Energy consumption and real GDP in asean

Introduction

Problem statements

Energy is crucial for both human life and production activities, with its demand increasing significantly in recent years This growing demand has led to conflicts over energy resources, such as oil reserves in oceans, highlighting the importance of energy in the global economy As a result, understanding the relationship between energy consumption and economic growth has become essential, prompting economists to conduct extensive research on this vital connection.

Numerous studies have explored the impact of energy on economic growth, with varying conclusions over the years While some researchers, such as Stern (1993) and Cleveland et al (2000), identified energy as a crucial factor influencing economic growth, others, including Solow (1974) and Cheng (1995), argued that economic growth is largely independent of energy, citing negligible energy costs (Belloumi, 2009) Traditional economic growth models, as noted by Stern (2011) and Pirlogea and Cicea (2012), often exclude energy from their frameworks, assuming a substitutable relationship between energy and capital Furthermore, research by Akarca and Long (1979; 1980), Glasure and Lee (1998), and Masih and Masih (1996; 1997; 1998) highlights the complex relationship between energy consumption and economic growth, which varies by country and time period However, Hong, Wijeweera, and B Charles (2013) point out two limitations in existing literature: the absence of a combined energy-based and mainstream economic model and the potential for omitted variable biases.

Numerous studies have explored the relationship between energy consumption and Gross Domestic Product (GDP) in developed countries, revealing both unidirectional and bidirectional connections In contrast, research on developing countries presents mixed findings; while some studies indicate a significant relationship, others suggest no correlation exists Notably, in the ASEAN region, the GDP of ten developing countries has shown consistent growth since 1974.

Between 1974 and 2014, several ASEAN countries, including Myanmar, the Philippines, Singapore, and Vietnam, experienced significant GDP growth alongside rapid increases in energy consumption Conversely, other nations, despite lower GDP growth, also saw rising energy use This study aims to explore the relationship between energy consumption and economic growth across ten ASEAN countries during this period.

Research objectives

This study aimed to determine the relationship between energy consumption and real GDP in Asean.

Finding the factors those influence on real GDP.

Is there the relationship between energy consumption and real GDP in case of countries in Asean?

What are factors those influence on real GDP and how ?

Scope of study

The study will be collected data of Asean from Word Bank which has the period from

This study analyzes the relationship between energy consumption and real GDP across ten ASEAN countries from 1974 to 2014 It incorporates explanatory variables such as imports, exports, CO2 emissions, capital, and human capital The research employs various methodologies, including panel cointegration tests, panel unit root tests, Granger causality tests using Vector Error Correction Models (VECM), and Fully Modified Ordinary Least Squares (FMOLS) to derive its findings.

This study is structured into six chapters: Chapter One introduces the topic, while Chapter Two provides a literature review focusing on growth theory and empirical studies Chapter Three details the data collection process and the methodology employed in the research Chapter Four examines real GDP and energy consumption in ASEAN, analyzing individual countries as well as the aggregate data The findings are presented in Chapter Five, and Chapter Six concludes the study.

The structure of study

Many papers in the past studied about the growth models, such as Neoclassical growth models, ecological economic models and endogenous growth models.

In his 1956 study, Solow introduced a growth model focusing on capital and labor as the primary factors of production, while deeming energy to have a minimal correlational role His model is recognized as a key example of Neoclassical growth theory, emphasizing the significance of these two inputs in economic growth.

Y= F(K,L) Where Y presents rate of production, K is stock of capital and L is labor.

Endogenous models, including those by Romer (1986), Galor and Weil (2000), and Lucas (2002), highlight the significance of human capital in driving economic growth and its crucial contribution to real GDP.

Stern's 2011 paper highlights the significance of energy in growth models, while omitting the contributions of traditional inputs such as capital and labor This approach positions Stern’s model within the framework of ecological economics, emphasizing the pivotal role of energy in sustainable growth.

This paper presents a comprehensive model derived from various empirical studies, integrating key factors from neoclassical growth theory, ecological economics, and endogenous growth models It incorporates capital, labor, energy, and human capital into a production function approach Notably, research by Stern (1993, 2000) and Shahiduzzaman and Alam (2012) demonstrates that this production function methodology offers a more thorough framework for analysis.

Literature Reviews

Growth theory

Many papers in the past studied about the growth models, such as Neoclassical growth models, ecological economic models and endogenous growth models.

In his 1956 study, Solow introduced a growth model that emphasized capital and labor as the primary factors of production, while downplaying the role of energy His work is a key contribution to Neoclassical growth theory, highlighting the importance of these two inputs in understanding economic growth dynamics.

Y= F(K,L) Where Y presents rate of production, K is stock of capital and L is labor.

Endogenous models, including the works of Romer (1986), Galor and Weil (2000), and Lucas (2002), highlight the significance of human capital as a crucial factor influencing real GDP These studies underscore the vital role that human capital plays in economic growth and development.

Stern's 2011 paper emphasizes the significance of energy in economic growth models, while excluding traditional inputs such as capital and labor This approach positions Stern’s model within the framework of ecological economics.

This paper presents a comprehensive model that integrates key elements from various empirical studies, including capital, labor, energy, and human capital, as outlined in neoclassical growth theory and ecological economic models By employing a production function approach, this model synthesizes insights from notable works, including those by Stern (1993, 2000) and Shahiduzzaman and Alam (2012), demonstrating that this methodology offers a more holistic understanding of economic growth dynamics.

2.2.1 nergy, economic growth and real GDP

Numerous studies have explored the connection between energy consumption and economic growth Kraft and Kraft (1978) analyzed U.S data from 1947 to 1974 and concluded that increased gross national product (GNP) leads to higher energy consumption In contrast, Akarca and Long (1979) found that greater energy consumption correlates with higher employment levels, but their subsequent 1980 study revealed no causal relationship between energy consumption and GNP Further research, such as Erol and Yu (1987a), also indicated no link between energy consumption and employment using data from 1973 to 1984 However, Murray and Nan (1992) found that increased energy consumption boosts employment Erol and Yu (1987b) examined causal relationships between energy consumption and real GNP across various countries, revealing a bidirectional relationship in Japan, a GNP-to-energy consumption relationship in Germany and Italy, an energy-to-GNP relationship in Canada, and no causal links in France and the U.K.

