Different impacts of industrialization on thecarbon emissions across seven asean countries

The research explores the complex interactions between industrialization and its impact on carbon emissions in seven ASEAN countries: Brunei, Indonesia, Malaysia, the Philippines, Singap

VIETNAM NATIONAL UNIVERSITY HO CHI MINH CITY

UNIVERSITY OF ECONOMICS AND LAW

PROJECT SCIENTIFIC RESEARCH OF STUDENTS

DIFFERENT IMPACTS OF INDUSTRIALIZATION ON THE CARBON EMISSIONS ACROSS SEVEN ASEAN COUNTRIES

Class: Economic Mathematics (222MA9308)

Lecturer: Dr Pham Hoang Uyen, Ms Vo Thi Le Uyen

Group:

No

1 Nguyen Ngoc Thao Nguyen K214020180 International

Economics Relation Leader

2 Le Phuong Thao K214020183 Economics RelationInternational Member

Economics Relation Member

ABSTRACT 3

CHAPTER 1: INTRODUCTION 3

1.1 Background 3

1.2 Research questions 4

1.3 Research subjects and scope of the study 4

1.3.1 Research subjects 4

1.3.2 Research scope 4

1.4 Previous research 5

CHAPTER II: METHODOLOGY 5

2.1 Research model 5

2.1.1 GDP per capita and carbon emissions 5

2.1.2 Export and carbon emissions 6

2.1.3 Import and carbon emissions 6

2.1.4 Technology innovation and carbon emissions 6

2.2 Research data 6

2.3 Implementation and results 7

2.3.1 Method 1: Using the logarithms form 7

2.3.2 Method 2: Using min, max formula 10

2.3.3 Method 3: Detecting outliers 12

2.3.3.1 Keeping outliers 13

2.3.3.2 Removing outliers 16

CHAPTER III: RESULTS AND DISCUSSION 20

CHAPTER IV: CONCLUSION 20

REFERENCES 21

The research explores the complex interactions between industrialization and its impact on carbonemissions in seven ASEAN countries: Brunei, Indonesia, Malaysia, the Philippines, Singapore,Thailand, and Vietnam, from 1990 to 2017 was investigated The results show a nuancedunderstanding of industrialization It explores the impact of GDP per capita on CO2 emissions,highlighting the different results observed depending on the level of economic development andindustrial structure Furthermore, the study reveals a complex relationship between trade openness andcarbon emissions, with countries exhibiting different emission patterns depending on their export andimport profiles Furthermore, the study highlights the important role of technological innovation inreducing CO2 emissions and highlights the importance of R&D efforts and the introduction of cleanertechnologies These results will deliver valuable insights to policymakers in the ASEAN region andprovide guidance in developing effective strategies to promote sustainable development, reduce carbonemissions, and drive technological progress towards a greener future.

Keywords: Carbon emissions, GDP per capita, export, import, technology innovation, ASEAN

CHAPTER 1: INTRODUCTION

1.1 Background

In recent years, the dangers of environmental pollution and climate change have increasingly posed asignificant risk to both human health and the ecosystem The economic development andindustrialization have been associated with the excessive consumption of fossil fuels and otherresources This growth has raised concerns among some scholars about the ecological deterioration(Chen et al., 2020a), particularly the substantial release of carbon dioxide (CO2) that is likely to takeplace mostly in seven ASEAN countries (Brunei, Indonesia, Malaysia, Philippines, Singapore,Thailand, and Vietnam)

Exhibit 1: CO2 emissions in seven ASEAN's countries during 1990–2017 (unit: kiloton)Between 1990 and 2017, the ASEAN region's carbon emissions showed significant fluctuations,influenced by various socio-economic and environmental conditions Certain countries have recordedrelatively stable carbon emissions, mainly in industries such as oil and gas Rapid industrialization andurbanization have led to significant increases in emissions in some areas The presence of strongmanufacturing and export sectors has contributed to significant increases in emissions in somecountries Emissions are also increasing in other countries due to population growth and industrialexpansion On the other hand, countries with smaller, service-oriented economies showed relativelystable emission patterns While the manufacturing and energy sectors have played important roles indriving emissions growth in certain countries, rapid industrialization and urbanization have contributed

to large increases in emissions in others These diverse developments highlight the importance ofimplementing sustainable strategies to address environmental challenges across the ASEAN region

1.2 Research questions

This paper aims to address the following research inquiries:

(1) Does industrialization have a negative influence on the environment in many ASEAN nations? (2) What are the disparities in the composition of the industrial sector among the seven ASEANcountries, and how do these disparities contribute to variations in carbon emissions?

