Guest essay by Dr. Indur M. Goklany
Periodically we are flooded with reports of air pollution episodes in various developing countries, and claims of their staggering death toll, and consequent reductions in life spans. The Economic Times (India), for example, recently claimed:
Nov 09, 2017, 05.11 PM IST
With pollution levels in NCR 40 times the World Health Organization’s safe limit, your life expectancy could be cut short…
Killing you softly
The US embassy website said levels of the fine pollutants known as PM2.5 that are most harmful to health reached 703 — well over double the threshold of 300 that authorities class as hazardous. PM2.5 are particles with a diameter of 2.5 micrometers or less and a study recently found that a 10 µg/m3 (per cubic meter of air) increase leads to a 1.03 year reduction in life expectancy…
What happens if something is done about it
In Delhi, people could live as much as nine years longer if India met WHO standards, and six years longer if it met its own standards. Similarly, people in other metros like Kolkata and Mumbai could live for around 3.5 years longer if India complied with WHO standards.
Not to be outdone, the BBC has a video online, Bosnia’s silent killer — air pollution:
How credible are these reports? Are aggregate data on life expectancy consistent with such claims? Here I will focus on air pollution from PM2.5, which are generally regarded to be the deadliest form of air pollution (and the subject of the above reports).
Consider that Delhi might have the worst respirable suspended particulate matter (RSPM) air quality among India’s major cities (Figure 1). But — surprise — it also has the second highest life expectancy among India’s states (73.2 years for 2010-2014)!  By contrast, average life expectancy for India is 67.9 years.
Figure 1: Concentration of respirable suspended particulate matter (RSPM) in residential and industrial areas in major Indian cities. Source: Hosamane SN, Desai GP. 2013. Urban air pollution trend in India-present scenario. International Journal of Innovative Research in Science, Engineering and Technology 2: 3738-47.
Similarly, Beijing, hardly the cleanest place in China, has its 2nd highest life expectancy, 82.0 yrs, behind Shanghai (83.2 yrs in 2016). The national average was 76.3 yrs in 2015. By contrast, Hawaii, the state with the highest life expectancy in the U.S., had a life expectancy of 81.2 yrs in 2014, while the U.S. average was 78.8 yrs in 2015!
Trends in Population Exposure to PM2.5 and Life Expectancy
CHINA & INDIA
Figures 2 and 3 show trends in estimates of mean population exposure to ambient PM2.5, life expectancy, CO2 emissions (a surrogate for industrial activity and fossil fuel use), and GDP per capita (a surrogate for both income and economic well-being) for China and India, respectively. [Although I would not rely on these PM2.5 exposure estimates for any quantitative purposes, I will assume that they are good enough to identify broad qualitative trends.]
Figure 2: China — Trends in (1) GDP per capita (in constant PPP adjusted 2011 international dollars); (2) CO2 emissions (in million metric tons of carbon), (3) population weighted annual ambient exposure to PM2.5 based on Brauer, M. et al. 2016, for the Global Burden of Disease Study 2015; (4) life expectancy (in years). Sources: CO2 data from CDIAC; all other data from the World Bank’s World Development Indicators.
Figure 3: India — Trends in (1) GDP per capita (in constant PPP adjusted 2011 international dollars); (2) CO2 emissions (in million metric tons of carbon), (3) population weighted annual ambient exposure to PM2.5 based on Brauer, M. et al. 2016, for the Global Burden of Disease Study 2015; (4) life expectancy (in years). Sources: CO2 data from CDIAC; all other data from the World Bank’s World Development Indicators.
These figures show:
· Life expectancies increase even as both GDP per capita and CO2 emissions increase. [Note that China’s life expectancy declined toward the beginning and end of the 1960s. This was probably because China still had not conquered hunger and food supplies per capita were consequently low.]
- Life expectancies have gone up despite increases in ambient PM2.5 exposure.
- GDP per capita tracks fairly well with CO2 emissions.
BOSNIA-HERCEGOVINA (BOSNIA for short)
Bosnia-Hercegovina is an interesting case, as can be seen by Figure 4. To put this figure in context, consider that Bosnia underwent drastic social, political and economic turmoil during the 1980s and through the early 1990s. Following Tito’s death in 1980, Yugoslavia disintegrated into its constituent pieces, and Bosnia-Hercegovina emerged as an independent state in 1992. Shortly thereafter, the Bosnian War got underway. It ended formally in 1995.
During this period of upheaval, GDP per capita and fuel use declined temporarily, as did CO2 emissions and mean population exposure to PM2.5. Not surprisingly, because of war casualties and, possibly, declines in GDP per capita and fossil fuel use, life expectancy declined somewhat. By 1994, GDP per capita had started to climb again, and so did fossil fuel use and life expectancy. However, PM2.5 exposure stayed more or less constant through the early 2000s, after which it increased (i.e., deteriorated), yet life expectancy has continued to increase.