Research by Masih and Masih (1996), Soytas and Sari (2003), Yoo (2005, 2006a, 2006b, 2006c), Yoo and Jung (2005), Chen et al (2007), and Zachariadis (2007) indicates that in Malaysia, Singapore, and the Philippines, there is no causal relationship between energy consumption and economic growth However, a bidirectional relationship exists in Pakistan, South Korea, and Taiwan, while Indonesia experiences increased growth leading to higher energy consumption, and the opposite is true for India, Thailand, and Sri Lanka The findings from Masih and Masih (1996, 1997, 1998) reveal contrasting results across different countries Additionally, Stern (2000) identified a unidirectional causality in the United States, and Soytas and Sari (2003) reported varied outcomes across nations; specifically, no causal relationship in Canada, Indonesia, Poland, the United Kingdom, and the United States, bidirectional causality in Argentina and Turkey, unidirectional causality in France, West Germany, and Japan, and mixed results in Italy and South Korea, with a bidirectional relationship noted in Canada.

Previous studies on the relationship between energy consumption and output or employment face the issue of omitted variable bias due to reliance on bivariate causality tests, leading to potentially misleading statistical results (Stern, 2000; Payne, 2010) In their 1984 research, Yu and Hwang addressed this by incorporating employment into their model, utilizing both Sims and Granger causality tests Their findings indicated that increased employment drives higher energy consumption, while no causal link was found between energy consumption and GNP Similarly, Stern's 1993 study acknowledged this issue and included employment and capital in his model, demonstrating that growth in real GDP correlates with increased energy consumption.

In their 2008 study, Narayan and Smyth investigated the relationship between capital formation, energy consumption, and real GDP Their findings, derived from panel cointegration analysis, revealed a significant cointegration among these variables Additionally, they identified a positive long-run causal relationship, indicating that both capital and energy consumption contribute to the growth of real GDP.

A study conducted in 2009 revealed a cointegration between per capita GDP and per capita energy consumption The research, which analyzed panel data from 88 countries spanning from 1975 to 2003, found a significant two-way causality in both the short and long run between GDP growth and energy consumption growth.

Research by Adhikari and Chen (2012) demonstrates a long-run relationship between economic growth and energy consumption across 80 countries, categorized into upper middle, lower middle, and low-income groups Utilizing panel unit root, panel cointegration, and panel dynamic ordinary least squares methods, the study found that all groups exhibited a cointegrated relationship Notably, in low-income countries, economic growth was identified as a driver of energy consumption, while upper and lower middle-income countries displayed a unidirectional relationship where energy consumption influences economic growth.

2.2.2 papers that applied different methodologies

Over the past decade, traditional OLS methods have been commonly used to estimate parameters and conduct statistical tests; however, these methods fail to account for the unique characteristics of time series data, such as potential endogeneity and non-stationarity, leading to misleading results (Granger and Newbold, 1974) To address these issues, new econometric techniques, including the Engle-Granger (1987) and Johansen-Juselius (1990) cointegration and error-correction models, have been developed to better examine the relationship between energy consumption and economic growth For instance, Francis et al (2007) applied these models in their research on Haiti, Jamaica, and Trinidad and Tobago, revealing a bidirectional causality between energy consumption and economic growth Conversely, Oh and Lee (2004a, 2004b) found inconsistent conclusions regarding this relationship in their studies on Korea, highlighting the importance of using appropriate methodologies.

Lee (2005), Yet Chen et al (2007), Mehrara (2007), Narayan and Smyth (2007), Lee and Chang (2008) and Lee et al (2008) applied Panel unit root and cointegration

1 7 tests in their paper According Payne (2010), this approach provides additional power by merging the cross-section and time series data allowing for the heterogeneity across countries.

Harris and Sollis (2003) highlighted limitations in traditional cointegration methods like Engle-Granger and Johansen-Juselius for analyzing the relationship between energy consumption and economic growth, particularly concerning low power and size properties in small samples To address these issues, newer models such as the autoregressive distributed lag (ARDL) model and bounds testing approach, along with the Toda-Yamamoto and Dolado-Lütkepohl long-run causality tests, have been utilized in recent studies For instance, Altinay and Karagol (2005) discovered unidirectional causality in Turkey using the Dolado-Lütkepohl test, while Lee (2006) reported varying results across countries with the Toda-Yamamoto test, indicating no causal relationship in Germany, Sweden, and the UK, a bidirectional relationship in the US, and a unidirectional effect in Belgium, Canada, and Switzerland Zachariadis (2007) employed both the ARDL bounds test and Toda-Yamamoto causality test, revealing conflicting results regarding the causal links between sector-specific energy consumption and income/output measures in several countries including Canada, France, Germany, Italy, Japan, the UK, and the US.

In their 2008 study, Huang et al utilized dynamic panel estimation to analyze energy consumption across different income levels They categorized their data into three groups: low-income, middle-income, and high-income panels The findings revealed that there was no causal relationship between energy consumption and real GDP per capita in the low-income panel.

1 8 causing economic growth positively for the second group, and energy consumption causing economic growth negatively for the last group.

In a study by Hung-Pin (2014), vector error correction models and autoregressive distributed lag bounds testing were utilized to examine the causal relationship between economic growth and renewable energy consumption across nine OECD countries from 1982 to 2011 The findings revealed varied results by country: the USA exhibited both long-run and strong unidirectional causality from economic growth to renewable energy, while Germany and the UK showed long-run and strong unidirectional causality from renewable energy to economic growth Additionally, the USA and Japan demonstrated long-run unidirectional causality from economic growth to renewable energy In contrast, Germany, Italy, and the UK indicated long-run unidirectional causality from renewable energy to economic growth, with Italy and the UK also displaying short-run unidirectional causality from economic growth to renewable energy.

