INTRODUCTION
Problem statement
Energy is essential for economic activities, leading to significant interest among researchers in the relationship between energy consumption and economic growth, particularly as industrial activity has become increasingly crucial for growth While economists focus on this nexus, climate activists are concerned about energy consumption's environmental impact, especially from nonrenewable sources, which contribute to greenhouse gas emissions and drive global warming and climate change.
Since the Kyoto Protocol came into effect in 2005, developed economies have faced increased pressure to address their significant contributions to greenhouse gas (GHG) emissions The United Nations Framework Convention on Climate Change (UNFCCC) mandates that these countries reduce their collective GHG emissions by 5.2% from 1990 levels, with a target of a 29% reduction by 2010 Achieving these goals is essential for mitigating the adverse environmental impacts of economic activities.
The Kyoto Protocol's effectiveness in addressing global warming and climate change is hindered by conflicts among major economies, as energy conservation policies are seen to significantly impact their economic performance Key emitters such as the United States, China, and India have declined to sign the protocol, fearing a loss of competitive advantage over non-ratifying countries.
Governments often hesitate to acknowledge the impacts of climate change due to concerns about reducing greenhouse gas (GHG) emissions and its potential effects on economic growth U.S leaders have contended that adhering to the Protocol could result in the loss of approximately five million jobs and a significant decline in gross domestic product (GDP) (Broehl, 2005) However, as criticism mounts and countries that comply with the Protocol face severe natural disasters annually, major nations will increasingly find it difficult to overlook the necessity of the Protocol.
Despite opposition from some industrialized nations, over 140 countries, including the European Union, have ratified the Kyoto Protocol and implemented new energy policies to meet their emission targets Renewable energy technologies are proving to be the most effective means for these countries to satisfy their increasing energy demands while achieving global greenhouse gas reduction goals The transition from nonrenewable to renewable energy is not just about meeting emission levels; it is becoming a catalyst for technological advancement and improved energy efficiency The International Renewable Energy Agency projects that enhanced renewable energy policies could double the share of renewables in global energy consumption by 2030 at no additional cost, while the Intergovernmental Panel on Climate Change estimates that 80% of the global energy supply could be derived from renewable sources by 2050.
Renewable energy policies are rapidly advancing worldwide due to their environmental benefits and potential cost savings The Renewable Energy Policy Network for the 21st Century (REN21) highlights this significant growth in the sector.
Over 100 countries, including major economies, have established national goals for renewable energy generation and consumption (Broehl, 2005) This shift from nonrenewable to renewable energy, particularly in developed nations, has sparked significant research in the field of energy economics.
Numerous studies have explored the link between energy consumption and economic growth, primarily focusing on general energy consumption Research utilizing time series data has consistently demonstrated a significant positive relationship between energy consumption and GDP, as evidenced by findings from Stern (2000) and Stresing, Lindenberger, and Kummel (2008), among others Similarly, panel data analyses conducted by Lee and Chang (2007), Narayan and Smyth (2008), and Apergis and Payne (2009) have corroborated these results.
Alper et al (2013) confirmed positive associations between energy and economic growth using micro data from 47 US states In 2010, Apergis and Payne focused on renewable energy and found similar results However, Apergis and Payne (2012) and Tugcu, Ozturk, and Aslan (2012) expanded their analysis to include both renewable and nonrenewable energy, revealing differing outcomes While Apergis and Payne identified a significant positive correlation between both energy types and GDP, Tugcu and colleagues reported mixed results when applying classic and developed production models.
Despite variations in samples, data, and models across empirical studies, the primary research methods employed—cointegration analysis and Granger causality tests based on Cobb-Douglas production functions—remain consistent Notably, these studies often overlook the relationship between renewable and nonrenewable energy consumption, failing to determine whether they act as substitutes or complements Given the similarity in analytical approaches, the results tend to converge Furthermore, the limited number of studies focusing on renewable energy is disproportionate to its increasing significance in economic activities and environmental protection.
This research aims to analyze the impact of nonrenewable and renewable energy consumption on the GDP of 34 OECD countries, employing a novel approach compared to previous studies Utilizing a multiple-input, one-output stochastic distance function, this study focuses solely on GDP as the output, diverging from earlier research that considered multiple outputs Unlike prior studies that relied on micro data from U.S electric companies, this research will utilize macro data from OECD countries The stochastic distance function will facilitate the estimation of the effects of both energy types on GDP, with a particular emphasis on the partial effects between nonrenewable and renewable energy consumption Additionally, the study will compute average technical efficiency, efficiency change, technical change, and productivity change across OECD nations.
Objective and research questions
This study aims to estimate the impact of both nonrenewable and renewable energy consumption on GDP Utilizing a quantitative approach with panel data from OECD countries, the research seeks to answer three key questions.
1 How does nonrenewable and renewable energy consumption affect OECD countries’ GDP?
2 What is the relationship between nonrenewable energy consumption and renewable energy consumption in OECD countries?
3 How does the productivity of OECD countries change in the whole estimated period? tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Scope of the thesis
This study will mainly examine the effect of nonrenewable, renewable energy consumption on GDP in 34 OECD countries, utilizing the panel data from 1990 to
In 2012, all energy-related data was sourced from the U.S Energy Information Administration (EIA), while GDP, capital, and labor statistics were obtained from the World Bank's World Development Indicators database.
Structure of the thesis
The research consists of six chapters, beginning with a literature review in Chapter 2 that examines the economic and environmental effects of energy consumption through empirical studies, while also introducing the stochastic distance function used in this thesis Chapter 3 details the properties of the stochastic distance function and the methodology for calculating productivity changes, followed by an overview of energy consumption and supply in OECD countries Chapter 5 presents the empirical results, divided into two sections: the first addresses data-related issues, and the second discusses regression and computation outcomes Finally, Chapter 6 summarizes the key findings and offers policy implications and suggestions for future research expansion.
LITERATURE REVIEW
Definition and classification of energy
Energy exists in various forms, leading to multiple definitions based on research contexts, such as thermal and nuclear energy (Nigg, MacIntosh, and Mester, 2000) The most prevalent definition describes energy as "the ability to do work" (US EIA, 2015) In physics, energy is understood as a property of objects that can be transferred or converted into different forms, but it cannot be created or destroyed (Kittel and Kroemer, 1980) This aligns with the First Law of Thermodynamics, which states that the total energy of a system remains constant unless altered by mechanical work or heat transfer, emphasizing that energy cannot be created or annihilated (Denker, 2013).