Figure 4: Bosnia — Trends in (1) GDP per capita (in constant PPP adjusted 2011 international dollars); (2) CO2 emissions (in million metric tons of carbon), (3) population weighted annual ambient exposure to PM2.5 based on Brauer, M. et al. (2016), for the Global Burden of Disease Study 2015; (4) life expectancy (in years). Sources: CO2 data from CDIAC; all other data from the World Bank’s World Development Indicators.
The above figures (2–4) also indicate that for each of the three countries, it’s not evident that PM2.5 shortens lifespans or, if it does, its effects are more than overwhelmed by increases in life expectancy enabled directly or indirectly by economic growth (which is underpinned by fossil fuel use).
[The figures also show that CO2 emissions rose much more rapidly in each country than PM2.5 — see the following table. This suggests that each society had determined formally or informally that they would rather first obtain the benefits associated with fossil fuel use before turning their attention to reducing PM2.5 exposure or, for that matter, foregoing the benefits of fossil fuel use. This should be kept in mind when one develops estimates of the willingness to pay for co-benefits from PM2.5 reductions as part of any analysis of the costs and benefits of CO2 reductions, or in the calculations of the social cost of carbon. In other words, real world data does not support the notion that people are willing to pay for reductions in PM2.5 (or other pollutants) at the expense of foregoing fossil fuel use except in extraordinary circumstances. This notion might have been valid once upon a time for colorless and odorless gases, and had the effects of air pollution been unknown, but it can no longer be considered true today given the wide coverage of pollution matters in the local and international media and the internet, and the emphasis on renewable sources and pollution controls.]
|China||India||Bosnia & Hercegovina|
|Increase in PM2.5 exposure from 1995 to 2014 = ΔPM2.5||17%||18%||17%|
|Increase in CO2 emissions from 1995 to 2014 = ΔCO2||210%||176%||648%|
|Ratio of growth from 1995–2014
Let’s now look at data from the United States.
Figure 5: United States — life expectancy (yrs), GDP per capita (1990 International, PPP-adjusted $), SO2 and PM10 emissions (million short tons), PM2.5 (mean annual exposure, μg/m3). Sources: Updated from Goklany. The improving state of the world: why we’re living longer, healthier, more comfortable lives on a cleaner planet, Cato Institute (2007) using: Haines, Michael R. , “ Expectation of life at birth, by sex and race: 1850–1998 ”; Historical statistics of the United States, colonial times to 1970, US Department of Commerce, Bureau of the Census (1975); CDC (2016); CDIAC (2017); World Bank Data Bank (2017).
In addition to trends in estimates of mean exposure to ambient PM2.5, life expectancy, CO2 emissions, and GDP per capita, Figure 5 also shows trends in sulfur dioxide (SO2) and PM10. [SO2 is a proxy for sulfate aerosols, which would be a component of PM2.5 and also PM10. Note the dramatic reductions in PM2.5 since 1990.]
This figure shows that life expectancy has been increasing, with occasional setbacks, since at least 1850. These improvements got steadier after 1880, a period during which fossil fuel use began to take off, but with occasional relapses, e.g. 1916–1918 during World War I and the Spanish flu epidemic, the Depression era years, and World War II years). But since World War II, the more or less steady increase in life expectancy has been punctuated by fewer and smaller relapses.
Over the entire period, 1850–2015, air pollution levels would have first increased as society’s reliance on fossil fuels and industrialization increased. Then, one by one, the various pollutants were reduced. As noted elsewhere, the order in which these pollutants peaked seem to more or less follow the order in which society perceived (or became aware) of their negative impacts.
Setting aside the steep decline in life expectancy from 1916–1918, Figure 5 reinforces the observations made from Figures 2 through 4, namely: (a) life expectancy is better correlated (and improves) with GDP per capita and CO2 emissions, than pollution levels, and (b) and it has continued to improve, notwithstanding trends in pollution from fossil fuel combustion. The information on Figure 5 is broadly consistent with data on deposition from various forms of airborne particulate pollution from the Arctic from 1788–2003, shown in Figure 6.  The latter probably reflects a composite of industrial and forest fire activities in both the U.S. and Canada. [Significantly, because of the steady increase in urbanization since the late 18th century, mean exposure to PM would have risen more steeply than indicated by emissions alone; this casts further doubt on the purported detrimental effect of PM on life expectancy.]