2.2.3 papers that was added more variables

Several researchers have enhanced their models by incorporating additional variables, leading to varying outcomes Notable studies by Yu and Jin (1992), Cheng (1996), Paul and Bhattacharya (2004), and Pirlogea and Cicea (2012) included measures of capital and labor, while Glasure and Lee (1995) introduced wages and energy prices, followed by real money supply and government spending in 1996 These investigations revealed contrasting findings: some indicated no long-term cointegration or causal relationship between energy consumption and economic growth (Yu and Jin, 1992; Cheng, 1996), whereas others identified a bidirectional relationship (Glasure and Lee, 1995, 1996; Paul and Bhattacharya, 2004).

Review of empirical studies

This paper examines variables derived from existing literature, focusing on real gross domestic product (real GDP) as the dependent variable The independent variables include energy consumption, capital, human capital, imports, exports, and CO2 emissions The energy input (E) represents primary energy usage prior to its transformation into other fuels, calculated as indigenous production plus imports and stock changes, minus exports and fuels supplied to international transport Energy consumption data is aggregated and measured in kilotons of oil equivalent Capital input (K) is defined by gross capital formation, which includes costs for fixed asset extensions and net inventory changes Human capital (H) is assessed through educational processes, measured by tertiary education enrollment as a percentage of the total population Research by Amiri and Gerdtham indicates a long-run unidirectional causality from exports and imports to economic growth, leading to their inclusion in the model The import (IM) variable captures the value of goods and services received from abroad, while the export (EM) variable reflects the value of goods and services provided to other countries Additionally, Papież (2013) identifies a bidirectional causality between CO2 emissions and economic growth, prompting the inclusion of CO2 emissions (C) in the model, which are measured in metric tons per capita from fossil fuel combustion and cement production.

Data and Methodology

Data

This paper examines variables from existing literature, focusing on real gross domestic product (real GDP) as the dependent variable, while the independent variables include energy consumption, capital, human capital, imports, exports, and CO2 emissions The energy input variable (E) represents primary energy usage prior to conversion into other fuels, calculated as indigenous production plus imports and stock changes, minus exports and fuels supplied for international transport Energy consumption data is aggregated and expressed in kilotons of oil equivalent Capital input (K) reflects gross capital formation, encompassing costs for fixed asset extensions and net inventory changes Human capital (H) is assessed through educational processes, specifically measuring total enrollment in tertiary education as a percentage of the total population According to Amiri and Gerdtham, there exists a long-run unidirectional causality from exports and imports to economic growth, thus both variables are included in the model The import (IM) variable accounts for the value of all goods and services received from abroad, while the export (EM) variable represents the value of goods and services provided to other countries Papież (2013) indicates a bidirectional causality between CO2 emissions and economic growth, leading to the inclusion of CO2 emissions (C), measured in metric tons per capita from fossil fuel combustion and cement production Data will be sourced from the World Bank (2015), covering the period from 1974 to 2014, with definitions and descriptions presented in Tables 2 and 3.

Variable Variable name Unit Definition

Gross Domestic Product Million US Dolar Gross Domestic Product is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources.

IM Import Million US Dolar This variable is represent the value of all goods and other market services received from the rest of the world.

EX Export Million US Dolar This variable represent the value of all goods and other market services provided to the rest of the world.

K Capital Million US Dolar The gross capital formation that is the cost of extensions to the fixed assets of the economy plus net changes in the level of inventories.

Human capital represents the percentage of the population engaged in education and training, specifically reflected in the enrollment rates of tertiary education This metric is crucial for understanding the overall educational attainment within a society, highlighting the importance of investing in human capital for economic development.

EC energy consumption, measured in kilograms of oil, represents the primary energy available before it is converted into other fuels This metric can be calculated by adding indigenous production, imports, and stock changes, then subtracting exports and the fuels supplied to ships and aircraft for international transport.

C CO2 emission Metrictons per capita The variable is that stemming from the burning of fossil fuels and the manufacture of cement.

Table 2 : Descriptive statistics on data

(million USD) (kg of oil) (million USD) (percented (million USD) tons per (million USD) of total ) capita)

Methodology

The initial step involves testing for the presence of a unit root in the variables While numerous studies have explored unit root testing, this paper employs the Levin et al (2002) test, commonly known as the LLC test This method is based on the Augmented Dickey-Fuller (ADF) test, with the hypothesis focused on determining the unit root characteristics of the data.

At the beginning, he performed ADF regressions for each cross-section I, that is:

Which alternative hypothesis corresponds to � �� being stationary.

In the context of an LLC model, we consider T time periods and N panel members, where Δ denotes the first difference operator and Y represents the dependent variable The term ε signifies white noise, while p indicates the lag order, which can be determined using Hall's method from 1990.

After that, LLC ran two auxiliary regression ∆� �� and ∆� ��−1 to generate the two residuals those are ê �� and �̂ ��−1 He calculated the regression standard error

�̂ �� in (2) and normalized these two residuals by �̂ �� aiming to control for the heterogeneity across individuals.

The long-run variance for Model (1) can be estimated under the null hypothesis of a unit root by calculating the ratio of long-run to short-run standard deviations.

In this equation, LLC said that K is the truncation lag parameter and the sample covariance weights � �̅� depend on the choice of Bartlett kernel

The long-run standard deviation to the innovation standard deviation for each country i was defined and estimated that:

And the average standard deviation ratio is:

� �=1 Final, he computed the panel test statistic by running the pool regression: ẽ �� = ��̃ ��−1 + � ��

And the test base on regression t-statistic for testing � = 0 with �

Following the unit root test, we will employ Pedroni's (1999) methodology to examine the presence of cointegration, indicating a long-run relationship among the variables Pedroni introduced seven panel cointegration statistics, of which four are specifically categorized as panel cointegration statistics.

