Nexus between Energy Consumption, CO2 Emissions and Economic Growth in India

This paper examines the interdependency between energy consumption, CO2 emissions and economic growth in India post-LPG period. It also explores the Environmental Kuznets Curve (EKC) hypothesis and its applicability in India. The study uses annual time series data from 1990-2014 and applies econometric modelling techniques like Perron's Root Unit test and Vector Error Correction Model.

2
ABSTRACT
The role of the Energy is pretty much significant in the development of the economy
and is also a key for sustainable Development and there is a causality among the energy
economic growth (GDP), consumption (EC), & an overall emissions of CO2. Since the LPG
reforms (Liberalisation, Privatisation and Globalisation) in 1991, India has changed itself
from a shut economy to an open economy. The international trade, FDI & set up of Special
Economic Zones have led to growth of the economy. Growth of the economy is measured out
by the increase in GDP on the basis of per capita but we miss out on the environmental
degradation that follows the economic growth. High rate of economic growth implies high
level of energy consumption. This energy consumption leads to CO2 emissions, climate
change – global warming, mineral depletion, exhaustion of natural resources and depletion of
the ozone layer. The paper attempts to research the nexus between consumption of energy,
CO2 emissions and growth of economy of a developing economy like India; over the post-
LPG period. The purpose is to evaluate the linkage between economic growth and
consumption of energy. I also attempt to establish if the EKC hypothesis holds in case of
India.
Keywords- environment, natural resources, Globalisation, mineral depletion, CO2
emissions, energy consumption, GDP, EKC, etc.

3
INTRODUCTION
Energy is a strategic commodity due to which its role is significant in the
development of a country. In India the energy system has evolved significantly over the last
six decades, keeping up the hopes of 1.2 billion people, inside the democratic framework
along with the globally integrated economy and the environmentally sensitive administration.
The rising energy demand has put humongous pressure on the limited resources as well as
necessitated the optimum use of the resources. Since 1991, India has been pursuing a
reformed development mission (Ministry of Statistics & Program Implementation -
Government of India, 2018). Significant effort has been made into improving the availability
of energy, as a support to the country’s developmental initiatives.
India is the third biggest in consumption of primary energy after China and USA with
the consumption of 5.5% worldwide offer in 2016. The growth of the economy depends
significantly on the long run renewable energy availability which is both affordable and
accessible (India’s energy consumption to grow faster than major economies, 2018). Since
Liberalization, the Indian economy witnessed sustainable growth and this sustained economic
growth of the country has placed a huge demand on the energy resources.
In India, the supply of energy however falls short of the energy in demand. The
heightened consumption of energy has made India one of the major emitters of greenhouse
gases. The Government of India realizes the importance to limit the global warming by
reducing the greenhouse gas emissions. To limit the emission over the coming decades India
is beginning to focus on lowering the carbon intensity over the long term period. The plan is
to reduce consumption of energy and CO2 emissions without significantly affecting the
growth of economy.
The energy conservation ensures energy security and reduction in the greenhouse gases. If the
consumption of energy leads to economic development then the economy is energy
dependent and if the energy falls then this would significantly affect the level of income. But,
if the development of the economy leads to energy consumption then the dependency of
economy on energy is very less. If no relationship exists between them then implementation
of the energy conservation policies will be easy and will not lead to an effect which is
unfavourable on the output. The attempt is to experimentally explore the connection between
environmental degradation, energy consumption, and growth of economy.

4
Background
There was a major impact raised on the economy due to globalization which changes
the world where every nation is free to trade with each other. Today, world nations are
associated with each other through gas, oil, and coal based resources (Chang and Berdiev,
2011. p. 817).. In the specified way, nations are there in the close relations by method of trade
of the energy and in this particular sense; the energy can be translated as a huge component
of the globalization.
The standardized economic theory treats capital and labour as the primary inputs of
production process and fails to consider energy as the production factor (Kurtz and Fustes,
2014. p. 24). It is dealt as an intermediate good for labour and capital and the point of this
statement is that if primary input is not energy then the price of energy and the availability of
energy are not significant for the economic growth. When we look from the perspective of
neoclassical, it could be contended that raising the cost of the energy by even an overall
factor of 2 would only diminish the term GDP by a few amount. As stated by Okun (1974,
1975), on the neoclassical thinking line, contended that the energy when contrasted with
other production inputs, establishes only relatively 'a small cost share' in the overall total
yield.
Perry (1975, 1977) suggested that energy is one of the main components of
manufacturing so it is possible that the energy prices might not affect the productivity and the
output of the economy. If such is the situation then it is possible that an increase in the energy
prices would easily substitute capital with labour and it might not lead to any negative impact
on energy and production (Ebohon, 1996). The other side of the argument by Berndt
andWood (1975) reflects that if energy and labour are substitutable then the existence of
correlative relationship among energy and capital raises its significant way above the cost
share. The dynamic effect of these connections on output and productivity puts light on the
way that energy impact economic growth.
The EKC hypothesis expresses that with the increase in income, there is increase in
the emissions as well until some point of income is reached post which emissions begin to
decay. The EKC hypothesis indicates emissions as a element of income, which presumes that
multidirectional causality keep going from growth of economy to emissions. In any case, it is
possible that causation could keep running from emissions towards economic growth
whereby emissions happen in the process of production and, as a result, there is increase in
income.

