|Poor Visibility on China's Air
December 2007 Far Eastern Economic Review
by Vaclav Smil
When I lived in Hong Kong in the year 2000 the first thing I did every morning right after getting
up, usually between 5:30-6:00 AM, was to check the Air Quality Index on the Hong Kong's
Observatory Web site.
This was a ritual, rather than an informative, exercise: chances were very high that even at that
early hour most stations would already have high readings which in Hong Kong means the index of
suspended particulate matter and nitrogen oxides was 55-85. Only Tap Mun Chau (a small island in
the territory's northeastern waters, far away from major urban sources of pollution) would sneak
in sometimes with low readings, particularly if there was a brisk southeast wind. Seven years
later, I checked the Observatory's air quality readings from Canada at 5:30 a.m. of Hong Kong
time: every station, including Tap Mun Chau, already had high air-pollution readings, and Mong
Kok's nitrogen oxides were already very high.
This unwelcome reality is explained by a two-fold nature of Hong Kong's air pollution problem. The
region itself is still a considerable source of particulates and sulfur and nitrogen oxides
(especially from inadequately controlled emissions from cars and diesel-fueled trucks and ships)
and as a day progresses traffic congestion results in repeated rise of concentrations in the most
densely inhabited neighborhoods during the late afternoon hours when photochemical atmospheric
reactions create exceptionally high levels of smog. But even in 2000 it was clear that whatever
steps Hong Kong will take to reduce its own considerable emissions such steps would not suffice to
make the region's air fairly clean again. Only massive reduction of emissions that are transported
from Guangdong could do that, as at least 80-85% of all pollutants in Hong Kong's air now
originate in China.
But, as everybody concedes, such a step is unlikely to take place in the near future. Guangdong's
cities and countryside within a radius of about 200 kilometers from Hong Kong are the single
largest source of China's manufactured exports and transforming that region into an area of low
emissions will be a very costly task that would take decades to accomplish. This means that Hong
Kong's citizens will continue to live with the new "normal" and will find it increasingly
impossible to see Kowloon across the Strait from the Central. So will the tourists-instead of a
stunning panorama from the Peak, they may see no further than high rise apartment buildings a few
hundred meters below their platform.
But Hong Kong does not have China's most polluted air. Several northern cities are contenders for
that title, and the capital is not far behind. When temperature inversions limit the mixing layer
above Beijing, tall apartment blocks on the other side of a multilane thoroughfare often appear as
just ghostly silhouettes, and rare is the summer day when Xishan (Western Mountains, about 30
kilometers west of the downtown) can be clearly seen from Meishan (Coal Hill, just north of the
Research has shown that the psychical impact of closed horizons and visibly dirty air is
considerable, and the economic impact on lost tourist income is self-evident. Why visit a place
that can be more sensed than seen?
Limited visibility may be the most immediate indicator of China's poor air quality, but by far the
greatest damage done by polluted air is to human health, to plants and to economic activity. We
can now get a better appreciation of how bad some of these impacts are, and how difficult it is to
assess the damage they create, thanks to a series of quantitative studies published in Clearing
the Air: The Health and Economic Damages of Air Pollution in China, a new book edited by Mun S. Ho
and Chris P. Nielsen and published by the MIT Press in 2007.
Contributions in this book are a part of the China Project whose various programs have been
conducted by the Harvard University Center for Environment since the late 1990s. Assessment
studies in the book deal with estimating health effects of air pollution in China, with local
population exposures to pollutants from different sources (industrial, transportation, electricity
generation), with economic value of air-pollution-related health risks and with sector allocation
of emissions and damage and policies to control them. Those who are familiar with similar studies
done in Western countries will immediately realize how challenging, and how inherently uncertain,
many of these assessments are.
The standard approach in assessing health damage has been first to identify all sources and
atmospheric concentrations of offending pollutants in a study region. Then estimate as accurately
as possible the actual extent of population exposure to the effects of air pollution (in this case
to suspended particulate matter and sulfur and nitrogen oxides derived mostly from coal
combustion). And finally express the resulting health and economic damage by using a suitable
valuation method. Every step has its own problems, and one of the book's strengths is that it
makes clear the extent of the uncertainties encountered.