LLC said that those are based on pooling along what is commonly referred to as the within- dimension:

Panel non-parametric (PP) t-Statistic

The group mean panel cointegration statistics, as noted by the author, are derived from pooling methods that focus on the between-dimension.

Group non-parametric (PP) t-Statistic:

In the first step, he estimated the panel cointegration regression:

According to Pedroni, T represents the number of observations over time, N denotes the number of individual members in the panel, and M indicates the number of regression variables Subsequently, the residuals are collected for further analysis.

The second step, computing the residual for the differenced regression

The third step, using the Newey-West (1987) estimator to calculate the long-run variance

The last step, he divided seven statistics above into two parts those are non-parametric and parametric statistic In part one: With the non-parametric those are

� � , � � , � �� , �̃ � , �̃ �� Using the residual �̂ �,� that calculated from the first step for computing the long-run variance of �̂ �,� that is denoted �̂ � After that, computing �̂ � and �̂ � :

In term of part two: For the parametric those are � �� , �̃ �� Estimating �̂ �,� = �̂ �

∑ �=1 �̂ �,� ∆�̂ �,�−� + �̂ �,� and compute �̂ � that is the simple variance of �̂ �,�

This section analyzes the causality direction between real GDP and energy consumption within a panel context, following the application of the panel unit root test and panel cointegration test.

In the absence of cointegration among the variables, it indicates a lack of long-term relationship between them Consequently, this paper will employ the vector autoregression (VAR) model, originally developed by Granger in 1969, to analyze the data.

Cointegration between variables indicates a long-run relationship, necessitating the estimation of a panel vector error correction model (VECM) to conduct Granger-Causality tests The two-stage procedure developed by Engle and Granger (1987) is utilized to analyze both short-run and long-run relationships between the variables In the initial step, the long-run parameters of the model are estimated to derive the residuals.

�� ,� is the error correction term that derived from the long-run relationship of Eq

(7).In the second step, estimating the Granger Causality test with the dynamic vector error correction model:

Where ∆ denotes the first difference of the variable and q is the lag length that is chosen by Akaike information criterion (AIC), Schwarz information criterion (SC) and Hannan-

Quinn information criterion Besides that, real GDP denoted GDP, C is CO2 emission, K is capital, H is human capital, and EC is energy consumption, IM is import and EX is export.

3.2.4 Fully modified ordinary least square (FMOLS)

To estimate the long-run parameters of a cointegration relationship, the Fully Modified Ordinary Least Squares (FMOLS) technique can be employed Developed by Pedroni (2001, 2004), the FMOLS estimator effectively addresses issues of endogeneity and serial correlation in the regressors According to Pedroni, this method provides a robust framework for analyzing long-term relationships between variables.

Where t = 1,….,T and I = 1,… N � �,� and � �� according Pedroni that they are must be cointegrated with slope � � � � is allowed may or may not be homogeneous across i.

Overview of Energy Consumption

Individual country

Figure 2 : The real Gross Domestic Product in Brunei from 1974 to 2014

Figure 3 : The Energy Consumption in Brunei from 1974 to 2014

The real GDP of Brunei has shown a consistent upward trend from 1974 to 2014, with notable peaks in 1980, 2008, 2012, and 2014, despite some fluctuations In 2014, the real GDP reached its highest point, reflecting overall growth during this period Meanwhile, energy consumption in Brunei remained stable from 1976 to 2005, but experienced significant growth after 2005, culminating in a rapid increase that reached a peak in subsequent years.

2013 Two figure also show that at the peak of the real GDP energy consumption of this country also increasing.

Figure 4 : The real Gross Domestic Product in Cambodia from 1974 to 2014

Figure 5 : The Energy Consumption in Brunei from 1974 to 2014

Between 1995 and 2014, Cambodia experienced a slight but stable increase in real GDP, despite the overall growth being minimal In contrast, energy consumption during this period showed significant fluctuations, highlighting the challenges associated with data collection and the issues of missing data.

From 1995 to 2003, energy consumption showed a steady increase; however, there was a significant decline from 2003 to 2006 This trend shifted dramatically in 2009, when energy consumption surged, peaking in 2014 The period between 2004 and 2008 may have experienced challenges related to energy supply, yet it is evident that energy consumption and real GDP followed a similar trend throughout these years.

Figure 6 : The real GDP in Indonesia from 1974 to 2014

Figure 7 : The Energy Consumption in Indonesia from 1974 to 2014

From 1974 to 2014, Indonesia experienced a notable rise in both real GDP and energy consumption, as illustrated in figures 5 and 6 However, the rate of increase in energy consumption outpaced that of real GDP, indicating a significant reliance on energy resources as the country's economy grows.

Figure 8 : The real GDP in LAO PDR from 1974 to 2014

Lao faces a significant challenge due to the lack of available energy consumption data According to Figure 7, real GDP in Laos showed a gradual increase from 1984 to 2007, followed by a dramatic surge between 2008 and 2014.

Figure 9 : The real GDP in Malaysia from 1974 to 2014

Figure 10 : The Energy Consumption in Indonesia from 1974 to 2014

Figures 8 and 9 illustrate the trends in real GDP and energy consumption in Malaysia from 1974 to 2014 During this period, Malaysia experienced a consistent rise in real GDP, while energy consumption saw an even more pronounced increase, doubling within just seven years This indicates that energy consumption grew at a faster rate than real GDP during this timeframe.

Figure 11 : The real GDP in Myanmar from 1974 to 2014

Figure 12 : The Energy Consumption in Myanmar from 1974 to 2014

From 1974 to 1995, Myanmar's real GDP was notably low, but after 1995, it experienced significant growth, increasing twofold by 1999, fourfold by 2002, and eightfold by 2004 compared to 1995 levels In contrast, energy consumption remained stable from 1974 to 1984, saw a rise in 1985 and 1986, and then decreased from 1987 to 1991 Following 1993, energy consumption began to rise again, peaking in 2007 This indicates that while real GDP steadily increased during this period, energy consumption exhibited instability.

