CC BY
4
Вестник Российского университета дружбы народов. Серия: Экономика
ISSN 2313-2329 (Print); ISSN 2408-8986 (Online) 2024 Vol. 32 No. 2 287-302
http://journals.rudn.ru/economics
DOI: 10.22363/2313-2329-2024-32-2-287-302 EDN: IXXILJ UDC 339.9
Research article / Научная статья
Evidence from the South African Energy Sector on the Impact of Gas Consumption and Technologies on the Environment
Collin L. Yobe <E El, Binganidzo Muchara Э
University of South Africa, Graduate School of Business Leadership, Cnr Janadel and, Alexandra Ave, Midrand, 1686, South Africa
Abstract. Legacy emissions from fossil fuel consumption signify the lasting impact of past carbon dioxide (CO2) emissions on present-day emissions. Given that the current emission levels are also high; it has become urgent to deal with this crisis. This study aims to examine the effects of gas consumption, legacy CO2 emissions, energy decoupling, and population on carbon dioxide emissions in South Africa using the modified IPAT identity and the Markov Switching Dynamic Regression analysis. Integrating additional variables into the modified IPAT identity uncovered evidence from the South African energy sector on the impact of gas consumption on the environment. The Markov Switching Dynamic Regression Model (MSDRM) utilised annual data from the South African energy sector from 1966 to 2020, collected from diverse sources. Results indicate that the Gas model's probability (i.e., 0.8475) would persist in high-emissions states over time. The MSDRM results showed that gas consumption suggests a statistically significant negative relationship between gas consumption (-0.0461) and CO2 emissions, meaning that despite the decrease in CO2 emissions from using gas, it does not imply instant reversals in the ambient CO2 as to reduce the overall CO2, likely contributed from other CO2-emitting fuels. The MSDRM results showed that legacy CO2 emissions positively impact (I) current CO2 emissions and that decoupling (T) leads to increased CO2 emissions—the latter relationship indicating likely energy rebounding. These findings highlight the need to prioritise interventions and strategies targeting the factors with higher probabilities of contributing to sustained high emissions, which may involve implementing policies to transition away from high-emission sources while exploring alternatives and adopting cleaner energy sources. The results emphasise the challenge of decoupling economic growth from high CO2 emissions and underscore the importance of sustained efforts to address and mitigate climate change.
Keywords: Gas consumption, energy decoupling, South African energy sector, modified IPAT identity
© Yobe C.L., Muchara B., 2024
Tfva"! This work is licensed under a Creative Commons Attribution 4.0 International License https://creativecommons.org/licenses/by-nc/4.0/legalcode
Article history: received January 18, 2024; revised February 29, 2024; accepted March 12, 2024.
For citation: Yobe, C.L., & Muchara, B. (2024). Evidence from the South African energy sector on the impact of gas consumption and technologies on the environment. RUDN Journal of Economics, 32(2), 287-302. https://doi.org/10.22363/2313-2329-2024-32-2-287-302
Влияние потребления газа и технологий на окружающую среду: опыт энергетического сектора Южной Африки
К.Л. Йобе <Е И, Б. Мучара ©
Университет Южной Африки, Высшая школа бизнес-лидерства, ЮАР, 1686, Мидранд, авеню Джанадель и Александра
Аннотация. Остаточные выбросы от потребления ископаемого топлива свидетельствуют о долгосрочном воздействии прошлых выбросов углекислого газа (CO2) на современные объемы. Учитывая, что текущие уровни выбросов остаются на высоком уровне, необходимо обращать внимание на накопительный характер данного процесса. Представленное исследование направлено на изучение влияния потребления газа и остаточных выбросов CO2 на выбросы углекислого газа в Южной Африке с использованием модифицированной идентификации IPAT, а также скрытой модели Маркова. Внедрение дополнительных переменных в модифицированную идентификацию IPAT выявило доказательства влияния потребления газа на окружающую среду в энергетическом секторе Южной Африки. в модели динамической регрессии маркова (MSDRM) использованы ежегодные данные из энергетического сектора Южной Африки с 1966 по 2020 г., собранные из различных источников. Результаты показывают, что вероятность модели (0,8475) будет сохраняться в состояниях с высокими выбросами с течением времени. Результаты MSDRM показали, что потребление газа указывает на статистически значимую отрицательную связь между потреблением газа (-0,0461) и выбросами CO2, означающую, что, несмотря на снижение выбросов CO2 при использовании газа, это не влечет за собой мгновенных обратных изменений в атмосфере. Вероятно, такие изменения вызваны другими источниками выбросов. Результаты MSDRM показали, что остаточные выбросы CO2 положительно влияют (I) на нынешние выбросы CO2 и что декаплинг (T) приводит к увеличению выбросов CO2. Полученные результаты подчеркивают необходимость разработки приоритетных «зеленых» стратегий, направленных на борьбу с источниками с устойчивыми высокими выбросами.
Ключевые слова: потребление газа, энергетический сектор Южной Африки, IPAT
История статьи: поступила в редакцию 18 января 2024 г., проверена 29 февраля 2024 г., принята к публикации 12 марта 2024 г.
Для цитирования: Yobe C.L., Muchara B. Evidence from the South African energy sector on the impact of gas consumption and technologies on the environment // Вестник Российского университета дружбы народов. Серия: Экономика. 2024. Т. 32. № 2. С. 287-302. https://doi.org/10.22363/2313-2329-2024-32-2-287-302
Introduction
The creation of strategies to reduce disruption and guarantee a smooth integration process is necessary for the effective integration of renewable energy sources (Abas et al., 2015; Benli et al., 2020; Du et al., 2019; Neves, Marques, 2021; Stancin et al., 2020; Zou et al., 2016).
