| Research Background:Since the last few decades global economy has been experiencing significant economic growth;although this growth is not balanced everywhere in the world.Some countries are achieving high economic growth;some countries are experiencing low economic growth and even some countries are failing to achieve economic growth.The major economic drivers;Gross Domestic Product(GDP),Foreign Direct Investment(FDI),export,import and population are growing more rapidly over the last few decades,more specifically since the 1950s.Undoubtedly,economic development brings many blessings for the people by creating new employment opportunities,offering diversified and huge numbers of goods and services,better health care,advanced technological experience,better education,better transportation privileges,alleviating poverty and so on which certainly improve the standard of living of people across the world.In contrast,economic development has few but significant dark sides;for instance emitting excessive CO2 emissions which is the root cause of increasing global temperature,polluting air and deteriorating environment,altogether excessive CO2 emission is an influential threat for the people,plant and planet.Considering the above facts,this study aims to investigate the causal relations between CO2 emission and major economic drivers;GDP per capita,FDI,international trade-export&import,and population(POP)for the ninety-eight selected countries together and income group wise four sub-groups of countries.Research Gap:Although there is ample research work and literature on this field for individual economies,nations,countries and groups of countries,nevertheless there is adequate room to work on this significant research area for some valid reasons.First,research outcomes in the mentioned field are inconclusive and widely varied.Second,existing research mostly carried on individual countries and a few research carried on small groups of countries and those are mostly regional groups of countries.For instance,research carried out on BRICS countries,MENA countries,G-7 countries,South and South East Asian countries(SESA)and so on.We find only two research works which are closely related to the present studies based on different income groups for limited sample countries.Hence,present research aims to analyse causal relationships for a comprehensive sample of countries with reference to low income economies,lower middle income economies,upper middle income economies and high income economies.Third,to the best of our knowledge only Begum et al.(2015)incorporate population as a CO2 emission driver in a research conducted on an individual country,Malaysia.We did not find any other research which incorporates population as a CO2 emission driver;hence we incorporate population as a CO2 emission driver in this research and investigate the causal relationship between them.Hopefully,present study will fulfil the identified research gap and contribute new knowledge to the existing body of research literature in this field.Research Methods and Econometric Model:In order to diagnose the causal relationships between CO2 emissions and major economic variables;present study applies Panel Vector Autoregression(PVAR)econometric model to data of ninety-eight sample countries from the different groups of income;which includes thirteen low income economies,twenty lower middle income economies,twenty-six upper middle income economies and thirty-nine high income economies.However,many scholars like Acheampong(2018);Koengkan,Losekann,&Fuinhas(2018);Lin&Xu(2018);Wang,Wang,Du(2019);Ahmed and Khder(2014)also apply pane VAR model for the same purposes.The present research was conducted based on secondary data for the period of 2000-2018 which was collected from the websites of World Bank(WB),Union of Concerned Scientists,Our World in Data and so on.Data analysis started using descriptive statistics which reveals mean,maximum,minimum,standard deviation,skewness,kurtosis values of the respective variables and Eviews 10 was used to run panel unit root test,panel cointegration test,optimal lag selection and PVAR.Data Analysis and Results:Thesis started with an overview of long-run trend analysis of major variables selected for the study which shows insightful trends and significant changes occur over the long-run;mostly variables,CO2 emissions,GDP,FDI,export,import and population experience upward trend.In some cases,variables experience minor and sharp fluctuations,for instance FDI and international trade experience fluctuating trends in recent years;most probably due to economic shock,industrial and agricultural production shock,regional and global financial crisis.In contrast,some variables,for instance GDP,CO2 emissions and population,experience a smoothly rising trend in the long horizon.Next,the study conducts analysis of descriptive statistics and reveals that the upper middle income group of countries emit the highest average CO2;high income countries emit the second highest CO2 emission during the study period.Low income group of countries emits the lowest average CO2 whereas the lower middle income group of countries emits third highest CO2 annually based on the data from the year 2000 to 2018.According to these results,it can be concluded that higher the country’s income higher the country’s CO2 emission.Study also reveals that GDP,FDI,import and export volumes are positively correlated with CO2 emission,which means higher the GDP,FDI,import and export higher the CO2 emission and vice versa.The empirical estimation under VAR model begins with the inspection of stationarity properties of the selected factors by applying a number of alternative panel unit root tests.Considering the characteristics of panel data,a unit root test conducted to inspect the variables whether data are stationary or nonstationary before establishing the Panel VAR model.The reliability of the Panel VAR model estimation depends on the stationarity of the variables.Hence,the present study employs Augmented Dicky Fuller(ADF)and Phillips-Perron(PP)-Fisher Chi-square unit root test.Panel unit root test results for full panel of ninety-eight sample countries show CO2,EX,IM and GDP are non-stationary under level unit root test as p-values are greater than 5%and thus fail to reject null hypothesis that means these variables are non-stationary.On the other hand,FDI and POP are stationary as P values are less than one percent(1%),means null hypotheses are rejected at one percent(1%)level of significance.However,under first difference all the p-values are less than 1%which means null hypotheses are rejected at the 1 percent level of significance and thus data are stationary.Panel unit root tests results for four sub-groups of panel,namely low income economies,lower middle income economies,upper middle economies and high income economies confirm stationarity of data at first difference.Since data are stationary at first difference,therefore data meet the criteria to run Panel VAR and avoid generating spurious regression outcomes.Since,econometric models suggest that dependent variable rarely depends on the independent variable instantaneously in the field of economics;hence after specifying the model and stationarity test,VAR requires to find optimal lag length for running the data analysis purpose.Mostly dependent variable responds to independent variable with laps of time which is called lag.Variables selected for the current study are in the same nature which might respond with laps of time.It is also