Figure 13: The real GDP in Philippines from 1974 to 2014

Figure 14 : The Energy Consumption in Philippines from 1974 to 2014

The real Gross Domestic Product (GDP) of the Philippines has shown a consistent upward trend, as illustrated in Figure 12 Notably, from 2000 to 2010, the real GDP nearly tripled, reflecting significant economic growth This upward trajectory continued from 2010, with the GDP increasing nearly twofold, highlighting the ongoing economic expansion in the country.

Between 1974 and 2012, energy consumption exhibited significant fluctuations, initially rising from 1974 to 1983, then declining until 1993 It peaked in 2000, followed by a downward trend from 2001 to 2009, before experiencing a slight increase in the last three years of the period.

Figure 15: The real GDP in Singapore from 1974 to 2014

Figure 16 : The Energy Consumption in Singapore from 1974 to 2014

From 1974 to 2004, Singapore's real Gross Domestic Product (GDP) showed a slight increase, followed by a dramatic rise of nearly twofold between 2005 and 2010, peaking in 2014 In contrast, energy consumption exhibited fluctuations, with an upward trend from 1974 to 1994, followed by a decline from 1995 to 2003 Energy consumption peaked in 2004 before experiencing a decline until 2014.

Figure 17: The real GDP in ThaiLand from

There is a same trend in real

GDP and energy consumption in Thailand In detail, the figure

14 show that the real GDP in ThaiLand increased from

1994 to 1996, and have a slight decreasing after

1996 but continue strongly increasing by two times in

2013 Besides that, energy consumption also have a incline tendency This figures increased from 1974 to 1997 and slight deline in 1998 and continue increasing to reach a peak at 2014.

Figure 19: The real GDP in Viet Nam from 1974 to 2014

Figure 20 : The Energy Consumption in Thailand from 1974 to 2014

Figures 18 and 19 illustrate a consistent trend between real GDP and energy consumption in Vietnam Specifically, Figure 18 demonstrates that Vietnam's real GDP has steadily increased from 1985 onwards, highlighting the correlation between economic growth and energy usage in the country.

From 2005 to 2014, Vietnam experienced significant economic growth, with real GDP in 2014 reaching double the value of 2008 Energy consumption also showed a notable upward trend, particularly from 2001 to 2010, culminating in a peak in 2014 that was twice the level of 2005 In contrast, energy consumption remained stable from 1974 to 1997, highlighting the dramatic changes in the country's energy needs and economic development over the past decade.

Country in aggregate

This section provides a comprehensive analysis of the real GDP and energy consumption of ten ASEAN countries, highlighting key statistics such as the mean, median, minimum, and maximum values, which are detailed in Tables 3 and 4 below.

Table 3 : Description Energy Consumtion in Asean from 1974 to 2014 (kg of oil equivalent)

BRUNEI CAMBODIA INDONESIA MALAYSIA MYANMAR PHILIPPINES SINGAPORE THAILAND VIETNAM

Table 4 : Description Gross Domestic Produc in Asean from 1974 to 2014 (million USD)

BRUNEI CAMBODIA INDONESIA LAO_PDR MALAYSIA MYANMAR PHILIPPINES SINGAPORE THAILAND VIETNAM

Table 3 illustrates the energy consumption trends in ASEAN from 1974 to 2014 During this period, Brunei exhibited the highest energy consumption, with a mean of 7,032.932, a median of 6,995.904, a maximum of 9,695.714, and a minimum of 3,846.213 In contrast, Myanmar recorded the lowest energy consumption, with a mean of 272.630, a median of 270.893, a maximum of 307.443, and a minimum of 246.102.

During the analyzed period, Indonesia achieved the highest Gross Domestic Product (GDP) with a total of 10,235,737, a mean of 249,652, and a median of 140,001 Notably, Myanmar recorded its peak real GDP at 1,563,760, while Laos reported the lowest real GDP, with a mean of 3,272 and a median of 1,769 Despite Indonesia's impressive GDP figures, the country utilized less energy compared to others Conversely, Brunei, despite having a lower GDP, consumed a significantly higher amount of energy during the same period.

Results

Panel Unit Root test

The LLC (2002) test will be utilized to assess the presence of a unit root or stationarity in the data The null hypothesis posits that a unit root exists, while the alternative hypothesis suggests its absence The optimal lag length will be determined automatically using the Schwarz Information Criterion, and the test will be conducted under the assumption of an intercept without a trend Results will be presented in Table 5.

Table 5: Result of panel LLC (2002) unit root test

Table 2 presents the results of the LLC (2002) unit root test, indicating that all variables, including real GDP (15.0878, p-value 1.0000), energy (5.27597, p-value 1.0000), capital (9.44829, p-value 1.0000), human capital (3.87612, p-value 0.9999), export (12.9803, p-value 1.0000), import (9.60736, p-value 1.0000), and CO2 emissions (4.44319, p-value 1.0000), are non-stationary at their levels However, after taking the first difference, all variables become stationary at the 1% significance level, indicating that they are integrated of order one, I(1).

Panel cointegration test

When all variables are integrated of order one, I(1), the Pedroni (1999) test can be applied to assess cointegration among them This test determines the optimal lag length using the Schwarz Information Criterion and assumes the presence of an intercept without a trend The null hypothesis posits that no cointegration exists between the variables, while the alternative hypothesis suggests that cointegration is present The findings will be presented in Table 6.

