82
E. N. Udemba
2.2 Theoretical Background
Theoretically, this study is anchored on environmental Kuznets curve (EKC). EKC
is a hypothesis coined from Kuznets curve of testing the income inequality among
the farmers and white-collar workers. Simon Kuznets, [15]. He (Kuznets) tested
the income inequality and came to conclusion that as rural farmers move to urban
cities for white-collar jobs, the per capita income of the farmers who later joined
the white-collar jobs will grow to meet the income of the rich at a point where
curve exist. At this point, the income inequality is reduced. After establishment of
this hypothesis by Kuznets, some energy and environmental economics [10, 21, 29]
adopted this hypothesis to study the impact of economic growth on environment.
It is hypothesized that as the economy grow the quality of the environment will be
impacted negatively until it gets to a certain point where the masses will embrace
the awareness of environmental performance and begin to work toward betterment
of the environment. That particular point is called environmental Kuznets curve.
3 Data, Methodology, and Model Specification
A country specific data (1979–2018) of China is applied in this study. The selected
variables applied for efficient investigation of the ecological performance in China
are per capita of ecological footprint (this comprises six components namely; carbon
footprint, built-up land, cropland, grazing land, forest land, and fishing grounds),
GDP per capita (constant, 2010 U$), agriculture, forestry, and fishing, value added
(constant 2010 US$), urban population and energy use (million tonnes oil equivalent).
The energy use is the summation of different fossil fuel energy sources (oil; natural
gas and coal, all measured in million tonnes oil equivalent) applied in China. Apart
from ecological footprint per capita and energy use which are sourced from the Global
Footprint Network and [25] British Petroleum (BP) statistical review, respectively,
all other variables are sourced from World Bank Development Indicators (WDI). All
series are converted to natural logarithm (Fig. 1).
Brief summary of data and variable is displayed in Table 1.
Methodologies adopted in this study are descriptive statistics, the test of unit
root, autoregressive distributive lag (ARDL), diagnostic tests, and granger causality
test. Descriptive statistics and unit root tests [6, 22]; Kwiatkowski-Philips-SchmidtShin [16] were employed to account for the stationarity and normality of the data,
respectively, while ARDL-bound tests [23, 24] and granger causality were employed
for cointegration, short and long run and for forecasting estimations. Also, diagnostic
tests were employed for robust checking and the accuracy of the methodologies and
techniques applied in this study.
The model specification is anchored on ARDL as proposed by [23, 24]. First, the
econometric and empirical specifications are expressed as follow:
E. N. Udemba
2.2 Theoretical Background
Theoretically, this study is anchored on environmental Kuznets curve (EKC). EKC
is a hypothesis coined from Kuznets curve of testing the income inequality among
the farmers and white-collar workers. Simon Kuznets, [15]. He (Kuznets) tested
the income inequality and came to conclusion that as rural farmers move to urban
cities for white-collar jobs, the per capita income of the farmers who later joined
the white-collar jobs will grow to meet the income of the rich at a point where
curve exist. At this point, the income inequality is reduced. After establishment of
this hypothesis by Kuznets, some energy and environmental economics [10, 21, 29]
adopted this hypothesis to study the impact of economic growth on environment.
It is hypothesized that as the economy grow the quality of the environment will be
impacted negatively until it gets to a certain point where the masses will embrace
the awareness of environmental performance and begin to work toward betterment
of the environment. That particular point is called environmental Kuznets curve.
3 Data, Methodology, and Model Specification
A country specific data (1979–2018) of China is applied in this study. The selected
variables applied for efficient investigation of the ecological performance in China
are per capita of ecological footprint (this comprises six components namely; carbon
footprint, built-up land, cropland, grazing land, forest land, and fishing grounds),
GDP per capita (constant, 2010 U$), agriculture, forestry, and fishing, value added
(constant 2010 US$), urban population and energy use (million tonnes oil equivalent).
The energy use is the summation of different fossil fuel energy sources (oil; natural
gas and coal, all measured in million tonnes oil equivalent) applied in China. Apart
from ecological footprint per capita and energy use which are sourced from the Global
Footprint Network and [25] British Petroleum (BP) statistical review, respectively,
all other variables are sourced from World Bank Development Indicators (WDI). All
series are converted to natural logarithm (Fig. 1).
Brief summary of data and variable is displayed in Table 1.
Methodologies adopted in this study are descriptive statistics, the test of unit
root, autoregressive distributive lag (ARDL), diagnostic tests, and granger causality
test. Descriptive statistics and unit root tests [6, 22]; Kwiatkowski-Philips-SchmidtShin [16] were employed to account for the stationarity and normality of the data,
respectively, while ARDL-bound tests [23, 24] and granger causality were employed
for cointegration, short and long run and for forecasting estimations. Also, diagnostic
tests were employed for robust checking and the accuracy of the methodologies and
techniques applied in this study.
The model specification is anchored on ARDL as proposed by [23, 24]. First, the
econometric and empirical specifications are expressed as follow: