Environmental Engineering Reference
In-Depth Information
that would lead to a new trajectory. In the case of CO
2
, the oil price shock
was the impetus that changed its trajectory. Second, a national capacity
existed for a rapid and persistent change under the proper stimuli. In the
case of CO
2
, the change in its trajectory happened within a single year and
continued at a particular emission level despite the continued economic
growth and the decline in oil prices.
Stern and Common (2001) criticized the EKC models by saying that they
lacked adequate specii cation due to the omissions of a number of variables
that were correlated with GDP. Their criticism was based on the results of
statistical tests from their study on the relationship between sulphur emis-
sions and per capita GDP. They applied i xed and random ef ects models
to a subset of panel data from 73 countries over the period 1960 to 1990.
They used data from the ASL and Associates (1997) database, consisting
of data on sulphur emissions for most of the countries of the world over
the period 1850 to 1990. An EKC relationship was found in both models,
however, the turning points were much lower when using data for only the
OECD countries (i xed ef ects: $9239, random ef ects: $9181), than when
estimating EKC for the whole sample (i xed ef ects: $101166, random
ef ects: $54199). Similarly the estimated turning points were extremely
high for non-OECD countries (i xed ef ects: $908178, random ef ects:
$343689), implying a monotonic EKC relationship between per capita
income and sulphur emissions in the case of non-OECD countries and the
whole sample. They said that their i ndings suggested that the inclusion of
trade variables would be important as OECD countries could outsource
the production of pollution-intensive products to the rest of the world.
However, inclusion of global macroeconomic trends and shocks, such as
the oil crisis of the 1970s, seemed unimportant.
In a study by Halkos (2003) using the same sample data from ASL and
Associates (1997) as Stern and Common (2001), he empirically tested
the relationship between environmental damage from sulphur emissions
and per capita GDP. He applied two econometric methods: random coef-
i cients and the Arellano-Bond Generalized Method of Moments (A-B
GMM). The per capita GDP data were derived from the Penn World
Table in Summers and Heston (1991), together with the population data.
The sample of countries chosen by Halkos in his study represented 81
per cent of the world's population. He found an EKC pattern relation-
ship in the case of an A-B GMM model with turning points well within
the sample for all cases (global, OECD and non-OECD), ranging from
$2805 to $6230. However, unlike the results found by Stern and Common
(2001), the turning point for only the OECD countries was higher than
those for the global and non-OECD countries. On the other hand, no
support for the EKC hypothesis was found using the random coei cients