Energy can be classified in various ways depending on the context of study In classical mechanics, energy is divided into two main types: kinetic energy, which is associated with the motion of an object, and potential energy, which is related to an object's position within a force field or the configuration of a system's components.
Energy can be classified into various types, including thermal, radiant, kinetic, electrical, chemical, nuclear, and gravitational energy However, these classifications often overlap, as certain forms of energy, such as thermal energy, can contain both kinetic and potential energy components.
In this study, basing on its renewable ability, energy is classified as nonrenewable and renewable energy The descriptions of the two sources will be delivered in two following sections
Nonrenewable energy, as defined by the US EIA (2015), refers to energy sources that cannot be replenished quickly, making them finite resources.
Nonrenewable energy is derived from four primary sources: crude oil, natural gas, coal, and uranium (nuclear energy) The first three, known as fossil fuels, were formed millions of years ago from the remains of ancient plants and animals, subjected to heat and pressure over time These resources are consumed faster than they can be replenished, making their extraction increasingly costly Uranium, the fourth source, is utilized in nuclear power plants where its atoms are split to produce heat and electricity.
Most of energy used in daily living or production activities is acquired from nonrenewable sources, for example, 90% of total energy consumed in the US in
2014 is nonrenewable energy (US EIA, 2015) The huge and continual demand for nonrenewable energy, especially petroleum, is originated from the invention of internal combustion engines in the 17 th century Nowadays, in spite of the creation of new green technologies, infrastructure and transportation systems which use combustion engines are still globally prominent The ceaseless consumption of nonrenewable energy at the current rate is recognized as the primary cause for global warming and climate change (National Research Council, 2010)
Renewable energy is an energy source which can be easily and naturally recreated over a short time scale Different from limited energy sources like fossil fuels, humanity can replenish renewable energy from biological regeneration or other natural recurring mechanisms (US EIA, 2015)
Renewable energy is generated from five major sources below:
Solar energy, transformed to electricity and heat
Geothermal energy, radiated from heat from the earth’s core
Biomass from plants, including trees’ firewood, corn’s ethanol, vegetable oil’s biodiesel
Hydropower as known as water power, generated from falling water or fast running water through hydroelectric turbines
Renewable energy sources are widely distributed across various geographical areas, unlike petroleum, which is concentrated in specific regions like the Middle East Large-scale renewable energy projects are particularly beneficial for rural and remote areas, as they can drive economic growth in these communities (Leone, 2001).
Since 2004, global renewable energy production has experienced annual growth rates between 10% and 60%, particularly in wind power technology, significantly impacting global energy consumption A 2014 REN21 report indicated that renewable energy comprised 19% of total global energy consumption, with traditional biomass contributing 9%, heat energy 4.2%, hydroelectricity 3.8%, and the remaining 2% sourced from wind power, solar energy, geothermal energy, and biomass (Sawin et al., 2014).
Energy consumption and growth
Classical production models traditionally overlook energy as a crucial element in driving economic growth, focusing instead on capital and labor as the primary factors of production In these models, energy is often regarded merely as an intermediate factor rather than a fundamental component.
The rapid development of energy-consuming equipment has made energy a crucial factor for economic growth, drawing significant attention from energy economists Research by Cleveland, Costanza, and Kaufmann (1997) highlights that future economic performance will heavily depend on net energy production from various fuel sources, suggesting that classical economic models may need adjustments to account for biophysical constraints Beaudreau (1995) criticized traditional growth models for undervaluing energy's importance, asserting that production cannot function without it By incorporating energy into the model, he found that the gap between output growth rates and aggregate input growth rates, known as the Solow residual, was nearly eliminated, indicating that combined input growth indexes could largely explain manufacturing growth in the US, Germany, and Japan.
From an engineering economics perspective, Pokrovski (2003) noted that manual labor is increasingly being replaced by energy-driven machines across various sectors of modern economies, leading to production inputs being defined by capital, labor, and energy services Supporting this view, Thompson (2006) emphasized that energy serves as a crucial production input that integrates physical capital and labor to generate aggregate output.
To be concluded, modern economic activities require energy as a compulsive input
Excluding energy consumption out of augmented production function would result in a lack of judgment (Lee and Chang, 2007)
In the effort of demonstrating the role of energy use in economic growth, Apergis and Payne (2009, 2012); Arbex and Perobelli (2010); Lee and Chang (2007, 2008);
Narayan and Smyth (2008), Stern (2000), Stresing, Lindenberger, and Kummel (2008), and Yuan et al (2008) have identified four key hypotheses regarding the energy-growth relationship The first, known as the growth hypothesis, posits that energy consumption plays a crucial role in driving economic growth, suggesting that a reduction in energy use could negatively impact growth Consequently, energy policies should focus on promoting green energy to mitigate pollution The second hypothesis, the conservation hypothesis, indicates that economic growth can lead to increased energy consumption, implying that energy-saving policies may not necessarily hinder growth.
The feedback hypothesis illustrates a reciprocal relationship between GDP growth and energy use, evidenced by a bi-directional nexus between the two Implementing energy conservation policies may lead to a decline in economic growth, while economic performance can influence overall energy consumption Conversely, the neutrality hypothesis posits that energy consumption does not significantly affect growth, as there is no causality between the two In this context, energy policies aimed at reducing consumption would not impact economic growth.
Empirical research on the relationship between energy consumption and economic growth often employs cointegration analysis and Granger causality tests within an expanded production model that includes capital, labor, energy consumption, and additional factors such as Research and Development and Education The primary aim of these studies is to investigate the long-term cointegrating relationship and the causal dynamics between GDP and energy consumption.
Numerous studies confirm a positive long-run cointegration between GDP and energy consumption, although the causality differs based on the samples analyzed For instance, Stern (2000) demonstrated a bi-directional relationship between GDP and energy using US time series data from 1948 to 1994, a finding echoed by Alper et al (2013) in their analysis of annual data from 47 US states between 1997 and 2009 It is important to note that while a bi-directional relationship may exist in the short run, it may not hold in the long run, and vice versa Specifically, Yuan et al (2008) found that both aggregated energy consumption and disaggregated consumption of coal and oil positively influence total output in the long run, while GDP positively impacts total energy, coal, and oil usage, but only in the short run.
On the other hand, one – way effect from the use of energy on economic performance is the most frequent result derived from studies such as Lee and Chang
Research by Narayan and Smyth (2008) and Apergis and Payne (2009) analyzed panel data from 16 Asian countries, G7 nations, and Central America, revealing a significant positive impact of energy consumption on GDP Specifically, a 1% increase in energy consumption correlates with a GDP rise of 0.12-0.39% in G7 countries.
In conclusion, the majority of researchers support the growth hypothesis, while some provide evidence for the feedback hypothesis, and only a minority endorse the conservation and neutrality hypotheses This highlights the significant impact of energy on economic growth.
2.2.2 Environmental effects of energy consumption
Energy consumption, while contributing positively to economic growth, is primarily responsible for global warming and climate change due to greenhouse gas (GHG) emissions from fossil fuel use In 2013, approximately 32 billion tons of carbon dioxide (CO₂) were released into the atmosphere from burning fossil fuels, along with additional air pollution The detrimental effects on the global environment and human health are estimated to cost the world $4.9 trillion per ton.
CO₂ is responsible for an estimated loss of 150 billion US dollars (Ottmar, 2015) and is one of six greenhouse gases that significantly contribute to radiative forcing and global warming This increase in average surface temperature leads to climate change, which poses severe risks such as food and water shortages, rising sea levels, and ongoing flooding These impacts threaten the lives of billions, particularly in developing countries (Intergovernmental Panel on Climate Change, 2007).
Energy consumption not only harms the environment but also poses significant risks to human health, primarily through air pollution.
The exploitation and combustion of solid fuels, such as coal and biomass, pose significant health risks, particularly for impoverished households with limited access to green fuels and electricity (Smith et al., 2013).
The environmental consequences of energy extend beyond consumption to include the exploitation processes involved A clear example is the harvesting of firewood for charcoal production, which results in deforestation due to overharvesting This destruction eliminates vital atmospheric protection, specifically the CO₂-absorbing cover Furthermore, uncontrolled harvesting adversely affects biodiversity and contributes to soil erosion (Rowan, 2009).
Considering the externalities associated with nonrenewable energy consumption, particularly from fossil fuels, the cost of electricity generation from coal or oil could potentially double, while the cost from natural gas may increase by 30%.
(Dones et al., 2005) On the other hand, with increasing demand of energy to satisfy economic as well as living activities, energy resources have been exhausted
Consequently, nonrenewable energy is no longer free source but more and more expensive because the exploitation becomes costly eventually It is a serious threat to energy security
Therefore, the searching for new energy sources, which is not only easy to be refilled up but also friendly to the environment, is an urgent issue to all nations
Renewable energy has been widely considered as the sustainable source which can satisfy the production demand and environmental protection requirements
Nonrenewable and renewable energy consumption and economic growth
The growing adoption of renewable energy in economic activities, particularly in developed nations, has captured the interest of economists regarding its impact Building on methodologies from prior energy economics research, many scholars have utilized cointegration and Granger causality tests to examine the effects of renewable energy consumption on economic performance.
Apergis and Payne (2010) put renewable energy consumption, represented by renewable electricity consumption, into the production side model of a panel data of
A study of 11 countries in the Eurasian region found a long-run equilibrium between GDP and renewable energy consumption, indicating a bi-directional causality between these variables in both the short and long run Utilizing the fully modified ordinary least squares method for heterogeneous cointegrated panels, the authors revealed that a 1% increase in renewable energy usage would result in a 0.195% rise in real GDP.
In 2012, Apergis and Payne expanded their research to include both nonrenewable and renewable energy sources across a sample of 80 countries, yielding results consistent with their 2010 study They identified a long-run cointegrating relationship and bidirectional causality between energy consumption—both renewable and nonrenewable—and GDP growth Their findings indicate that both energy types have a statistically significant and positive impact on economic growth, with a 1% increase in energy consumption correlating with substantial economic benefits.
Real GDP is projected to increase by 0.384% and 0.371%, highlighting the crucial role of energy in the economy Despite the rise of renewable energy sources, nonrenewable energy continues to have a more substantial impact on economic growth.
Digging deeper into this area, Tugcu, Ozturk, and Aslan (2012) adopted two different production models on the annual data of Group of Seven (G7) countries
The article discusses two types of functions: the classic function that incorporates capital, labor, and both nonrenewable and renewable energy consumption as inputs, and a modified function that additionally includes research and development along with human capital.
The autoregressive distributed lag approach to cointegration was employed to assess the impact of nonrenewable and renewable energy on the economic performance of G7 countries from 1980 to 2009 Unlike previous studies that utilized the Granger causality test, the authors implemented a newer causality test method developed by Hatemi to explore the relationship between energy consumption and GDP growth This study yielded different results compared to most prior research in the field The long-run estimation indicated that neither nonrenewable nor renewable energy consumption significantly contributed to economic growth during the examined period The researchers conducted analyses on the entire sample as well as individual countries, finding bi-directional causality in all seven countries under the classical production model, while mixed results emerged when a modified production model was applied.
For OECD countries, a recent study was conducted by Shafiei, Salim, and Cabalu
(2014) to check if economies significantly benefits from the use of nonrenewable and renewable energy and to compare the influence of each source on total output
The study examined two key outputs: GDP and industrial output, highlighting the significant role of the industrial sector in economic growth and its substantial share of total energy consumption In addition to employing cointegration and Granger causality tests, the research utilized the recently developed dynamic ordinary least squares technique.
Regression analysis indicates that both energy sources play a significant role in driving GDP growth in OECD countries However, a comparative assessment of their impacts reveals nuanced differences.
Nonrenewable energy sources continue to dominate and significantly influence developed countries A 1% increase in renewable energy consumption correlates with a 0.024% rise in real GDP, while a similar increase in nonrenewable energy consumption results in a 0.245% boost Notably, renewable energy has a negligible impact on industrial output, whereas a 1% increase in nonrenewable energy use elevates output by 0.171% Additionally, the Granger causality test indicates a mutual causality between both renewable and nonrenewable energy consumption and real GDP in both the short and long term.
In summary, research on the energy-growth nexus emphasizes the critical importance of energy, particularly nonrenewable and renewable sources, in contemporary economic activities Despite the increasing adoption and advantages of renewable energy, nonrenewable sources continue to play a vital role, contributing more significantly to national economies Additionally, studies indicate that the fundamental inputs of capital and labor have a substantial positive impact on GDP growth.
The current research on renewable energy consumption remains limited despite its rapid growth Additionally, the methodologies employed are often repetitive across various studies, leading to similar outcomes that fail to capture the true impact of renewable energy and its relationship with nonrenewable energy consumption Consequently, there is a pressing need for more diverse and innovative approaches in this area of study.