Table 6: Panel Pedroni (1999) cointegration test

Notes: ***,**,* indicate statistical significance at 1%, 5%, and 10% level of significance, respectively

Table 3 indicates that three Within-dimension statistics are significant at the 1% level: the panel rho-Statistic (0.0034), panel PP-Statistic (0.0000), and panel ADF-Statistic (0.0000) Additionally, two Between-dimension statistics are also significant at the 1% level: the group PP-Statistic (0.0006) and group ADF-Statistic (0.0034) These results lead to the rejection of the null hypothesis and the acceptance of the alternative hypothesis, confirming the presence of cointegration among the variables Consequently, it can be concluded that a long-run equilibrium relationship exists between the variables, indicating that changes in one factor will affect the other over time.

Granger causality test

The presence of cointegration indicates a long-run relationship among the variables, prompting the estimation of a panel vector error correction model (VECM) to conduct Granger-Causality tests The findings are presented in Table 7 below.

Table 7: Result of Granger causality test with VECM

Independent variable Dependent variable ΔGDP ΔGDP ΔEnergy 0.898 (0.5320) ΔCapital 2.559 (0.0333) ΔHuman capital 0.257 (0.9742) ΔExport 0.82 (0.5916) ΔImport 0.523 (0.8277) ΔCO 2 emission 1.067 (0.4146)

Dependent variable ΔEnergy ΔGDP 2.089 (0.0744)* ΔEnergy ΔCapital 3.679 (0.0054) ΔHuman capital 1.9 (0.1032) ΔExport 3.513 (0.0070) ΔImport 2.506 (0.0364) ΔCO 2 emission 4.952 (0.0008)

Table 4 indicates a significant short-run causal relationship from capital to real GDP at the 5% level, suggesting that changes in capital can lead to changes in real GDP This finding aligns with Narayan and Smyth (2008), who also identified capital formation as a Granger cause of real GDP Additionally, the analysis reveals no long-run causal relationship between energy consumption and real GDP, indicating that fluctuations in energy consumption do not impact real GDP in the short run.

In the short run, changes in human capital, exports, imports, and CO2 emissions do not have a causal impact on real GDP This indicates that fluctuations in these factors do not significantly affect the real GDP within a brief timeframe.

In the long run, there is no causal relationship between energy consumption and real GDP, indicating that changes in energy consumption do not result in changes in real GDP over time.

The energy model reveals a significant short-run relationship between real GDP and energy consumption at the 10% level, as well as a long-run causal relationship at the 1% level, indicating that changes in real GDP influence energy consumption over both time frames Additionally, in the short run, there are significant causal relationships from capital, exports, and CO2 emissions to energy consumption, with significance levels of 1% for capital and exports, 5% for imports, and 1% for CO2 emissions Conversely, human capital does not exhibit any causal relationship with energy consumption in the short run, suggesting that changes in human capital do not affect energy consumption during this period.

Fully modified ordinary least square (FMOLS)

The panel cointegration test indicates a significant cointegration relationship among the variables, allowing us to estimate the long-run parameters of the model using the Fully Modified Ordinary Least Square (FMOLS) technique The results are presented in Table 8 below.

Table 8: Result of Fully modified ordinary least square

All variables are measure in natural logarithms.

Table 5 demonstrates that, in the long run, capital, energy, and exports positively influence real GDP at a 1% significance level, while human capital also contributes positively to real GDP.

5% significant level, negative effect from CO2 emission to real GDP at 1% significant level and negative effect from import to real GDP at 5% significant level.

An increase of 1% in capital leads to a 0.558% rise in real GDP, indicating that higher investment in physical capital positively impacts long-term economic growth This finding aligns with Narayan and Smyth (2008), which states that capital formation is a significant contributor to real GDP Observations show that developed countries invest more in physical capital, a trend that may also be reflected in developing countries within ASEAN.

Human capital plays a crucial role in the economy, demonstrated by a significant positive relationship with real GDP An increase of 1% in human capital correlates with a 0.026% rise in real GDP While this impact may be less than that of physical capital, human capital remains a vital economic factor Therefore, it is essential for developing countries to prioritize investments in education, as this will contribute to long-term growth in real GDP.

Numerous studies have explored the relationship between energy consumption and real GDP, revealing significant findings Research by Chien-Chiang Lee (2005) indicates both long-run and short-run causalities from energy consumption to GDP, while Narayan and Smyth (2008) demonstrate that energy consumption positively Granger-causes real GDP in the long run Additionally, Glasure and Lee (1998) identify a bidirectional causality between GDP and energy consumption Utilizing Fully Modified Ordinary Least Squares (FMOLS) analysis, it is found that a 1% increase in energy consumption leads to a 0.293% increase in real GDP, highlighting the importance of energy consumption for GDP growth in ASEAN countries However, given the environmental drawbacks of non-renewable energy sources, policymakers must implement strategies to promote the use of renewable energy to support sustainable economic growth.

The findings indicate a significant positive relationship between exports and real GDP, with a 1% increase in exports resulting in a 0.616% rise in real GDP This suggests that developing countries in ASEAN should focus on enhancing their exports of goods and services globally to stimulate economic growth.

By contrast, there is negative effect of export to real GDP with high signigicant level The result shows that when export increased 1% that lead to real GDP decreasing

0.205% that means the countries in ASEAN should decrease import the product and services from the other countries if they want to increase their real GDP in the long run.

The findings indicate a significant negative impact of CO2 emissions on real GDP, with a 1% increase in emissions resulting in a 0.421% decline in GDP For ASEAN countries to promote long-term economic growth, it is crucial to reduce CO2 emissions Policymakers should implement regulations to manage and limit the emissions produced by firms.

In the long run, a 1% increase in capital, energy, exports, and human capital contributes to real GDP growth of 0.558%, 0.293%, 0.617%, and 0.0262%, respectively Conversely, a 1% rise in CO2 emissions and imports results in a decline in real GDP by 0.421% and 0.205%, respectively.

Conclusions and Policy Implications

Conclusions

The goal of this paper is investigate the relationship between energy consumption and real GDP in Asean The panel Asean data was collected from World Bank from 1974 to

2014 and applying the panel unit root test LLC (2002), panel cointegration test Pedroni

(1999) and VECM Granger causality test.