5
It is also postulated that in the current case the trade assumes a important part in
overall development of a nation as an exports act as injections and imports as spillages in the
economy. Exports tend to create spill over effects in the production process which results in
greater factor productivity which will further lead to increase in comparative advantage and
specialisation.
The long-term connection between the development of economy, energy consumption
and CO2 emissions along with their heading of causality may vary from one nation to the
other because of the methodological differences and the country specific conditions. There
might exists useful challenges in cumulating the varied genuine energy consumption forms as
their exist contrasts in the units of estimate. The trasfomraiton is dependent upon the quality
of energy and the study attempts to discover the joining between the energy utilization,
development and CO2 emissions.
LITERATURE REVIEW
The connection among the energy consumption and the economic growth has been set
by numerous eminent researchers of India and beyond. There is abundant literature and
scientific perception which says that prime mover is energy of economic growth and is
critical for the change from the developing economy to a developed economy.
Developing nations traditionally over the world have depended mainly on import of
innovation from nations which are developed rather than domestic Research and
Development for driving their change in technology. In the paper “Assessment of R&D & its
overall impact on the manufacturing industries of the India” Sikdar and Mukhopadhya
emphasises that R&D is important for a developing country like India. Their investigation
derives the elasticity of industry-level aggregate factor productivity regarding the R&D
overall contention of the intermediates of foreign and domestic region, a huge positive impact
and with low and high R&D intensive industries on productivity was set up by the R&D
source of the inputs.
The literature on the ecological quality and growth of economy centres’ around the
examination of Environmental Kuznet’s Curve (EKC) existence. In the words of Kuznets
(1955) demonstrates a reversed U-shaped connection among the overall economic growth &
the pay imbalance which was later adjusted into a U-shaped connection between economic
growth and ecological quality. The Environmental Kuznet’s Curve estimates a transformed
U-shaped connection among the income level & the level of pollution which demonstrates

6
that as the pollution level rises, the ecological quality falls apart however it enhances as the
per capita income progresses.
In the paper “Note & Comments: On the overall Relationship among the GNP and
Energy” – a unidirectional causality was discovered by Kraft Kraft (1978) in U.S. between
consumption of income to energy by using data for duration from 1947–1974. Chen et al.
(2007) supported the above findings. A relationship between environmental productivity and
national income was detected by him in China. As stated by Lee (2005; 2006) inferred that
there exists both short-run & long run causalities between consumption of energy and GDP,
but no evidence existed of vice versa causality. These findings suggests that there might be
adverse effects of economic growth on conservation of energy which can be either transitory
or a permanent trend in developing nations.
Linh and Lin (2014) inspected the dynamic connection among the overall
consumption of the energy, CO2 emissions, Economic growth and FDI for Vietnam in the
period 1980 to 2010 in light of cointegration, Granger causality tests and Environmental
Kuznets Curve (EKC) approach. Their outcome were not ready to help the Vietnam’s EKC
theory, anyway the test results Granger causality & cointegration shows a dynamic
relationship among the emissions of the co2, FDI, energy consumption & the economic
growth.
The connection among the utilization of growth & energy of economy has been an overall
subject of incredible enquiry as the energy is thought to among the economic growth’s
critical driving forces in all economies (Pokharel, 2006).
As stated by Aqeel & Butt (2001) surveyed causal connection among usage of energy
and the overall growth of the economic in the Pakistan and they declared that like the other
creating nations- Pakistan too is an energy intensive developing nation, & like vast majority
of the other economies of the non-oil producing regions- Pakistan’s need of the energy are
also additionally satisfied by the expansive import quantities.
Renjish and Vimala (2016) in their paper analysed the usage of energy in various
economic sectors and concentrated on its immediate impact on the economic growth of India,
other than looking over the impact of different types of growth of energy consumption on the
overall growth of private investment & private consumption as different segment of the
growth of GDP. The particular study proposed, in reducing natural gas & consumption of the
oil especially in the economy consumption sectors, thereby conserving energy and
diminishing its requirements by the way of the demand-side management and by adopting
technologies which are more energy efficient in all areas.