Renewable energy has been shown by numerous scholars to positively influence GDP growth; however, its effects on technical efficiency and productivity change remain an unresolved issue This aspect is crucial for policymakers when considering decisions regarding national energy structures.
Further research is essential to address this issue and provide governments with substantial evidence, ultimately aiming to develop the most effective policies.
17 that not only mitigates the aftermaths of energy consumptions on environment but also enhances economic growth.
Productivity change and the Stochastic distance function
The definitions and explanations in this section are taken from OECD Glossary of Statistical Terms (2015)
Productivity is defined as the ratio of output to input, and productivity change (PC) refers to the variations in this ratio PC encompasses the effects of technical efficiency, allocative efficiency, disembodied technical change, and economies of scale This paper will assess PC through the measurement of technical efficiency (TE), efficiency change (EC), and technical change (TC), providing explanations of TE, EC, and TC for a comprehensive understanding.
Efficiency refers to the optimal performance of a production process, encompassing both engineering efficiency (TE) and economic efficiency, known as allocative efficiency Achieving full efficiency in engineering terms is essential for maximizing productivity.
Technical efficiency (TE) signifies that a production process has achieved the maximum output possible with the current technology and available inputs When an economy or firm attains TE, it operates on its production frontier Efficiency change (EC) involves moving towards or away from this best practice, which is the production frontier, and entails the process of addressing technical inefficiencies in production.
Total Change (TC) represents the variation in output volume that a production process can achieve while maintaining the same input levels It indicates the shifts in the production frontier over time, which can occur as either inward or outward movements.
18 happens due to various reasons, such as the change in technology, organization, regulation or the production constraint like input prices
2.4.2 Productivity change measurement and stochastic distance function
PC assessment often relies on Malmquist indices, but this method has notable limitations The stochastic distance function is introduced as an alternative approach to measure PC, effectively addressing the shortcomings associated with the Malmquist index.
The Malmquist output and input PC indices, developed by Caves, Christensen, and Diewert in 1982, are applicable for production scenarios with multiple inputs and outputs, regardless of returns to scale These indices are derived from output and input distance functions, reflecting changes in the maximum output level or minimum input level required However, they cannot be computed when the production form is defined by a nonparametric production frontier As an alternative, the authors propose the Tornqvist output and input indices, which can be constructed as the geometric mean of two Malmquist indices using price and quantity data, eliminating the need for translog parameters that define the production frontier.
Fọre, Norris, and Zhang (1994) expanded upon the previous research by summarizing their findings into two key points Firstly, they elucidate the calculation of Malmquist indices within a nonparametric framework for the production frontier This is achieved through the application of nonparametric linear programming techniques, specifically data envelopment analysis, to model distance functions based on input and output data.
Not using either translog functional structure or data for price, PC is directly computed by taking the geometric mean of two Malmquist productivity indices
The study by Caves, Christensen, and Diewert reveals that the PC index can be broken down into two components: Total Cost (TC) and Economic Cost (EC).
19 has the big limitation is that constant returns to scale is required on the frontier technology
Caves, Christensen, and Swanson (1981) introduced a stochastic approach applicable to flexible production functions through a translog cost function They demonstrated the dual relationship between transformation and cost functions, establishing that productivity change (PC) assessment, based on input-oriented distance, can be expressed as the negative time rate of change in the cost function Consequently, PC can be quantified as the negative time derivative of the analyzed translog cost function.
This method relies on the significant assumption that all firms must operate at their optimal technological efficiency for the economic cost (EC) to be zero Furthermore, the authors did not advance the methodology to allow for a direct estimation of the input distance function.
The authors built upon the foundational work of Fare, Norris, and Zhang (1994) and Caves, Christensen, and Swanson (1981) to develop a stochastic input distance frontier method for computing productivity change (PC), which is defined as the sum of total cost (TC) and efficiency change (EC) A key advantage of this approach is that it does not impose any requirements on returns to scale In contrast to previous nonstochastic methods, which attributed all deviations from reference technology to inefficiency and resulted in significant volatility in PC, EC, and TC, this study employs a parametric model that accounts for noise As a result, the authors observed reduced fluctuations in the temporal patterns of PC and its components.
To prove their arguments on the differences of measuring PC using these two methods, Atkinson, Cornwell, and Honerkamp (2003) utilized the panel dataset of
A study of 43 US electric utilities from 1961 to 1962 analyzed three inputs—fuel, labor, and capital—and two outputs: residential and industrial-commercial electricity The findings from two comparative methods indicated positive and comparable annual rates of productivity change (PC).
There are significant differences regarding the importance of TC and graduation-related topics.
The analysis of productivity growth reveals that the average annual rate of productivity change (PC) is 1.04% according to the Malmquist indices approach, compared to 0.56% from the stochastic distance function approach While the Malmquist indices method shows a balanced contribution of technical change (TC) and efficiency change (EC) to productivity gains, the stochastic distance function approach attributes most gains to TC Additionally, significant discrepancies in the temporal patterns of PC, TC, and EC are observed between the two methods, with the Malmquist indices exhibiting greater volatility This inconsistency is likely due to unresolved noise in the Malmquist indices method.
The use of stochastic distance functions to measure productivity change (PC) has gained recognition among many researchers This method was implemented by Atkinson and Dorfman (2005), incorporating sulfur dioxide (SO₂) emissions as a negative output.
The main outcome inferred from this research is the negative EC over the examined period which is greatly ascribed to the exertions of firms in cutting SO 2 emissions
ECONOMETRIC MODEL
Stochastic Distance Function form
This section follows Atkison, Cornwell, and Honerkamp (2003) Considering a country’s production technology where N nonnegative good inputs are combined, to create M nonnegative good outputs,
The country’s production technology can be written in term of an input correspondence as following: in which is the set of input requirement
For a given country, the input distance function is translated as the maximum scale factor essential for x t to be on the frontier of
(1) happens if and only if ≥ 1 This is because of the assumption of inputs’ free disposability More clearly, it is
Above function is served to compute the PC using Malmquist indices
Atkison, Cornwell, and Honerkamp (2003) expressed (1) for country c over period t under a more typical of an econometric model:
The model discussed by Fare and Primont (1996) highlights the duality between the input distance function and the cost function, emphasizing their interrelated nature in economic analysis.
The authors present a vector of input prices and assume that the cost function is multiplicative This allows the cost function to be expressed in a specific form, where one component is treated as a random variable and the other as a function of the input prices.
Similarly, the distance function can be defined as the same context:
The stochastic input distance function can be expressed by adjusting divergences from 1 due to technical inefficiency, as inferred from the equality that occurs only when \( x_{ct} \) belongs to the isoquant of the input requirement set.
(2) Therefore, if (2) is given as a functional form, it can be analyzed through econometrical methods after inputs are set under linear homogeneity.
Parametric specification
The translog functional form is utilized to provide a flexible approximation of the distance function Consequently, the empirical model for country \(c\) during period \(t\) is expressed as follows.
23 is the time dummy for specific year
The composite error is the additive error, combining two elements One is u ct ≥ 0, called one – sided element, and the other is v ct ,called standard noise with zero mean
Many parametric restrictions are imposed on (3) First of all, the symmetric requirements include following conditions:
Besides, the linear homogeneity property of input quantities suggests that:
Incorporating country-specific dummy variables, as suggested by Le and Atkinson (2010), into the model allows for a relaxation of the strong distribution assumption regarding the error terms \$u_{ct}\$ and \$v_{ct}\$ With a sample comprising 34 distinct OECD countries, each characterized by unique geographical, demographic, and regulatory attributes, the use of dummy variables effectively accounts for these national differences.
Computing partial effects among variables
Following Le and Atkinson (2010), Agee, Atkinson, and Crocker (2012), the implicit function theorem allows us to analyze the partial impact of one variable on another variable
Firstly, we take the partial derivative of function (3) with respect to each variable, including both output and input variables, i.e., and , respectively
Then, the impact of an input on an output is – with
The impact of an input on another input is – with
Computing technical efficiency, efficiency change, technical change and
This section follows Atkison, Cornwell, and Honerkamp (2003) EC, TC and PC are measured in terms of percentage changes
The measurement of TE, EC, TC, and PC is based on the results from the estimation of equation (3) Since non-negativity is not initially imposed on the one-sided element \( u_{ft} \) during the estimation, it is addressed later by adjusting the fitted model to determine the frontier intercept This adjustment involves adding and subtracting \( t \) from equation (3), allowing it to be rewritten accordingly.
(4) in which is the estimation of the frontier distance function at period t and
The technical efficiency of country c during period t, denoted as TE ct, is derived from equation (4).
With TE ct from (5), the change in TE, EC ct , is computed as following:
(6) This is the catching – up rate to the frontier from period t – 1 to period t
Technical change, TC ct , is the difference between the examined frontier distance function in two periods: t and t – 1, outputs and inputs holding constant TC is measured through below function:
Therefore, if the frontier intercept, i.e., t , changes, it will affects both TC and EC
Finally, with given TC ct and EC ct ; productivity change of country c at time t, PC ct , is constructed as:
PC ct = EC ct + TC ct
Model specification
This study expands the traditional production function by incorporating nonrenewable and renewable energy consumption alongside capital and labor as inputs, focusing solely on GDP as the output In this context, capital, labor, nonrenewable energy, and renewable energy are represented by the variables K, L, NE, and RE, respectively.
The dataset spans 23 years, from 1990 to 2012, resulting in 23 time-specific dummy variables labeled as year1 through year23 To prevent the dummy variable trap, only 22 of these variables are included in the model, with year1 excluded This approach allows for the calculation of average total cost (TC) for countries from year2 (1991) to year23 (2012).
As explained in Section 3.2, country – specific dummy variables are added to (3)
The OECD sample comprises 34 countries, resulting in 34 dummy variables labeled as d1, d2, …, d34 However, similar to the approach taken with time dummy variables, only 33 of these country variables are incorporated into the model.
The empirical model for country c over period t from (3) is specified as following:
The equation represents a comprehensive model where the dependent variable is influenced by various factors, including the natural logarithm of GDP, capital, labor, energy, and renewable energy Each factor is weighted by its respective coefficient, indicating its impact on the outcome Additionally, the model incorporates quadratic terms for each variable, capturing non-linear relationships, as well as interaction terms that reflect the combined effects of different inputs on the dependent variable This multifaceted approach allows for a nuanced understanding of the dynamics at play in the economic system.
The equation models the relationship between GDP and various factors, including capital (K), labor (L), natural resources (NE), and renewable energy (RE), with coefficients representing their respective impacts It incorporates time variables, indicating how these relationships evolve over the years, specifically from year 2 to year 23 The analysis highlights the significance of each factor in contributing to GDP growth, emphasizing the dynamic nature of economic influences over time.
The equation incorporates a series of terms involving logarithmic functions of various variables, such as \( \ln L \), \( \ln NE \), and \( \ln RE \), each multiplied by coefficients \( \alpha \) and indexed by years ranging from 2 to 23 Additionally, it includes quadratic terms for each year and specific variables \( d1 \) to \( d33 \), all contributing to the overall model.
The restrictions imposed on this model to meet linear homogeneity properties in section 3.2 are defined below:
10 α yk + α yl + α yne + α yre = 0 tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Furthermore, the symmetric restrictions are imposed on (8), so the following requirements must be satisfied: α kl = α lk ; α kne = α nek ; α kre = α rek ; α lne = α nel ; α lre = α rel ; α nere = α nere
In accordance with the methodologies established by Le and Atkinson (2010) and Agee, Atkinson, and Crocker (2012), we conducted estimations based on the distance model (8) and performed several tests We tested the null hypothesis asserting that all squared terms of inputs and their interaction terms are collectively equal to zero Additionally, we examined the null hypothesis that the interaction terms between outputs and inputs are also collectively equal to zero.
Before analyzing the partial effects of K, L, NE, RE, and GDP, we calculate the derivatives of the stochastic distance function concerning the natural logarithm of each variable for every country on an annual basis.
According to Le and Atkinson (2010) and Agee, Atkinson, and Crocker (2012), the derivatives discussed are averages weighted by the total of good outputs, with GDP being the sole output in this study The partial effect of inputs on GDP can be analyzed, while the interaction between inputs is also considered Since all variables are expressed in natural logarithm form, these partial effects can be interpreted similarly to elasticity Additionally, the average Total Efficiency (TE), Efficiency Change (EC), Technical Change (TC), and Pure Technical Change (PC) of OECD countries are calculated as averages weighted by GDP each year.
CHAPTER 4: ENERGY CONSUMPTION AND SUPPLY IN
This chapter is divided into three sections The first section provides an overview of the OECD and its global significance across various dimensions The subsequent sections focus on energy consumption and supply within OECD countries, with a particular emphasis on renewable energy sources.
OECD versus non – OECD
The Organization for Economic Cooperation and Development (OECD) was founded in 1960 when the United States and Canada joined 18 European nations to promote economic development It officially came into existence in 1961 with the implementation of the OECD Convention.