The result was got from the cointegration tests is that there is a cointegration relationship or long-run relationship between the variables.

The VECM Granger causality test reveals a long-run causal relationship from real GDP to energy consumption, alongside a short-run unidirectional causality in the same direction Additionally, the test indicates short-run causality from capital to real GDP, as well as from capital, exports, imports, and CO2 emissions to energy consumption Notably, there is no evidence of causality from energy consumption to real GDP, suggesting that policies aimed at controlling or limiting energy consumption in ASEAN countries will not adversely affect real GDP in either the short or long run Conversely, increases in real GDP are likely to drive energy consumption over time.

The findings from the fully modified ordinary least squares analysis indicate that variables such as capital, energy, exports, and human capital have a significant positive impact on real GDP in ASEAN countries This suggests that increased investment in these areas can lead to long-term GDP growth Conversely, the analysis also reveals a significant negative effect of CO2 emissions and imports on real GDP, highlighting the importance for these countries to minimize CO2 emissions and imports to enhance their long-term economic performance.

Non-renewable energy plays a crucial role in both the economy and daily life, yet its limited availability and negative environmental impact raise concerns Research indicates that there is no direct causal relationship between energy consumption and real GDP in both the short and long run Therefore, it is essential to implement policies aimed at regulating non-renewable energy consumption, particularly among businesses.

There exists both a long-term and short-term causal relationship between real GDP and energy consumption Therefore, policymakers should implement strategies that promote the use of renewable energy sources, such as solar, wind, and hydro energy, which are environmentally friendly and help conserve non-renewable resources.

The policy makers should have some strategy to encourage the investment of capital, human capital and increasing export to make the increasing real GDP in the future.

The thesis reveals significant gaps in data sourced from the World Bank regarding ASEAN countries, including the absence of energy consumption statistics for Laos and human capital data for Singapore, along with incomplete information for other nations Additionally, while labor is a crucial factor, its data remained static at a baseline level, leading to its exclusion from the model.

This thesis focuses on a single method for testing panel unit root and panel cointegration due to the author's knowledge limitations, allowing for a thorough examination of the variables' characteristics.

Akarca, A.T., Long, T.V., 1979 Energy and employment: a time series analysis of the causal relationship Resources and Energy 2(2–3), 151–162.

Akarca, A.T., Long, T.V., 1980 Relationship between energy and GNP: a reexamination. Journal of Energy and Development 5 (2), 326–331.

Altinay, G., Karagol, E., 2005 Electricity consumption and economic growth: evidence from Turkey Energy Economics 27(6), 849–856.

Asafu-Adjaye, J., 2000 The relationship between energy consumption, energy prices, and economic growth: time series evidence from Asian developing countries Energy Economics 22(6), 615–625.

Bowden, N., Payne, J.E., 2010 Sectoral analysis of the causal relationship between renewable and non-renewable energy consumption and real output in the US Energy Sources, Part B: Economics, Planning, and Policy 5(4), 400–408.

Chen, S.T., Kuo, H.I., Chen, C.C., 2007 The relationship between GDP and electricity consumption in 10 Asian countries Energy Policy 35(4), 2611–2621.

Cheng, B.S., 1996 An investigation of cointegration and causality between energy consumption and economic growth Journal of Energy and Development 21(1), 73–84.

Dolado, J.J., Lütkepohl, H., 1996 Making wald tests work for cointegrated VAR systems Econometric Theory 15(4), 369–386.

Engle, R.F., Granger, C.W.J., 1987 Co-integration and error correction: representation, estimation, and testing Econometrica, 55(2), 251-276.

Erol, U., Yu, E.S.H., 1987a Time series analysis of the causal relationships between US energy and employment Resources and Energy 9(1), 75–89.

Erol, U., Yu, E.S.H., 1987b On the causal relationship between energy and income for industrialized countries Journal of Energy and Development 13(1), 113–122.

Francis, B.M., Moseley, L., Iyare, S.O., 2007 Energy consumption and projected growth in selected Caribbean countries Energy Economics 29(6), 1224–1232.

Ghali, K.H., El-Sakka, M.I.T., 2004 Energy and output growth in Canada: a multivariate cointegration analysis Energy Economics 26(2), 225–238.

Ghosh, S., 2002 Electricity consumption and economic growth in India Energy Policy 30(2), 125–129.

Glasure, Y.U., 2002 Energy and national income in Korea: further evidence on the role of omitted variables Energy Economics 24(), 355–365.

Glasure, Y.U., Lee, A.R., 1995 Relationship between US energy consumption and employment: further evidence Energy Sources 17(5), 509–516.

Glasure, Y.U., Lee, A.R., 1996 The macroeconomic effects of relative prices, money, and federal spending on the relationship between US energy consumption and employment. Journal of Energy and Development 22(1), 81–91.

Glasure, Y.U., Lee, A.R., 1998 Cointegration, error correction, and the relationship between GDP and energy: the case of South Korea and Singapore Resource and Energy Economics 20(1), 17–25.

Granger, C.W.J., Newbold, P., 1974 Spurious regressions in econometrics Journal of Econometrics 2(2), 111–120.

Harris, R., Sollis, R., 2003 Applied Time Series Modelling and Forecasting Wiley, Chichester.

Hondroyiannis, G., Lolos, S., Papapetrou, E., 2002 Energy consumption and economic growth: assessing the evidence from Greece Energy Economics 24(2), 319–336.

Huang, B.N., Hwang, M.J., Yang, C.W., 2008 Causal relationship between energy consumption and GDP growth revisited: a dynamic panel data approach Ecological Economics 67(1), 41–54.

Johansen, S., Juselius, K., 1990 Maximum likelihood estimation and inference on cointegration with applications to the demand for money Oxford Bulletin of Economics and Statistics 52(2), 169–210.

Kraft, J., Kraft, A., 1978 On the relationship between energy and GNP Journal ofEnergy and Development 3(2), 401–403.

