; while those related to morbidity are not correlated with any specific diseases. From an empirical viewpoint, data related to household air pollution are either lacking (e.g. indoor PM_2.5 concentration) or partial (e.g. types of stoves and cooking practices), which imposed reliance on information from other areas.

This analysis is part of a broader study conducted by the World Bank, which estimated the overall cost of environmental degradation in Morocco [9] . It addressed the degradation related to several natural resources (categories): water, air, agricultural land, waste management, coastal zones, and forests. The results showed that the total cost of environmental degradation was about 3.52 percent of the country’s GDP in 2014. Air pollution was found to be the second most important type of degradation in the country, after water overexploitation and pollution (Figure 4).

The paper points to the following key conclusions:

・ The health impacts from PM_2.5 exposure is a pressing environmental challenge in the country, costing society US$1.14 billion, or 1.05 percent of the country’s GDP in 2014. In relative terms, the results are lower than those obtained in other countries of the region (3.58 percent of Egypt’s GDP; 2.48 percent of Iran’s GDP), and in the region as a whole (2.2 percent of the

Middle East and North Africa’s regional GDP)⁹ [8] .

・ Exposure to PM_2.5 caused an estimated 5,450 deaths on average, of which about 75 percent are due to ambient air pollution and 25 percent are due to household air pollution. The cost of ambient air pollution is particularly high in cities like Casablanca, Tangier, and Marrakesh, due to large populations exposed to high levels of PM_2.5 concentration. At the same time, household air pollution is a significant problem for the 18 percent of the rural households that use wood and coal for cooking and heating.

・ The most vulnerable groups for premature deaths from air pollution are adults over 55 years old (due to ischemic heart disease and stroke) and children under 5 (due to acute lower respiratory infections). Special attention should be targeted to these age groups when designing programs for reducing health impacts from air pollution.

Morocco has made considerable progress in establishing air pollution monitoring systems in several urban areas. It is important to continue the assessment and analysis of ambient data by taking into account the expansion of cities and the industrial sites [37] ; as well as to conduct monitoring of household air pollution in key rural areas where use of solid fuels for cooking and heating is common. Concrete options to reduce the cost of ambient air pollution should first be implemented in the most affected cities: Casablanca, Tangier, Marrakesh. Existing studies indicate several ways of reducing ambient pollution, depending on the source. For example, air pollution from the transport sector can be decreased through transport system improvements (e.g. improving public transportation, encouraging alternative modes of transport―including non-motorized transport, improving traffic management) and through vehicle level improvement (e.g. improving fuels and technology) [38] . These options, as well as others, need to be carefully considered and tailored to the context of each city, in order to achieve healthier lives and better economic opportunities.

This paper is a result of a broader analytical work carried out by the World Bank, which estimated the cost of environmental degradation in Morocco. The authors gratefully acknowledge the support of the Ministry of Sustainable Development in Morocco, in particular Ms. R. Chafil, Mr. M. Maktit and Mr. S. Maliki, as well as the contribution of the Directorate of National Meteorology and the Ministry of Health. Special thanks are given also to Ms. E. Strukova, Mr. C. Sall, Mr. A. Khattabi, Mr. A. Jorio, and M. S. Belghazi for their support.

Croitoru, L. and Sarraf, M. (2017) Estimating the Health Cost of Air Pollution: The Case of Morocco. Journal of Environmental Protection, 8, 1087-1099. https://doi.org/10.4236/jep.2017.810069