The OECD, comprising 34 countries across America, Europe, and the Asia-Pacific, includes many of the world's most developed nations alongside emerging economies such as Chile, Mexico, and Turkey This diverse membership grants the OECD significant influence in global economics and politics, as well as a substantial role in the global energy supply and demand.
In 2013, OECD countries represented 18% of the global population, 47% of the world's GDP, and 40% of total primary energy supply (TPES), as illustrated in Figure 4.1 Remarkably, despite comprising only one-fifth of the global population, OECD nations generated approximately half of the world's economic output.
Figure 4.1: OECD versus Non – OECD in terms of population, GDP, total primary energy supply and production
Source: International Energy Agency (IEA) (2015)
OECD also ranks number one in terms of energy – intensive region in the world
In 2013, the OECD's total primary energy supply (TPES) per capita was 4.2 tonnes of oil equivalent (TOE), significantly higher than the global average of 1.9 TOE This disparity can be attributed to several factors, including the OECD's large industrial and service sectors that require substantial energy consumption, nearly complete electrification within the group, and a high rate of vehicles per household (IEA, 2015).
The energy intensity of OECD countries is generally lower than the global average, despite their high per capita energy consumption This indicates a more advanced technological development, enabling these nations to produce additional output with less energy consumption (IEA, 2015).
Energy consumption in OECD countries
Generally, total final consumption of energy, which is the sum of all energy provided to final consumers in all sectors, in the OECD has increased by time
Countries in America lead global energy consumption, followed by Europe, while Asia and Oceania rank last These trends are illustrated in Figure 4.2 This disparity in energy usage highlights significant regional differences in energy demand.
30 general decoupling of economic performance by region from the observation of energy consumption over time
Figure 4.2: Total final energy consumption by region in OECD (1971 – 2013)
According to the IEA (2015), the final energy intensity, which is calculated by dividing total final energy consumption by GDP, has been steadily decreasing over time, as illustrated in Figure 4.3 Notably, this ratio in 2013 was less than half of what it was in 1997, indicating a significant improvement in energy efficiency within the OECD over the years.
Figure 4.3: Final energy intensity in OECD (1971 – 2013)
Source: IEA (2015) tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
On the other hand, the ratios of final energy intensity are very different among countries, based on their economic structures and the effectiveness of energy consumption in each country
Figure 4.4 illustrates the energy intensities across various economic sectors, highlighting that the industry and transportation sectors typically represent the largest share of total energy consumption, with their rankings fluctuating annually For instance, the IEA (2015) reports that in 2013, transportation accounted for approximately one-third of total final energy consumption, whereas in 1971, industrial activities dominated with 41%, while transportation only represented 24%.
The residential sector is the third largest consumer of energy Over time, energy intensities in this sector, which measure final energy consumption relative to GDP adjusted for purchasing power parity (USD PPP), have shown a decreasing trend This decline highlights the ongoing improvements in energy consumption efficiency across various sectors.
Deeper studying in national level should be conducted to explain the reasons
Figure 4.4: Sectorial energy intensities in OECD (1971 – 2013)
Source: IEA (2015) tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
4.2.2 Renewable energy consumption versus nonrenewable and total energy consumption
In overall, the consumption of total energy and each source of energy tended to increase during the investigated period in this paper (1990 – 2012) Figure 4.5 delivers a clear picture of this
Figure 4.5: Energy consumption in OECD (1990 – 2012)
Nonrenewable and total energy consumption went up relatively fast from 1990 to
Between 2008 and 2010, energy consumption declined due to the global economic recession Following the recovery, the overall energy usage, particularly of nonrenewable sources, has experienced a slowdown However, the share of renewable energy, although still small, has consistently increased each year, rising from approximately 7.82%.
From 1990 to 2012, energy consumption saw a significant decline, dropping to 10.93% during the economic downturn Notably, while the overall use of energy and nonrenewable sources fell sharply, renewable energy consumption continued to rise steadily This trend can be attributed in part to the influence of the Kyoto Protocol, which came into effect in 2005.
In term of renewable energy consumption by sector, Figure 4.6 demonstrates the shares of renewable energy used in sectors of OECD countries in 1990 and in 2013
Nonrenewable Renewable Total energy tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
33 to show the changes in sectorial renewable energy consumption over time
Electricity plants represent the largest share of total renewable energy consumption, although this share has decreased from 51.6% in 1990 to 48.8% in 2013 This decline is accompanied by a reduction in the Residential, Commercial, and Public sector, as well as in Combined Heat and Power systems.
Figure 4.6: Sectorial renewable energy consumption in OECD in 1990 and
The adoption of renewable energy in final consumption sectors, particularly in transportation, has seen significant growth, increasing from 0.05% in 1990 to 9.7% in 2013 This surge is largely driven by the use of biofuels for heat generation, which are directly consumed on-site in these sectors (IEA, 2015).
Energy supply in OECD countries
Figure 4.7 displays the TPES, energy production and net imports of energy through the period 1971 – 2014 TPES of the OECD has unsteadily grown over time From
From 1984 to 2007, there was a period of relatively stable annual growth, characterized by an average increase.
34 rate of 1.4% per year During the economic downfall in 2008 – 2009, TPES sharply dropped After this period, TPES tended to go around the same levels of the years
2000 It reached the level of 5238 Mtoe in 2014, which is 4% lower compared to the level in 2004, but 16% and 55% higher compared to the level in 1990, 1971, respectively
Figure 4.7: Total primary energy supply in OECD (1971 – 2014)
Unlike TPES, energy production in the OECD consistently increased over the period, while net imports remained relatively low and experienced significant fluctuations The rise in energy production outpaced energy consumption, resulting in a high level of self-sufficiency for the OECD, nearly reaching complete self-sufficiency when measured by the ratio of energy production to TPES.
2014 when this ratio is 99% (IEA, 2015)
4.3.2 Renewable energy supply in OECD countries
In 2014, Figure 4.8 illustrates the composition of Total Primary Energy Supply (TPES) in OECD countries, highlighting a significant increase in the share of renewable energy compared to nonrenewable energy sources This trend indicates a much higher growth rate for renewable energy supply within OECD nations from 1990 onwards.
In 2014, the former statistic was 3.3%, while the latter was only 0.5% (IEA, 2015).