Certainly the best illustration of why the results of this kind of studies have to be interpreted
with a great deal of caution is a comparison of air-pollution health-damage assessments for China
published during the past 10 years. A World Bank study in 1997 put the total number of premature
deaths due to air pollution at 178,000 a year in cities and at 110,000 in the countryside (mostly
due to poor indoor air quality from inadequately vented stoves). In contrast, a 2003 study led by
Stefan Hirschberg estimated as many as one million premature deaths a year, while the Harvard
project (by Messrs. Ho and Jorgenson) put the excess mortality due to particulate matter and
sulfur dioxide at about 94,000 in 1997. Such discrepancies make it virtually impossible to use
these dose-response studies in formulating any effective policies.
Another obvious problem with studies that inevitably take some time to complete is the truly mad
rate of China's energy consumption growth-it makes everything of historical interest in just a
matter of years. The Harvard book contains an interesting study that deals with the population
response to pollutants from electricity generation based on emissions in 2003. During that year
China's electricity generating capacity reached 391 gigawatts, of which 290 GW were from thermal
(mostly coal-fired) stations; three years later the total rose to 622 GW with 484 GW from thermal
stations, the latter capacity being 67% higher than it was in 2003. This means that between 2003
and 2006 China added nearly 200 GW of thermal capacity, or more than the total installed capacity
in France and the U.K.
A study based on 2003 data is thus already obsolete. If the capacity increase was matched by
emissions increase, we could simply prorate the results. In reality, increased use of flue-gas
desulfurization in China's large coal-fired power plants means that for some stations emissions of
sulfur dioxide emissions have actually declined substantially, while in other regions
concentrations of new and uncontrolled power plants have resulted in a serious deterioration of
air quality. And in all regions continuing urbanization has exposed a larger percentage of
population to potentially harmful concentrations.
Moreover, there is no reliable way to assess long-term nationwide trends. Not surprisingly,
official statistics make the situation look better. The latest report on the nation's air quality
in 2006, issued by the State Environmental Protection Agency in 2007, notes an overall
improvements in urban air quality as about 63% of 559 cities with regular monitoring now meet or
surpass the criteria for urban air (Category II and I in Chinese grading).
But the proportion is different in terms of people affected. As most of the large cities do not
meet the minimum requirements (belonging to category III or worse), in 2006 the split between
population subjected to acceptable and unacceptable levels was 52/48, with nearly 170 million
urban residents living in cities that even by China's relaxed standards are too polluted.
The reality is much worse because the Chinese standards have been far too relaxed. Particulate
concentrations of 200 micrograms/m3 are acceptable in residential areas, while the World Health
Organization's limit used to be just 90 micrograms/m3. Used to be, because the WHO set this limit
aside and no longer sets guidelines for total suspended particulates. Its latest air quality
guidelines issued in 2005 set levels for particles smaller than 10 micrograms annual mean no
higher than 70 micrograms/m3. Particles smaller than 2.5 micrograms can penetrate deeper into
lungs and their chronic effect is mostly responsible for premature mortality (studies have shown
that every additional microgram increase of these particles increases the probability of air
pollution-related mortality by 1%). who's maximum annual mean for these particles is no more than
35 micrograms/m3-but in Chinese cities their concentrations are commonly between 50-150
A misleading impression of reality can be obtained by checking SEPA's Website, which shows a daily
air quality index for China's major cities. On the day I checked the Hong Kong air pollution
index, the SEPA's air pollution index for Beijing stood at 159, in the "lightly polluted"
category. But this classification allows daily concentrations of particulates smaller than 10
micrograms to be as high as 350 micrograms/m3-while who's new guidelines for 24-hour mean level
are no higher than 150 micrograms/m3, less than half the Chinese value. Consequently, by WHO's
standard Beijing would be anything but "lightly polluted." Even more importantly, while high
levels of particulate continue to be a major problem, China is unique among the world's major
economies in having now the worst of both major air pollution types.
As the Western countries (and Japan) got richer, they relegated coal combustion to generation of
electricity in large power plants with highly efficient electrostatic precipitators that remove up
to 99.99% of all particulates, and later they equipped many of those plants with efficient
desulfurization. As a result, classical (London-type) smog consisting of high levels of
particulates and sulfur dioxide has been virtually eliminated in all affluent countries whose
principal air pollution problem is now photochemical smog, a product of complex, long-chained
atmospheric reactions that start with emissions of carbon monoxide, volatile hydrocarbons and
nitrogen oxides. By far the largest source of these pollutants is automotive traffic, and sunny
climates promote frequent formation of the smog. This means that the phenomenon is seasonal in
Toronto and Paris, but it persists for most of the year in subtropical and many tropical cities
with high concentrations of vehicular traffic, and also with high frequency of flights (jet
engines emit a great deal of nitrogen oxides).