Figure 6: Air pollution deposition in the Arctic, 1788–2003. (A) Annual average concentrations of black carbon (BC) and vanillic acid (VA). VA is an indicator of boreal forest fires. The gray shaded region represents the portion of black carbon (BC) attributed to industrial emissions, not boreal forest fires. (B) Annual average concentrations of BC and non-sea-salt sulfur (nss-S). Spikes in nss-S are from explosive volcanic eruptions (e.g., Tambora, 1816; Krakatoa, 1883; and Katmai, 1912). Source: McConnell et al. (2007)
An examination of historical trends for other countries indicates that for the most part each country follows the same general script outlined above, namely, industrialization, stoked by increases in fossil fuel use, has increased GDP per capita and is associated with increases in life expectancy, regardless of whether air pollution levels went up or down. The exceptions to this pattern would be countries that have easy access to alternative energy sources, e.g., hydropower, and/or nuclear. Also, there might be discontinuities in the general increases in GDP per capita, fossil fuel use and life expectancy for some countries during periods of major economic, social and political disruptions such as the collapse and disintegration of the Soviet Union and the restructuring of its satellite states.
To recap, the death toll from air pollution caused by fossil fuel combustion — and the resulting decline in life expectancy — are, to quote Mark Twain, greatly exaggerated. In fact, for whatever reason, life expectancy increases in association with fossil fuel use.
Finally, some may argue that while PM2.5 may not reduce life expectancy, it may actually make the population sicker. But this argument fails scrutiny.
The following table compares (unadjusted) life expectancy at birth in 1950 against “health-adjusted life expectancy” (HALE) in 2000 and 2015 for the U.S., the world’s two-most populous countries — India and China, and the world. [HALE is a measure that tries to combine the quantity of life (i.e., its length) with the quality of health experienced over that lifespan. The World Health Organization defines HALE as the “average number of years that a person can expect to live in full health by taking into account years lived in less than full health due to disease and/or injury.”] The table shows that HALE today substantially exceeds the unadjusted life expectancy in 1950. [1950 is shortly before the China and India began to industrialize in earnest. It is also at the start of a fresh burst of industrialization in the U.S. — see Figure 5). In other words, we are not only living longer, we are staying healthier for a longer period of time.
|Life expectancy in
|Health-adjusted life expectancy in 2000
|Health-adjusted life expectancy in 2015
|Atmospheric CO2 level (ppm)||311||370||401|
Unadjusted and health-adjusted life expectancy (HALE) for China, India, U.S., and the World. Health-adjusted life expectancy adjusts unadjusted life expectancy downward to account for the amount of time spent in an unhealthy condition and the severity of that condition. Sources: Maddisson (2001), p.30; ESRL Mauna Loa data, ftp://aftp.cmdl.noaa.gov/products/trends/CO2/CO2_annmean_mlo.txt WHO (2016), http://gamapserver.who.int/gho/interactive_charts/mbd/hale_1/atlas.html
So why the discrepancy between claims that PM2.5 (or air pollution more generally) reduces life expectancy, and the reality that life expectancy has actually increased, and continues to increase in some of the most polluted cities of the world despite increases in PM2.5?
A couple of reasons, which are not mutually exclusive, come to mind:
- The cumulative direct and indirect effects of economic development (and fossil fuel use) on life expectancy not only outweigh the effects of PM2.5, they also enable populations to reduce PM2.5, once more significant health threats are reduced.
- Life expectancy is based on data on real births and real deaths, whereas the mortality effects of PM2.5 are based on “statistical” deaths or, to use a term currently in vogue, “fake” deaths. As Steve Milloy is fond of asking, “Where are the bodies?”
In today’s world, claims of air pollution shortening life expectancy are fake news premised on fake deaths.
 Goklany IM. Have increases in population, affluence and technology worsened human and environmental well-being. The Electronic Journal of Sustainable Development. 2009;1(3):1; p. 15.
 McConnell JR, Edwards R, Kok GL, Flanner MG, Zender CS, Saltzman ES, Banta JR, Pasteris DR, Carter MM, Kahl JD. “20th-century industrial black carbon emissions altered arctic climate forcing,” Science. 2007 Sep 7;317(5843):1381-4.
 Goklany IM. Have increases in population, affluence and technology worsened human and environmental well-being. The Electronic Journal of Sustainable Development. 2009;1(3):1
 Estimates of deaths from air pollutants are based on epidemiological studies. However, these studies have several shortcomings. They include the fact that it is inappropriate to use outdoor monitors fixed in space to represent population exposure to PM2.5 (because most people move around, spend a substantial period of time indoors, and indoor and outdoor air quality may not always be the same for a variety of reasons). Steve Milloy has an extensive critique of a recent epidemiological air pollution study that illustrates many of these shortcomings.