7
METHODOLOGY AND ECONOMETRIC MODELLING
Methodology
The paper makes use data of annual time series for the duration starting 1990–2014
which include - Carbon dioxide emissions (metric tons per capita), the real GDP per-capita
(constant 2010 US $), trade openness (% of exports & imports Of GDP), energy consumption
(kg of oil equivalent / capita) & financial development (private sector total credit as an
overall ratio of the GDP) of India & total population (taken in the thousands) the data for
which is collected from the World Bank website.
For the purpose of the study, energy consumption, the growth trend of the CO2 emissions,
economic growth and trade has been depicted graphically (Fig 1)
To show the interdependency between consumption of energy and growth the two
variables are regressed upon each other and the fitted regression line is graphed (Table 1 and
Fig 2). The descriptive statistics, the standard deviation, the mean value, and the Variation of
coefficient of different variables is given below in Table 2.
Perron’s Root Unit test is been applied to examine the stationary property of time
series data & in order to establish the relationship of the long run among economic growth,
energy consumption, and Trade in India, CO2 emissions, Gregory Hansen structural break
integration test is used. In order to study the short run nexus between economic growth, CO2
emissions, energy consumption, & Trade in India -Vector Error Correction Model is used.
Econometric Modelling
The theory of EKC suggests that the effect of environmental is a reversed Ushaped
function of the income (GDP) and an overall logarithm of the marker is modelled as a
quadratic function of the logarithm of GDP. In view of the EKC hypothesis, there is
arrangement of a linear logarithm quadratic model to play out the connection among co2
outflow, economic growth, energy consumption and trade as follows:
Ln CO2 = β0 + β1 lnENCt + β2 ln GDPt + β2 ln GDPsq + β3 ln TOPt + νt
Where t = 1, 2, 3,….,T = Time period,
Xt = vector representing explanatory variables
Νt = error term whose assumption is serial uncorrelation.
According to the Environmental Kuznets Curve theory, the signs of lnENCt, ln GDPt are
expected to be positive because higher the ratio of consumption of energy and income, the
greater would be the co2 emissions. On the other hand, there is an expectation that ln GDPsq
will display a negative sign.

8
Perron (1989) Unit Root test
The macroeconomic data set is assumed to be stationary around a deterministic trend and the
idea of unit root test was introduced by Dicky and Fuller to check whether or not the current
data is stationary.
The procedure of Perron’s 1989 is been taken into account by a single exogenous break in
accordance with the current theory of the asymptotic distribution and perron utilises a
modified Dicky-Fuller unit root tests which incorporates the dummy variables to the account
for the purpose of the exogenous structural break.The performance is not dependent on the
intensity of the break. The results of the Unit Root test of the Perron model are established
using E-views and is depicted in table 3.
Johansen-Cointegration Test
The Johansen (1991) co-integration test is used to test whether or not the series of data are
co-integrated. Johansen test has been used in this paper with respect to the vector
autoregressive model. The test makes use of maximum eigenvalue tests and trace for
determining the number of co-integrated relationships and Table 4 displays the results of the
Johansen co-integration test along with null & alternative hypotheses.
Table 5 shows the estimated equation of the econometric model using OLS.
Granger Causality Test
The co-integration test above has established the equilibrium relationship which is long–run
among the emissions of the co2, economic growth, energy consumption & trade. Granger
causality test with respect to the VECM will assess whether or not the one variable historical
value will affect the present value of the other basic variables. The results will establish the
direction of the causal relationships between the variables in the model (Table 6).
EMPIRICAL RESULTS
Graphical Representation
Figure 1