(3) What are the primary factors that account for discrepancies in carbon emissions duringindustrialization across the chosen ASEAN countries?

1.3 Research subjects and scope of the study

1.3.1 Research subjects

The differential impact of industrialization on carbon emissions in seven ASEAN countries viavariation data evidence over the period from 1990 to 2017

Table 1: Definition of the data

1.3.2 Research scope

Due to the availability of the data, this study examines the scope of 7 ASEAN countries from

a major driver of carbon emissions globally, the specific effects and nuances of industrialization oncarbon emissions in ASEAN countries have not been extensively studied Previous research hasexplored the relationship between industrialization and carbon emissions in individual ASEANcountries, such as Indonesia (Setyawan et al., 2018), Thailand (Santikarn et al., 2020), and Vietnam (Le

et al., 2021) However, comparative studies that examine the differences and similarities in the impacts

of industrialization on carbon emissions across multiple ASEAN countries are lacking Such studies areessential because the factors that drive carbon emissions in one country may not be the same in anothercountry due to differences in industrial structures, economic conditions, and environmental policies.For instance, Vietnam has experienced rapid industrialization and urbanization in recent years, leading

to a surge in energy consumption and carbon emissions However, the country's policy focus onrenewable energy and energy efficiency has the potential to mitigate the negative impacts ofindustrialization on the environment and support sustainable development goals These variations maylead to different levels and patterns of carbon emissions across ASEAN countries, making it necessary

to explore the specific factors and mechanisms that influence the relationship between industrializationand carbon emissions within each country In a related study, Salman et al (2019) investigated theeffects of exports and imports on carbon emissions in ASEAN countries, but their findings do notdirectly address the research gap on the impacts of industrialization Therefore, further comparativeresearch is warranted to explore and analyze the different patterns, factors, and outcomes ofindustrialization on carbon emissions within the ASEAN context, offering valuable insights to informsustainable development practices and environmental policies in the region

CHAPTER II: METHODOLOGY

2.1 Research model

2.1.1 GDP per capita and carbon emissions.

Researchers and environmentalists have paid a lot of attention to studying the relationship betweenGDP per capita and carbon emissions The link between per capita GDP and carbon emissions is likeinverted-U shape curve, which is termed as the Environmental Kuznets Curve (EKC) (Kuznets, 1955;

Grossman and Krueger, 1991, 1995) Nevertheless, in this study, we mainly focused on 7 ASEANcountries (Brunei, Indonesia, Malaysia, Philippines, Singapore, Thailand, and Vietnam).

Hypothesis a: GDP per capita might have a negative relationship with polution levels

2.1.2 Export and carbon emissions.

In 1998, Schmalensee and his partners argued that a rise in export levels may lead to the depletion ofnatural resources, which would increase the carbon emissions from the combustion of fossil fuel andhence, consequent deterioration in environmental quality According to Sadorsky (2012), import andexport activities affect the consumption of energy in local areas

Hypothesis b: Export might have a negative relationship with polution levels

2.1.3 Import and carbon emissions.

Imported goods require a substantial network of transportation This network requires fuel to transportthe goods to various destinations Therefore, an increase in imported goods would requiretransportation machinery to consume more fuel and thus, raise carbon emissions Liddle (2018) statedthat imports have a considerable impact on consumption-based emissions in a panel of 102 countriesover a time period from 1990 to 2013

Hypothesis c: Import might have a positive relationship with polution levels

2.1.4 Technology innovation and carbon emissions.

Sohag et al (2015) stated that technology innovation helps to reduce energy consumption byaugmenting the energy efficiency of production operations Moreover, the study by Samargandi (2017)found that technology innovation plays an insignificant role in reducing carbon emissions in SaudiArabia The research conducted by Samargandi (2017) indicated that technological innovation has littleimpact on reducing carbon emissions in Saudi Arabia

Hypothesis d: Technology innovation might have a positive relationship with polution levels

Figure 1: The relationship between industrialization factors and carbon emissions

We derived the data from the World Development Indicators (WDI) covering the time period from

1990 to 2017 The dependent variable is carbon dioxide emissions (CO2) whereas, the independentvariables include GDP per capita (GDPpc), export (Ex), import (Im) and technology innovation (TI)

Table 2: Summary of the data from 1990 to 2017

2.3 Implementation and results

Before running the model, we chose 3 following distinguished methods to check the availability of thefollowing data:

