=Paper=
{{Paper
|id=Vol-1498/HAICTA_2015_paper41
|storemode=property
|title=Countries Clustering with Respect to Carbon Dioxide Emissions by Using the IEA Database
|pdfUrl=https://ceur-ws.org/Vol-1498/HAICTA_2015_paper41.pdf
|volume=Vol-1498
|dblpUrl=https://dblp.org/rec/conf/haicta/ChalikiasN15
}}
==Countries Clustering with Respect to Carbon Dioxide Emissions by Using the IEA Database==
https://ceur-ws.org/Vol-1498/HAICTA_2015_paper41.pdf
Countries Clustering with Respect to Carbon Dioxide
Emissions by Using the IEA Database
Miltiadis Chalikias1, Stamatis Ntanos2
1
Technological Educational Institute of Piraeus, Department of Business Administration,
Egaleo, 122 44, Greece, e-mail: mchalik@teipir.gr
2
Democritus University of Thrace, Department of Forestry and Management of the
Environment and Natural Resources, Orestiada, 68200, Greece, e-mail: sdanos@mycosmos.gr
Abstract. The purpose of this study is to use clustering variables according to
the Kaya identity, an equation involving energy consumption, economic
growth and carbon dioxide emissions. By using aggregate data from the
International Energy Agency for a dataset of developed and developing
countries, we perform clustering according to variables such as population,
gross domestic product (GDP), total primary energy supply (TPES) and total
CO2 emissions. We use the estimated clusters to have an overview of the
relationship between economic development and carbon dioxide emissions.
Keywords: Carbon emissions, Kaya identity, cluster analysis
1 Introduction
Rapid economic development during the last century led to an increase in the
concentration of greenhouse gases in the atmosphere, especially carbon dioxide
(hence CO2). Carbon emissions are a negative byproduct of burning fossil fuels
mainly for energy and transportation purposes. Global carbon emissions from fossil
fuels follow a rising pattern since 1900. Emissions increased by over 16 times
between 1900 and 2008 and by about 1.5 times between 1990 and 2008 (EPA, 2015).
Data for energy consumption provided by the International Energy Agency
concerning the year 2012, reveal that North America has a mean energy consumption
of about 6.8 toe / person (toe stands for tons of oil equivalent), the EU 3.46 toe /
person, while for non-OECD countries the index is 1.34 toe / person. These
important differences confirm that the level of prosperity and the development rate of
the countries associate positively with energy consumption. A higher standard of
living means more energy consumption and hence more CO2 emissions. European
Union countries have stabilised energy consumption during the last decade due to
Kyoto Protocol commitments. China on the contrary shows an impressive energy
consumption increase, due to intense industrialization. Carbon emissions on a global
scale reached 31.7 billion tons in 2012, from 14.08 in 1971, an increase of 125%.
Analyzing further in OECD and Non-OECD countries, we notice that emissions
increase in Non-OECD countries is significant higher. Specifically, while in OECD
countries there is a 30% CO2 emissions increase between 1971 and 2012, the
347
emissions have tripled in non-OECD countries during the same time period (IEA
2014).
2 Literature Review
According to the International Energy Agency, an equation, proposed by Kaya
(1990; 1997) can be used in order to estimate the level of human effect on earth’s
atmosphere. This index, known as Kaya identity states that aggregate CO2
emissions can be estimated as the result of four inputs: Carbon content of the energy
consumed (or carbon efficiency), Energy intensity of the economy, Production per
person (per capita GDP), and population.
(1)
.
By using this basic index, CO2 emissions can be calculated for a country, region or
on a global scale and we can also test different greenhouse gas emissions scenarios
scenarios.
According to a recent paper by Ntanos et al. (2015), the relation between
economic growth and CO2 emissions was tested by using One Way ANOVA. The
results indicated that there is a positive correlation between GDP, energy
consumption and CO2 emissions for the electricity sector and a negative correlation
for the transportation sector. According to Zafeiriou et al. (2011) economic growth
is linked with energy use while the uneven distribution of fossil fuels around the
globe may lead to countries conflict. The turn to renewable energy sources can help
in countries energy security and help environmental protection. More specific it was
found that the substitution of fossil fuels with biomass can potentially contribute to
the reduction of the greenhouse effect.
In an interesting study concerning CO2 emissions and financial performance of
firms, it was found that the performance of environmentally responsible firms is
negatively related to an increase of global CO2 emissions. This implies that there is a
category of “green” investors that are interested in the environmental attitude of
corporations. In our opinion this approach can be applied not only for firms but also
for countries evaluation (Sariannidis et al. 2013).
Concerning the contribution of renewable energy sources (RES) in reducing CO2
emissions there is a plethora of publications. In a paper revealing the attitudes of
Greek citizens on environmental issues, it was found that citizens are willing to
invest in RES, especially for residential applications. The public’s high level of
awareness on RES can reduce national carbon dioxide emissions and contribute to
GDP growth, by creating jobs and increasing people’s income (Tsantopoulkos et al,
2014).