9
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
0
200
400
600
800
1000
1200
1400
1600
1800
R² = 0.911843414801842
R² = 0.910600456158458
R² = 0.923367272276659
R² = 0.936290886790579
G r ow t h Tr e ndCO2 emissions (metric tons per
capita)
Linear (CO2 emissions (metric
tons per capita))
Energy use (kg of oil equivalent
per capita)
Linear (Energy use (kg of oil
equivalent per capita))
Trade (% of GDP)
Linear (Trade (% of GDP))
Linear (Trade (% of GDP))
Linear (Trade (% of GDP))
Linear (Trade (% of GDP))
GDP per capita (constant 2010
US$)
Linear (GDP per capita (constant
2010 US$))
The above figure depicts the growth trends of Co2 Emissions, Energy Consumption, GDP
and Trade over the period 1990-2014. R2 value indicates the coefficient of determination and
a high square signifies that the model explains variability of the response data around its
mean. For instance Energy Consumption has R2 of 0.9106 which indicates that 91.06%
variability is explained by the model itself. It is a goodness of fit.
Interdependence among Energy Consumption & the Economic Growth
To prove the theoretical discussion above, an empirical estimation of the two variables are is
provided here. The two variables are regressed and the regression line is fitted to check the
interdependence.
Table 1

10
The above table explains the interdependence among the energy consumption & economic
growth. The p value is highly significant which rejects our Null Hypothesis that the variables
are not interdependent and hence our alternate hypothesis – the variables are interdependent
holds. One percent change in GDP leads to 49.65% change in energy consumption. The
regression line is fitted below.
Figure 2
6.2 6.4 6.6 6.8 7 7.2 7.4 7.6
5.4
5.6
5.8
6
6.2
6.4
6.6
f(x) = 0.496537507780024 x + 2.74197079600704
LnGDP Line Fit Plot
Y
Predicted Y
Linear (Predicted Y)
LnGDP
Y
Descriptive Statistics
Table 2
Variables Mean Standard Deviation Coefficient of Variation
CO2 Emissions(CO2) 1.088993 0.295970704 27.17837768
Energy Consumption(ENC) 454.93 86.76688217 19.07257691
GDP 929.6904 348.2861314 37.4625941
Trade Openness (TOP) 34.42779 13.97812259 40.6012775
The Unit Root Test is performed in E-views. The result is as follows:
Table 3

11
The Null Hypothesis in the paper assumed that data series has unit root. The result show that
the variables are stationary with respect to the first difference at 5% level of significance.
This means the null hypothesis is rejected. Hence, all data series are integrated of order 1
(I(1)) and appropriate for further testing.
Co-Integration Test
Johansen Cointegration test is carried out in E-Views to check the co-integration between the
variables. The trace and maximum eigenvalue statistic show that there exists differences in
significance level of cointegration equations. The results are used in choosing the number of
cointegrated relationships. The result is as follows.

12
Table 4
The Trace test & the maximum Eigen value test indicates that there is one co-integrating
equation at 5% level of significance. The number of co-integrating equations for both the
Trace & Maximum Eigen Value Test are the same. The existence of cointegration in the
model suggests that OLS estimation is the best estimator and is also consistent with the
findings of Alves & Bueno (2003).
Estimated Regression of the Model
To determine the impact of independent variables on the dependent variable, we regress the
model in Stata. The result is as follows:
Table 5

13
In our model; lnCO2 is the dependent variable and energy consumption, GDP, sqGDP and
trade are independent variables. After regression we can see that all the variables are
significant at 5% significance level other than trade openness. An increase in energy
consumption by 1 percent would lead to high percentage increase in CO2 emissions. An
increase in GDP would also lead to an increase in CO2 emissions. Since coefficient of GDP
is positive, the coefficient of square of GDP is negative but it is significant as the p-value is
0.004. The trade openness is not significant in determining the level of CO2 emissions in our
model. R-square of 0.9947 indicates that 99.47% variation is explained by the model itself.
Granger Causality Test
Table 6 shows the short-run Granger causality result with the null hypothesis being that there
exists no causal relationship between each pair of variables. Some of the results support the
second hypothesis; which indicates the existence of short-run relationship between variables.
One bidirectional causality relationship and four unidirectional causality relationship between
the variables also exists.
sqGDP establishes causality with CO2 emissions at 10% level of significance.
sqGDP establishes causality with energy consumption at 10% level of significance.
TOP establishes causality with GDP at 5% level of significance
TOP establishes causality with sqGDP at 5% level of significance
CO2 emissions establishes causality with energy consumption at 10% level of significance.
A bi-directional causality cannot be established.

14
Table 6
The causality explains that as the GDP or the level of economic growth rises, the level of
CO2 emissions and energy consumption rises as well. Trade openness affects the level of
GDP. An increase or decrease in net exports will lead to significant changes in the GDP. CO2
emissions that is the level of pollution affects the level of energy consumption in the
economy.