2.3.1 Method 1: Using the logarithms form

As the data has “time series” issue, we use “d.variable” to create missing values before develop theeconometric model of carbon emissions following the logarithm form for seven ASEAN countries:

ln(d.CO2) = + ln(d.GDPpc) + ln(d.Ex) + ln(d.Im) + ln(d.TI)

To analyze the variances, we built the ANOVA table for logarithms, then infer the equation:

Table 3: The ANOVA table.

ln(d.CO2) = 57,48 - 10,03ln(d.GDPpc) – 0,0003ln(d.Ex) + 0,0004ln(d.Im) + 0,48ln(d.TI) (*)

It can be seen from the equation (*) that GDP per capita and export both have negative impacts on thecarbon emissions, while import and technology innovation show positive influences on the pollutionlevels, when other variable keeps constant respectively However, whether the data is reliable or notwould be checked through the assumptions about the error term :ԑ

Constant variance: In the “Breausch–Pagan / Cook–Weisberg test for heteroskedasticity”, although thep-value is relatively small, notwithstanding the null hypothesis is still rejected that the variance is notconstant (the probability is smaller than 0,05) This is also proved by the plot that the pattern of the datapoints is slightly narrow despite some outliers (Figure 2) VIOLATED

Figure 2: The plot of residuals

Normality: The p-value in the Shapiro-Wilk W test, which is 0.95, indicates that the error term is aԑnormally distributed random variable Similarly, the Kernel density plot shows that the error term isԑquite close to a normal distribution despite a trivial deviation from normality (Figure 3) The normal Q-

Q plot also proves the same result (Figure 4) NOT VIOLATED

Figure 3: The Kernel density plot Figure 4: The normal Q-Q plot of the residuals.Independence: The value of for a particular set of values for the independent variables is related toԑthe value of for any other set of valuesԑ (Figure 5) VIOLATED

Figure 5: The scatter plot of standardized residuals

Mean of ԑ: The error term is a random variable with mean or expected value of zero, that is E( ) = 0ԑ ԑ NOT VIOLATED

As two out of four assumptions are violated, then the results are unreliable or even misleading

2.3.2 Method 2: Using min, max formula

We apply the following formula to create new data:

x’ = Unlike the first method, we develop the econometric model of carbon emissions for seven ASEANcountries without using the logarithms:

CO2 = + GDPpc + Ex + Im + TI

To analyze the variances, we built the ANOVA table for logarithms, then infer the equation:

Table 4: The ANOVA table

CO2 = 0,22 - 0,58GDPpc – 1,64Ex + 1,81Im + 0,28TI (**)

It can be seen from the equation (**) that GDP per capita and export both have negative impacts on thecarbon emissions, while import and technology innovation show positive influences on the pollutionlevels when other variable keeps constant respectively However, whether the data is reliable or notwould be checked through the assumptions about the error term :ԑ

Constant variance: In the “Breausch–Pagan / Cook–Weisberg test for heteroskedasticity”, the p-value

is very small, which is equal to 0 Hence, the null hypothesis is rejected that the variance of the error

term is not constant This can also be seen in the plot that the pattern of the data points is not goingԑthe same way (Figure 6) VIOLATED

Figure 6: The plot of residuals

Normality: In the Shapiro-Wilk W test, the error term is not normally distributed as the p-value isԑvery small (0.00) Correspondingly, the Kernel density plot shows that the error term is not close to aԑnormal distribution but slightly follow a right skewness (Figure 7) The normal Q-Q plot also provesthe points on the plot clearly depart from a straight line (Figure 8) VIOLATED

Figure 7: The Kernel density plot Figure 8: The normal Q-Q plot of the residuals.

Independence: The value of for a particular set of values for the independent variables is related toԑthe value of for any other set of valuesԑ (Figure 9) VIOLATED

Figure 9: The scatter plot of standardized residuals

Mean of ԑ: The error term is not a random variable with mean or expected value of approximatelyԑzero, that is E( ) = 0.004 ԑ VIOLATED

As all of the assumptions are violated, then the results are unreliable or even misleading

2.3.3 Method 3: Detecting outliers.

In this method, we simultaneously focus on the influence of outliers and compare the differencebetween the results when keeping and eliminating the outliers The mean and standard deviation areutilized in this method by the following formula:

x’ = The econometric model of carbon emissions for seven ASEAN countries is developed:

CO2 = + GDPpc + Ex + Im + TI

The first step in our study is to build the graph box of all the variables to detect the outliers After that,

we would divide into two options, including keeping and removing outliers and make a comparison

Figure 10: The graph box of five variables

2.3.3.1 Keeping outliers.

We built the ANOVA table, then infer the following equation: