Environmental Engineering Reference
In-Depth Information
the factors underlying the pollution faced by urban dwellers. On the other
hand, they were trying to understand the environmental impact beyond
urban areas, and therefore the analysis using aggregate emissions was per-
ceived to provide greater insight. They concentrated on four air emissions:
SPM, SO
2
, NO
x
and CO across 30 countries and across time. Out of the
30 countries, 22 were categorized as high income, six were middle income
and two were low income. The emissions data constructed from fuel con-
sumption i gures were obtained from the World Resources Institute (1991)
and were measured in terms of kilograms per capita on a national basis.
The income data used were per capita GDP in 1985 US dollars. Their
i ndings showed a U-shaped curve of per capita emissions for all four air
pollutants and per capita GDP. However, they found substantially higher
turning points for the similar pollutants used by Grossman and Kreuger.
The estimated turning points were $8709 for SO
2
, $11 217 for NO
x
, $10 289
for SPM and $5963 for CO. They considered this to be reasonable given
their use of aggregate emissions data as opposed to urban atmospheric
concentrations. On the other hand, their results were widely dif erent from
those found by Panayotou, who also used aggregate emissions data. As
the turning points were high enough above the per capita income of most
countries, they concluded that global emissions of these pollutants would
continue to increase in the foreseeable future.
Cole et al. (1997) examined the relationship between per capita income
and a wide range of environmental indicators, which had indirect impact
on the environment using cross-country panel data sets. These indica-
tors were carbon dioxide, CFCs and halons, methane, nitrogen dioxide,
sulphur dioxide, suspended particulate matter, carbon monoxide, nitrates,
energy consumption, municipal waste and trai c volume. They arrived at
three main conclusions. First, the EKC existed only for local air pollut-
ants, while indicators with more global or indirect impact either increased
monotonically with income or had turning points at very high levels of per
capita income with large standard errors, unless they had been subjected
to a 'multilateral policy initiative', which implied that the process was not
automatic. Second, concentrations of local air pollutants in urban areas
peaked at a lower per capita income level than total emissions, which
implied that it was easier to improve the quality of urban air than to
reduce national emissions. Third, transport-generated local air pollutants
peaked at a higher per capita income level than total emissions per capita,
which implied the severity and need to tackle the problem in the transport
sector.
On the other hand, in a number of studies it has been found that CO
2
follows an EKC pattern. In a study by Holtz-Eakin and Selden (1995) in
which they examined the relationship of CO
2
emissions and GDP, they