15
CONCLUSION
The paper attempts to test the EKC theory in the Indian economy based on the empirical
results it can be concluded that as the level of GDP rises, the CO2 emissions rises, as the p-
value is highly significant. Causality exists between sq GDP and CO2 emissions. However
due to unidirectional relationships, a very strong inference cannot be made from the results.
The paper assessed the linkage between consumption of energy and economic growth
in India. The GDP growth rate fuels the rate of energy consumed in India, but the reverse is
not true. The energy policies made should be such that it reduces conventional and non-
renewable energy consumption, like crude oil and natural gas because importing these energy
resources leads to leakage in the economy. The government incurs huge expenditure in
importing the energy resources at the subsided rates, which has more implications for the
maintenance of a sound macroeconomic environment. However, if the resources are
consumed in small amounts then it would not only keep the environment clean but also
stabilise the financial position of the country. Therefore, effort should be made to use the
renewable energy sources for efficient utilization, which in turn would make the natural
resources sustainable.
The tests also detect that consumption of energy plays a significant role in trade but
also leads to increase in pollution level. Empirically energy consumption leads to increase in
economic growth in the short run which implies that the supply of energy is integral for the
sustainable economic growth and to meet the energy demand of the economy. There is an
urgent need to use energy conservation policies to reduce the CO2 emissions and other
pollutants in the economy.

16
REFERENCES
Alves, D. C. O., and R. D. Bueno. 2003. ‘Short-Run, Long-Run and Cross Elasticities of
Gasoline Demand in Brazil.’ Energy Economics 25:191–209.
Aqeel, A. & Mohammad Subihuddin Butt., "The Relationship Between Energy Consumption
and Economic Growth in Pakistan", Asia-Pacific Development Journal, 2001, Vol. 8,
pp.101-10.
Berndt, E. R. and D. O. Wood, "Engineering and Econometric Interpretations of Energy
Capital Complementarity.", American Economic Review, 1979, Vol.69, pp.342-54.
Chang, C.P., Berdiev, A.N. (2011), The political economy of energy regulation in OECD
countries. Energy Economics, 33(5), 816-825.
Chen, W. Y.,Z.X.Wu, J. K. He, P. F.Gao, and S. F. Xu. 2007. ‘Carbon Emission Control
Strategies for China: A Comparative Study with Partial and General Equilibrium
Versions of the China market Model.’ Energy 32:59–72.
Gregory, A.W. & Hansen, B.E. (1996). “Residual based tests for cointegration in models with
regime shifts”. Journal of Econometrics, 70, 99-126.
India’s energy consumption to grow faster than major economies. (2018). Retrieved from
https://economictimes.indiatimes.com/industry/energy/oil-gas/indias-energy-
consumption-to-grow-faster-than-major-economies/articleshow/56800587.cms
Johansen, S. (1988). “Statistical Analysis and Cointegrating Vectors”, Journal of Ecnomic
Dynamics and Control, 12(4), 231-254.
Kraft, J., Kraft, A., "On the Relationship between Energy and GNP", Journal of Energy and
Development, 1978, Vol. 3, pp.401-03.
Kurtz, D.V., Fustes, M. (2014), Globalization and energy: An anthropological perspective.
Journal of Globalization Studies, 5(2), 19-38.
Kuznets, S. (1955). Economic Growth and Income Inequality, American Economic Review.
45(1), 1-28.
Lee, C. C. 2005. ‘Energy Consumption and gdp in Developing Countries: A Cointegrated
Panel Analysis.’ Energy Economics 27:415–27.
Linh, D.H. & Lin, Shih- Mo (2014). “CO2 Emissions, Energy Consumption, Economic
Growth and FDI in Vietnam” Managing Global Transitions, 12(3); 219-232.
Ministry of Statistics and Program Implementation | Government Of India. (2018). Retrieved
from http://www.mospi.nic.in

17
Perron, P. (1989). “The Great Crash, the oil price shock and the unit root hypothesis”,
Econometrics, 57, 1361-1401
Pokharel, S. H., "An Econometrics Analysis of Energy Consumption in Nepal", Energy
Policy, 2006, pp.1-12.
Renjish Kumar V K , Dr. Vimala M, ―”Energy Consumption In India- Recent Trends”, Asia
Pacific Journal of Research Vol: I. Issue XXXVI, February 2016 pp no 140-
151,ISSN: 2320-5504, E-ISSN-2347-4793
Sikdar, C. and Mukhopadhyay, K. (2018) ‘Assessment of R&D and its impact on Indian
manufacturing industries’, Int. J. Computational Economics and Econometrics, Vol.
8, No. 2, pp.207–228.