Lee, C.C., 2005 Energy consumption and GDP in developing countries: a cointegrated panel analysis Energy Economics 27(3), 415–427.

Lee, C.C., 2006 The causality relationship between energy consumption and GDP in G-

Lee, C.C., Chang, C.P., 2008 Energy consumption and economic growth in Asian economies: a more comprehensive analysis using panel data Resource and Energy Economics 30(1), 50–65.

Lee, C.C., Chang, C.P., Chen, P.F., 2008 Energy-income causality in OECD countries revisited: the key role of capital stock Energy Economics 30(5), 2359–2373.

Masih, A.M.M., Masih, R., 1996 Energy consumption, real income and temporal causality: results from a multi-country study based on cointegration and error-correction modelling techniques Energy Economics 18(3), 165–183.

In their 1997 study, Masih and Masih explore the temporal causal relationships among energy consumption, real income, and prices in Asian energy-dependent newly industrialized countries (NICs) Utilizing a multivariate cointegration and vector error correction approach, the authors provide new evidence that enhances our understanding of how these economic factors interact over time Their findings contribute valuable insights to policy modeling in the context of energy economics.

Masih, A.M.M., Masih, R., 1998 A multivariate cointegrated modeling approach in testing temporal causality between energy consumption, real income, and prices with an application to two Asian LDCs Applied Economics 30(10), 1287–1298.

Mehrara, M., 2007 Energy consumption and economic growth: the case of oil exporting countries Energy Policy 35(5), 2939–2945.

Narayan, P.K., Smyth, R., 2007 Energy consumption and real GDP in G7 countries: new evidence from panel cointegration with structural breaks Energy Economics 30(5), 2331–2341.

Oh, W., Lee, K., 2004a Causal relationship between energy consumption and GDP revisited: the case of Korea 1970-1999 Energy Economics 26(1), 51–59.

Oh, W., Lee, K., 2004b Energy consumption and economic growth in Korea: testing the causality relation Journal of Policy Modeling 26(8–9), 973–981.

Paul, S., Bhattacharya, R.N., 2004 Causality between energy consumption and economic growth in India: a note on conflicting results Energy Economics 26(6), 977–983.

Payne, J.E., 2010 Survey of the international evidence on the causal relationship between energy consumption and growth Journal of Economic Studies 37 (1), 53–95.

Pirlogea, C., Cicea, C., 2012 Econometric perspective of the energy consumption and economic growth relation in European Union Renewable and Sustainable Energy Reviews 16(8), 5718–5726.

Sari, R., Ewing, B.T., Soytas, U., 2008 The relationship between disaggregate energy consumption and industrial production in the United States: an ARDL approach Energy Economics 30(5), 2302–2313.

Shiu, A., Lam, P.L., 2004 Electricity consumption and economic growth in China. Energy Policy 32(1), 47–54.

Stern, D.I., 1993 Energy and economic growth in the USA: a multivariate approach. Energy Economics 15 (2), 137–150.

Stern, D.I., 2000 A multivariate cointegration analysis of the role of energy in the US macroeconomy Energy Economics 22 (2), 267–283.

Stern, D.I., 2011 The role of energy in economic growth Ecological Economics Reviews

Solow, R M., 1974 Intergenerational equity and exhaustible resources Review of Economic Studies 41: Symposium on the Economics of Exhaustible Resources, 29–46.

Soytas, U., Sari, R., 2003 Energy consumption and GDP: causality relationship in G-7 and emerging markets Energy Economics 25(1), 33–37.

Soytas, U., Sari, R., 2006a Energy consumption and income in G7 countries Journal of Policy Modeling 28(7), 739–750.

Soytas, U., Sari, R., 2006b Can China contribute more to the fight against global warming? Journal of Policy Modeling 28(8), 837–846.

Soytas, U., Sari, R., 2007 The relationship between energy and production: evidence from Turkish manufacturing industry Energy Economics 29(6), 1151–1165.

Toda, H.Y., Yamamoto, T., 1995 Statistical inference in vector autoregressions with possibly integrated processes Journal of Econometrics 66(1–2), 225–250.

Yoo, S.H., 2005 Electricity consumption and economic growth: evidence from Korea. Energy Policy 33(12), 1627–1632.

Yoo, S.H., 2006a Causal relationship between coal consumption and economic growth in Korea Applied Energy 83(11), 1181–1189.

Yoo, S.H., 2006b Oil consumption and economic growth: evidence from Korea Energy Sources, Part B 1(3), 235–243.

Yoo, S.H., 2006c The causal relationship between electricity consumption and economic growth in the ASEAN countries Energy Policy 34(18), 3573–3582.

Yoo, S.H., Jung, K.-O., 2005 Nuclear energy consumption and economic growth in Korea Progress in Nuclear Energy 46(2), 101–109.

Yoo, S.H., Kim, Y., 2006 Electricity generation and economic growth in Indonesia. Energy 31(14), 2890–2899.

Yu, E.S.H., Jin, J.C., 1992 Cointegration tests of energy consumption, income, and employment Resources and Energy 14(3), 259–266.

Yuan, J., Kang, J., Zhao, C., Hu, Z., 2008 Energy consumption and economic growth: evidence from China at both aggregated and disaggregated levels Energy Economics 30(6), 3077–3094.

Zachariadis, T., 2007 Exploring the relationship between energy use and economic growth with bivariate models: new evidence from G-7 countries Energy Economics 29(6),1233–1253.

Limitations of the study

The thesis reveals significant gaps in data collected from the World Bank regarding ASEAN countries, including missing energy consumption statistics for Laos and human capital data for Singapore, along with incomplete information for several other nations Additionally, while labor is a crucial factor, its data remained stationary at a constant level, leading to its exclusion from the model.

This thesis focuses on a single method for testing panel unit root and panel cointegration due to the author's limited knowledge, allowing for a thorough examination of the variables' behavior.