35 renewable energy supply made a contribution of 9.2% in TPES Nevertheless, nonrenewable energy sources are still dominant with the largest portion belonging to oil
Figure 4.8: Composition of total primary energy supply in OECD (2014)
In 2014, the composition of renewable primary energy supply in OECD countries was dominated by bio-fuels and waste, which accounted for 55.2% of the total Hydroelectric power represented the second largest portion of renewable energy Within the bio-fuels and waste category, solid bio-fuels, including wood and charcoal, comprised the largest share at 37.9%.
Figure 4.9: Composition of total renewable primary energy supply (2014)
Source: IEA (2015) tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Compared with the world, while OECD countries made contribution of only 26.1% in global total renewable energy supply, they captured 39.1% of the global TPES in
As of 2013, renewable energy constituted only 9% of the total energy supply in OECD countries, significantly lower than the 49.6% in Africa, 29.2% in Non-OECD countries in America, 25.7% in Non-OECD countries in Asia, and 10.7% in China Figure 4.10 illustrates the distribution of renewable energy within the Total Primary Energy Supply (TPES) across OECD countries and other regions.
Figure 4.10: Renewable energy shares in TPES of OECD versus other regions
OECD countries are leaders in the development of new renewable energy sources, contributing 66.1% of the global renewable energy generation from solar, wind, tidal, municipal waste, biogases, and liquid biofuels in 2013, according to the IEA (2015).
EMPIRICAL RESULTS
Data
The study analyzes annual data on gross domestic product, capital, labor, and both nonrenewable and renewable energy consumption across 34 OECD countries from 1990 to 2012 This unbalanced panel dataset comprises a total of 760 observations.
Table 5.1 outlines the definitions and sources of the variables examined in this thesis, with the third column indicating the anticipated signs of the partial effects of capital, labor, nonrenewable energy, and renewable energy consumption on GDP, based on empirical studies discussed in Chapter 2.
Table 5.1: Variables definition Variable name Definition Expected sign Source
Data of GDP in dollar (constant
The variable is represented by the sum of gross value added by all resident producers in the economy, along with product taxes, minus any subsidies not included in product value GDP figures are expressed in US dollars and converted from domestic currencies using official exchange rates from 2005.
World Bank’s World Development Indicators
(2015) tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Variable name Definition Expected sign Source
Gross fixed capital formation, measured in constant 2005 US dollars, encompasses various investments such as land improvements, purchases of plant, machinery, and equipment, as well as the construction of infrastructure like roads and railways This category also includes the development of essential facilities, including schools, offices, hospitals, and both residential and commercial buildings.
The total labor force includes individuals aged 15 and older who are classified as economically active by the International Labor Organization This group encompasses all people who provide labor for the production of goods and services within a specific timeframe, incorporating both those who are employed and those who are unemployed.
Nonrenewable energy consumption is measured as the aggregate of the consumption of coal and coal products, oil, and natural gas in quadrillion Btu units
(2015) tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Variable name Definition Expected sign Source
Renewable energy consumption is measured as the aggregate of the consumption of wood, waste, geothermal, wind, photovoltaic, and solar thermal in quadrillion Btu units
Following the explanation in Chapter 3, all variables are converted to natural logarithm before conducting the econometrical estimation.
Descriptive analysis
The summary of descriptive statistics of all variables is reported in Table 5.2 below:
Table 5.2: Descriptive statistics of variables Variable Obs Median Mean Std Dev Min Max
The GDP and capital levels among OECD countries show minimal disparities between the median, mean, maximum, and minimum values, indicating a relative equality in growth However, countries with smaller populations, like Iceland and Luxembourg, exhibit a significant gap in labor values Despite this, the mean and median labor values remain closely aligned, suggesting that the labor forces across OECD nations are relatively uniform.
The disparities in median, mean, maximum, and minimum values of nonrenewable and renewable energy consumption are substantial, primarily due to variations in country size, population, and economic structures For instance, Iceland has a relatively low overall energy consumption due to its small population and size, with its economy largely driven by the fishery industry, which requires minimal energy (Sigfusson and Gestsson, 2012) In contrast, larger nations like the US, Japan, and Germany, characterized by significant population and size, have energy-intensive industries that create a high demand for energy.
The significant disparity in statistical values between nonrenewable and renewable energy consumption highlights the continued dominance of nonrenewable energy sources in OECD countries.
Figure 5.1 illustrates the correlations between nonrenewable and renewable energy consumption and GDP, showing a positive relationship between GDP and the consumption of both energy sources.
Figure 5.1: Correlation between GDP and nonrenewable, renewable energy consumption
Source: Author’s calculation Also from Figure 5.1, while GDP seem to spread around fitted line of nonrenewable energy consumption, it scatters farther from the line of renewable energy
-4 -2 0 2 4 lnne lngdp Fitted values tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
41 consumptions This suggests a relatively higher correlation of nonrenewable energy consumption than renewable energy consumption to GDP
The table below presents the correlation matrix among the variables, where Ln(GDP), Ln(K), Ln(L), Ln(NE), and Ln(RE) represent the natural logarithms of GDP, capital, labor, nonrenewable energy consumption, and renewable energy consumption, respectively.
Table 5.3: Correlation matrix between variables
Ln(GDP) Ln(K) Ln(L) Ln(NE) Ln(RE)
The analysis reveals a strong positive relationship between GDP and both nonrenewable and renewable energy consumption, with capital and labor also showing significant correlations with GDP Notably, capital exhibits the strongest connection to GDP, boasting a correlation value of 0.993 In contrast, renewable energy consumption demonstrates the weakest link to GDP, with a correlation value of 0.663, reflecting its recent development and limited contribution to economic growth.
To be concluded, the descriptive statistics displays that both nonrenewable and renewable energy consumption appear to be positively correlated with GDP
Nonrenewable energy consumption shows a stronger correlation with GDP compared to renewable energy consumption A more detailed econometric analysis of these relationships will be presented in the following section.
Regression results
5.3 Regression and calculation results 5.3.1 Partial effects among variables
The F-test was performed on the regression of the stochastic distance function to determine if the squared terms of inputs and their interaction terms are jointly equal to zero Additionally, the interaction terms between inputs and output were tested for joint significance The results showed p-values below 0.01 in both cases, allowing for the rejection of the null hypotheses at a 1% significance level This indicates that at least one squared term of inputs and one interaction term among inputs, as well as one interaction term between inputs and output, are statistically significant Detailed F-test results can be found in Appendix 3.