During winter months, when households and apartment blocks in northern part of China use coal for
heating and when limited mixing layer keeps pollutants from poorly controlled small coal-fired
power plants near the ground, many Chinese cities still experience high levels of classical
(particulate-sulfur dioxide) smog characterized by visible black dust fall and acrid sulfurous
smell. But in summer the same cities, now full of congested and poorly controlled vehicular
traffic, have recurrent high levels of photochemical smog.
Unfortunately, the Harvard studies have completely ignored the already severe and a rapidly
worsening problem of China's urban photochemical smog. The main product of these complex
atmospheric reactions is ozone, a highly reactive compound that has a number of adverse effects on
living organisms and on materials. Ozone impairs lung function, injures cells, limits work and
exercise capacity and lowers the resistance to bacterial infections; is effects on plants include
lower yields of both annual and tree crops (already documented in China), damaged leaves
(coniferous species are particularly susceptible), and reduced tree growth. High levels of ozone
also contribute to the deterioration of fabrics and rubber.
The best argument for reversing this decline of air quality would be to have a reasonably accurate
estimate of annual damages that would prove the benefits of remedial actions. Coming up with such
values is an extraordinary challenge and wide ranges of values came out even in the most
sophisticated Western studies. A study in the Harvard book shows that one tool often used in
economic valuations of environmental impacts would not work in China: a survey of contingent
valuation, the willingness to pay in order to reduce health risks of air pollution, found that the
Chinese respondents were willing to pay (in relative terms) only 1/10 to 1/1,000 of sums that
Americans were prepared to spend for the same purpose, and a surprisingly large percentage would
not pay anything.
As for the aggregate annual cost of China's air pollution, the Harvard study offers a central
estimate equivalent to about 1.8% of China's GDP in 1997, and the range of uncertainty as wide as
0.65% to 4.7% of the annual GDP a decade ago. Truly there is nothing new under the sun-in 1995 I
was the first Westerner to perform an exploratory assessment of economic costs of China's
environmental degradation (published by the East-West Center in Honolulu in 1996) and my
conservative estimate ended up with 0.6% of GDP as an equivalent annual cost of China's air
pollution in 1990. Can we assume that such conservative estimates worth less than 1% of China's
GDP are applicable today, a decade after Harvard's study benchmark year?
Higher urbanization and hence greater exposure of larger number of people to high air pollution
levels should have worsened the impact, but some successes in emission control have lessened
it-and a net result could not be established without further complex studies. But by the time
these are finished China's GDP, and emissions and exposure may be substantial multiples of today's
Consequently, it is unlikely that those inherently uncertain and easily disputable monetizations
of air pollution effects will have a decisive role in setting long-range policies. Nor am I sure
that international pressure can change anything. Even though the Pacific Ocean may be more than
10,000 kilometers across, on some days one-quarter of all particulates in Los Angeles air can be
traced to China, and this flux will certainly increase in years ahead. But the U.S. has no
leverage in these matters. That polluted air in California is just another part of a destructive
bargain (next to large domestic job losses in manufacturing and a perilously deepening trade
deficit) that the country has so unwisely struck in exchange for cheap and contaminated
Wal-Mart-type merchandise from China.
There is no doubt that new, more efficient techniques and new pollution controls have combined to
reduce the relative growth of China's air pollution during the past 15 years. From 2001-06, the
amount of sulfur dioxide removed from combustion gases nearly doubled and the capture of
particulates went up by nearly 70%.
These are considerable achievements but have been insufficient to prevent further absolute rises
in overall emissions-by 2006 sulfur dioxide releases were up by 30% and those of particulate by
about 10%. And because of runaway urbanization this means that the population exposed to harmful
air pollutant levels (as defined by WHO and not by SEPA) has increased. Unfortunately, as bad as
China's air quality is, an old clichˆm applies only too well: matters will have to get worse before
they will get better.
Mr. Smil is a professor in the Faculty of Environment at the University of Manitoba, and author of
25 books on energy, the environment and China.