348
3 Results
Cluster analysis was performed by using variables of CO2 emissions in tons per
capita, GDP in USD per capita and Total Primary Energy Supply (TPES) in tons per
capita. Firstly hierarchical cluster analysis was used and from the Agglomeration
schedule the number of clusters where estimated (it was found that 3 clusters exist).
With K-Means Analysis for the three clusters the following results were obtained:
Table 1. Cluster distribution with number of countries included in each cluster
Cluster Distribution
Number of
% of Combined % of Total
countries
1 91 65.0% 65.0%
2 36 25.7% 25.7%
Cluster
3 13 9.3% 9.3%
Combined 140 100.0% 100.0%
Total 140 100.0%
Three clusters were estimated as we can see in table 1. The countries names included
in the clusters are given below.
First Cluster In the first cluster 91 developing and underdeveloped countries are
included.
Second Cluster: Australia, Israel, Japan, Korea, New Zealand, Austria, Belgium,
Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Ireland,
Italy, Netherlands, Norway, Poland, Slovak Republic, Slovenia, Spain, Sweden,
Switzerland, United Kingdom, Belarus, Cyprus, Kazakhstan, Malta, Russian,
Federation, Turkmenistan, Malaysia, Singapore, Chinese Taipei, Hong Kong, China,
Islamic Rep. of Iran.
Third Cluster: Canada, United States, Iceland, Luxembourg, Brunei, Netherlands
Antilles, Trinidad and Tobago, Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, United
Arab Emirates.
Table 2. Cluster profiles (mean of GDP/cap, CO2 emissions/cap and energy consumption/cap)
Total Primart Energy Supply in toe
GDP per capita in USD CO2 Emissions in tons per capita per Capita
Std. Std. Std.
Mean Deviation Mean Deviation Mean Deviation
Cluster 1 7784.3 5095.6 2.29 1.93 1.07 .69
2 28942.5 10546.1 8.45 2.67 3.79 1.15
3 50515.0 26529.0 20.57 7.47 10.52 4.14
Combined 17192.9 17410.6 5.57 6.26 2.65 3.14
349
In table 2 we can see the average GDP/cap, the average CO2 emissions/cap and
the average energy consumption in toe/cap for each cluster. We observe that the 3d
cluster includes countries with the highest GDP which are the main world oil
producers, while the 2nd cluster includes developed countries and the 1st cluster
included undeveloped countries. We observe that CO2 emissions and energy supply
are positively related to economic development as expressed by GDP per capita.
4 Results
Using a database from IEA, we found that there is a correlation between economic
development and CO2 emissions. We observe that the clusters effectively divide
countries mainly depending on GDP per Capita criterion. Clusters depict rich
countries with a high production of oil, developed countries and undeveloped
countries. The results reveal that economic development is associated with intense
energy consumption and CO2 emissions and reinforce the conclusion that developed
countries are the major CO2 polluters.
References
1. EPA. (2015) Global Greenhouse Gas Emissions Data, Environmental
Protection Agency [available online],
http://www.epa.gov/climatechange/ghgemissions/
2. IEA. (2014) International Energy Agency, Data on CO2 emissions and Energy
Consumption in .xls form, [available online].
3. Kaya, Y. (1990) Impact of Carbon Dioxide Emission Control on GNP Growth:
Interpretation of Proposed Scenarios. Paper presented to the IPCC Energy and
Industry Subgroup, Response Strategies Working Group, Paris.
4. Kaya, Y. Yokobori, K. (1997) Environment, Energy, and Economy: strategies
for sustainability, United Nations University Press, New York, Paris, pp. 331
5. Ntanos, S. Arabatzis, G. Milioris, K. Chalikias, M. Lalou, P. (2015) Energy
Consumption and CO2 Emissions on a Global Level, Proceedings of the 4th
International Conference on Quantitative and Qualitative Methodologies in the
Economic and Administrative Sciences, 21-22 May, TEI of Athens, Egaleo,
Athens, Greece.
6. Sariannidis, N., Zafeiriou, E., Gianarakis, G and Arabatzis, G. (2013). CO2
emissions and financial performance of SR firms; The empirical survey of DJSI
with a non linear model. Business Strategy and the Environment, 22 (2): 109-
120.
7. Tsantopoulos, G., Arabatzis, G and Tampakis, S. (2014). Public attitudes
towards photovoltaic developments: Case study from Greece. Energy Policy,
71: 94-106.
350
8. Zafeiriou, E, Arabatzis, G. and Koutroumanidis, T., (2011). The fuelwood
market in Greece: An empirical approach. Renewable and Sustainable Energy
Reviews, 15 (6), 3008-3018.
351