The estimations results of four inputs and one ouput are presented in following table:
(Ln(RE)) 2 0.001 (0.000)*** tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Interaction terms among output and inputs
Notes: *** denotes significance at the 1% level Source: Author’s calculation
The partial derivative of the stochastic distance function with respect to GDP, as shown in Table 5.4, is –0.009, indicating a decrease in the distance function as GDP increases This aligns with the principle that higher GDP brings a country closer to its production frontier, reducing the gap between actual production levels and potential output In contrast, the partial derivatives of the distance function concerning capital, labor, and both nonrenewable and renewable energy consumption are positive, indicating that the distance function increases with these inputs This finding supports the conditions outlined for the input distance function (Atkinson, Cornwell, and Honerkamp, 2003).
Table 5.5 presents the partial impacts among variables, which are weighted averages based on GDP All variables are expressed in natural logarithm form, and the derivatives are calculated with respect to the natural logarithm of each variable.
The relationship between variables can be understood through elasticity According to Table 5.5, while holding other factors constant, increases in both nonrenewable and renewable energy consumption positively impact GDP, albeit to different extents Specifically, a 1% rise in nonrenewable energy usage results in a 0.51% increase in GDP, whereas a similar increase in renewable energy consumption only contributes an additional 0.03% to GDP These findings align with empirical studies presented in Chapter 2, which highlight that, despite the recent growth in renewable energy usage, nonrenewable energy remains the dominant driver of economic growth.
Table 5.5: Partial effects among variables
Partial effects of inputs on output
Capital has the most significant impact on GDP among all input factors, with a 1% increase in capital leading to an average 1% rise in output This demonstrates the efficient use of capital in OECD countries, which are primarily advanced economies In contrast, an increase in the labor force by 1% results in a slight decrease in GDP by 0.093%, a surprising finding that may warrant further investigation.
The decline in the significance of labor in OECD countries can be attributed to two main factors Firstly, these advanced nations heavily rely on technology in production, reducing the importance of manual labor Secondly, the growth of labor unions has been notable, as they strive to protect employee benefits amidst an expanding labor force This rise in union activity has led to an increase in labor strikes, which not only disrupt production but also impact social security Notably, dispute rates in the industrial sector, including manufacturing and construction, are typically twice as high as those in the service sector, with the exception of transportation Despite a growing employment rate in the service sector, the industrial sector continues to represent a significant portion of the labor force.
The recent increase in strike rates within the transportation sector, particularly in airlines and public transportation, has been noted However, the thesis indicates that the negative impact of labor on GDP is minimal, showing an insignificant partial effect.
The negative partial effect between renewable and nonrenewable energy consumption suggests that these two types of energy serve as substitutes Specifically, a 1% increase in renewable energy usage results in a 0.098% decrease in nonrenewable energy consumption.
Although the ratio of effect is small, the substitution between two energy sources contributes to ease the harsh impacts of nonrenewable energy consumption on the environment
Capital and labor significantly influence the consumption of renewable and nonrenewable energy sources A 1% increase in capital leads to a mere 0.202% rise in renewable energy consumption, while nonrenewable energy use surges by 3.389% Similarly, a 1% growth in the labor force results in a 3.52% increase in nonrenewable energy consumption, highlighting the stronger impact of capital and labor on nonrenewable energy compared to renewable sources.
0.152%, respectively These results again stress the dominance of nonrenewable energy in the economies of OECD countries
5.3.2 Technical efficiency, efficiency change, technical change and productivity change
Table 5.6 displays estimated technical efficiencies for OECD countries and their standard deviations They are computed from equation (5) in Chapter 3 and averages weighted for GDP
Table 5.6: Average technical efficiencies of OECD countries (1990 – 2012)
Weighted average 0.9973 Source: Author’s calculation tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
In 1990, the weighted average value of Total Efficiency (TE) among 34 OECD countries was 0.9965, suggesting that if an average country had utilized its inputs—capital, labor, and both renewable and nonrenewable energy— as effectively as the top-performing country, its GDP could have increased by approximately 0.355% This minimal increase indicates that the average OECD country operated at an efficiency level very close to that of the best performer in that year.
Figure 5.2 illustrates the fluctuations in average Total Efficiency (TE) from 1990 to 2004, peaking in the final two years before experiencing a decline starting in 2005 The global economic downturn between 2008 and 2012 likely contributed to this decrease, characterized by reduced capital investment and increased unemployment rates compared to earlier years.
Figure 5.2: Average technical efficiency of OECD countries (1990 – 2012)
The weighted-average annual rate of efficiency change (EC) for the entire period is notably high at 0.9973, indicating that if OECD's average country utilized all examined inputs as effectively as the best-performing country in the organization, its annual GDP could increase by approximately 0.266% from 1990 to 2012 Additionally, the small standard deviation values of technical efficiency (TE) measurements imply that the dispersion of TE is minimal throughout this period.
1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Table 5.7 presents the average Efficiency Change (EC), Technical Change (TC), and Productivity Change (PC) calculated using equations (6), (7), and (8) from Chapter 3 EC, defined as the difference in Technical Efficiency (TE) between consecutive years, reflects the progress of countries towards the production frontier The average EC follows a similar trend to the average TE, which increased among OECD countries from 1990 to 2004 but declined thereafter, resulting in positive EC values during the first period and negative values in the latter This consistent decrease in EC during the second period contributes to an overall negative annual TC of -0.0046% for the entire timeframe.
Table 5.7: Average efficiency change, technical change, productivity change of OECD countries (1991 – 2012)
Weighted average -0.000046 0.012107 0.012061 Source: Author’s calculation tot nghiep down load thyj uyi pl aluan van full moi nhat z z vbhtj mk gmail.com Luan van retey thac si cdeg jg hg
Total Change (TC) represents the difference in frontier distance functions between two consecutive years, t and t+1, while keeping outputs and inputs constant, highlighting the shift in the production frontier Unlike Efficiency Change (EC), TC rates are predominantly positive, except for the period from 1991 to 1995, leading to an average annual growth rate of 1.2107% These findings suggest that the average production frontier has consistently shifted outward over the analyzed period.
Productivity change (PC) is the sum of efficiency change (EC) and technical change (TC) Analysis from Table 5.7 indicates that, with the exception of the period from 1991 to 1995, the average country has experienced positive productivity changes over time The weighted average rate of PC at 1.2061% suggests that the productivity gains of OECD countries from 1990 to 2012 are entirely attributed to TC This indicates that the improvement in productivity is solely due to the outward shift of the production frontier The negative values of EC imply that OECD countries are operating very close to their best-practice levels, resulting in less incentive to invest in enhancing technical efficiency This finding aligns with the results of Atkinson, Cornwell, and Honerkamp (2003), although the contexts differ, as discussed in Chapter 2 regarding US electricity firms.