fact that too many lags lead to loss of degree of freedom.Therefore,before running Panel VAR,need to find optimum lag length.Albeit there is no hard-and-fast-rule on the choice of lag length nevertheless Akaike Information Criterion(AIC)value is one of the well accepted standards.According to this criteria,lower the AIC value,better the model.Applying Eviews software for VAR model,present research finds two periods lag length as optimal which is consistent with many previous researchers.Next,Johansen cointegration test conducted before setting "Panel VAR" for five panel data sets using Eviews to test the stationarity of data sets whether they have a stable,long-run relationship or not.The precondition for running panel cointegration is that model variables must be non-stationary at level unit root test but all the variables become stationary when these are converted into first differenced unit root test.All the five different data sets used in this research met this criterion.Panel cointegration test fails to reject null hypothesis for all five panel data sets which means model variables are not cointegrated in the long-run and capable for identifying causal relationship and meaningful results.Finally study reveals expected and satisfactory results for all the five different samples which are discussed here.VAR results for full panel data of ninety-eight countries identify both unidirectional and bidirectional causal relationship between CO2 emissions and major economic variables.Results show bidirectional causal relationships between GDP and CO2,GDP and FDI,GDP and export,population and GDP which mean variables affect each other.For instance,GDP contributes to CO2 emissions and CO2 emissions contribute to GDP per capita.Results also show unidirectional causal relationships between POP and CO2,EX and FDI,IM and EX,FDI and IM,GDP and IM which mean there is one way influence of independent variables on dependent variables.VAR results for panel data of thirteen low income countries identify both unidirectional and bidirectional causal relationships.Results disclose bidirectional causal relationship between FDI and import which denotes FDI influences import and import also affects FDI in low income countries.Results also reveal that there is a unidirectional causal relationship between export and import;besides unidirectional causal relationship exists between import and population which means independent variables affect dependent variables in one way only.It is worthy to note that CO2 emissions do not affect other variables and similarly other variables do not affect CO2 emissions in low income countries.These findings suggest that CO2 emission is not a sensitive issue in the sample low income countries,most probably due to poor FDI,GDP and international trade over there.VAR results for panel data of twenty lower middle income countries diagnose both unidirectional and bidirectional causal relationship.Results show a bidirectional causal relationship between GDP and export which means both variables affect each other.In contrast,results also show unidirectional causal relationships between export and CO2 emissions,FDI and CO2 emissions,GDP and CO2 emissions,import and export,population and export,GDP and import,FDI and GDP.More specifically,export,FDI,and GDP influence CO2 emissions,import and population affect export,GDP influences import and finally FDI influences GDP.These results suggest that CO2 emissions and major economic variables are more sensitive in lower middle income countries than low income countries.VAR results for twenty six upper middle income countries reveal only unidirectional causal relationships between GDP and CO2 emissions,GDP and export,GDP and FDI which means GDP affects CO2 emissions,exports and FDI in one way only.Finally,VAR results for thirty nine high income countries expose bidirectional causal relationships between GDP and export,population and GDP that means they affect each other.Results also find unidirectional causal relationships between imports and CO2 emissions,population and CO2 emissions,GDP and FDI,GDP and imports,population and imports,CO2 emissions and GDP which mean only independent variables affect dependent variables one way only.Based on the above results and discussion,we can claim that results significantly differ among the different income group wise panel data sets which are expected also.Policy Implications and Conclusion:Finally,study comes up with some pragmatic solutions about how CO2 emissions can be reduced substantially for ensuring a better and safe world in the future,First,innovating environmental friendly alternative sources of energies and developing energy efficiency technologies which do not emit excessive CO2 or any harmful elements to the environment at all will be the best strategy to combat the excessive CO2 emissions problem.In fact,necessity is the mother of invention and where there is a will there is a way;these quotations might dare to put the best efforts finding the best alternative sources of energy and developing energy efficient technologies.Eventually two hundred years back there was no usage of fossil fuel on earth and threats for the environment.Hence,alternatives to fossil fuel and CO2 emitting energy might be solar power,wind power,clean renewable energy and any other environmentally friendly energy in the future days.In this regard,scientists,experts,environmentalists,governments and policymakers;all the stakeholders should relentlessly work together and put their best efforts for the sake of sustainable development and a safe planet.Second,strict rules and regulations must be enacted,promulgated and implemented against excessive CO2 emissions and in favour of CO2 emissions reduction by the respective government and international organizations who works for environmental safety;for instance Intergovernmental Panel of Climate Change,United Nations Environment Program,World Meteorological Organization,Global Environment Facility,World Wide Fund for Nature,United Nations,World Health Organization and so on.Third,awareness concerning harmful impacts and consequences of excessive CO2 emissions must be raised among global leaders,politicians,policy makers,businessmen,and the public;so that they can contribute to reduce CO2 emissions from their respective positions.Fourth,apart from enhancing public awareness and enacting rules,imposing carbon taxes might help to reduce CO2 emissions.Carbon tax should be imposed on manufacturing and non-manufacturing firms involved in excessive CO2 emission which is also suggested by many researchers,experts and policymakers including Telli et al.(2008),Filipovic(2004)who suggested imposing a carbon tax.Therefore,countries can impose a certain carbon tax which is appropriate for the respective countries to curb CO2 emissions.In this context,Bluffstone(2003)suggests utilizing collected carbon tax revenue towards addressing environmental issues which will improve the quality of the environment.Moreover,each country should incorporate eco-friendly environmental concems and philosophy into her macroeconomic policies more seriously in order to control CO2 emissions.Finally,government and policymakers should encourage usage of environmentally friendly technologies and welcome FDI that will spread advanced and environmental friendly technologies across the countries.By putting the above initiatives into action and effective implementation together we can have a better and safe place for living in the future days. |
PDF Full Text Request