Energy rebound effects occur when an initial boost in energy efficiency is followed by an increase in energy use that returns to pre-involvement levels. These effects are particularly prevalent in places with high levels of energy use (Li R., Li S., 2021). One way to save energy is to replace incandescent lighting with energy-efficient LED fixtures. On the other hand, using more LED bulbs may lead to a greater total energy consumption.
Modelling environmental impact (I), i.e., carbon emissions due to population (P), affluence (A), and technology (T), using the IPAT identity (I = PAT) can help quantify GHG emissions and the factors driving them (Chertow, 2000; Pham et al., 2020; Skanberg and Svenfelt, 2022). This model, however, is limited in that it does not account for every element that could impact the environment. More advanced models that get around these restrictions and provide a deeper understanding of the variables influencing the environment have been created and made available.
Technological innovation significantly boosts economic growth, competitiveness, and value creation while addressing long-standing challenges like the health impacts of climate change, unsustainable economic expansion, and an ageing society. The research offers a theoretical understanding of the correlations between independent variables—population growth, consumption trends, and time-series trends of specific indicators—and CO2 emissions. The following have made a significant contribution to research. First, the methodological approach is the basis of the contribution. The study offers a novel effect assessment formula by providing a modified version of the IPAT identity. This identity allows additional and natural logarithm-converted variables to be incorporated into the model (Gujarati, 2004). Regression analyses that include a lagged dependent variable can capture persistence and dynamics in generating data (Wooldridge, 2010). It enables the relationship between the dependent variable's previous values and its current value to be captured (Gujarati, 2004), which can provide more precise results than considering only contemporaneous variables (Wooldridge, 2010), commonly employed in economics and other disciplines, and it can improve the goodness of fit of the model.
Furthermore, the study uses a technology variable that Data Envelopment Analysis acquired to estimate energy decoupling (DEA). The predicted DEA technology score considers South Africa's CO2 emissions and GDP per capita. Finding and measuring the effects of this variable is crucial as technology capabilities have made it possible to address several persistent problems. The third contribution of the research deals with endogeneity issues by employing a residual inclusion approach on Kaya identity variables utilising the MSDRM. The Kaya identity formula captures four variables— the human population, GDP per capita, energy intensity (per unit of GDP), and
carbon intensity. These determine the overall emissions of greenhouse gases (per unit of energy consumed).
Finally, utilising the modified version of the IPAT identity, the study reveals how structural breaks in CO2 emission data can be quantified using Markov-Switching Dynamic Regression Models (MSDRM). Since the current state of the data-generating process (DGP) is unknown, these models calculate the likelihood of various unobserved states of the DGP. Therefore, this study aims to examine the effects of gas consumption, legacy CO2 emissions, energy decoupling, and population on carbon dioxide emissions in South Africa using the modified IPAT identity and the Markov Switching Dynamic Regression analysis.
Exploring the Impact of Energy Decoupling in South Africa
Decoupling has become more significant as a strategy for resolving the problem of excessive carbon dioxide (CO2) emissions since it isolates economic expansion from resource consumption and environmental degradation. Numerous studies, particularly in South Africa, have looked at the relationship between population growth, decoupling, and the use of fossil fuels in relation to CO2 emissions.
Hubacek et al. (2021) demonstrated that, as the majority of wealthy countries have shown, it is possible to fully decouple GDP from production-based emissions. A promising development that some countries have managed to achieve is complete decoupling. CO2 emissions are released into the atmosphere even by nations that are completely disconnected from the environment. Decoupling might not be enough to solve the problems that greenhouse gas emissions are causing for the environment. They also point out that decoupling might only be temporary and that the decoupled countries might eventually resume increasing their emissions; as a result, consistent and continuous efforts are needed to keep decoupling trends going.
Data and methods Conceptual framework
Because of the disproportionate effects of coal, oil, and gas on the environment, it is important to look at the use of each fossil fuel. For example, coal usually creates more CO2 and local air pollution per unit of energy than other fossil fuel types. Simultaneously, gas has been seen as a good substitute that tackles issues related to environmental effects.
The formula for measuring human impact, the IPAT identity, is shown below.
I = P x A x T, (1)
where: I = impact; P = population; A = affluence; T = technology.
Econometric model
The econometric models specified CO2 emissions using the extended IPAT model using the MSDRM. The researchers determined the MSDRM for gas consumption. The comprehensive IPAT model allows for additional variables to be included.
A one-input and one-output DEA model — i.e., GDP per capita and CO2 Emissions, respectively — estimated the technical efficiency for energy decoupling. The estimated technical efficiency denotes technologies used to decouple GDP per capita and CO2 Emissions. Additionally, the specified model output-orientated DEA and assumed Constant Returns to Scale.
Data sources for the variables
The annual data for South Africa were acquired from different sources, covering the different annual periods. Table 1 shows the attributes of the other variables utilised in the research.
Table 1
Data sources for the variables
Variables Unit Variable period Source
Carbon dioxide Emissions Per Annual production-based (Andrew, Peters, 2021;
emissions of CO2, measured in tonnes. 1750-2020 Project, 2021)
Capita
Gas consumption Fossil fuel consumption is given in terawatt-hour equivalents (TWh). 1960-2020 www.BP.com
Population Population 1950-2021 https://population.un.org/wpp/
GDP at purchaser's prices is the
sum of gross value added by all
resident producers in the country
plus any product taxes and https://datacatalog.worldbank.
GDP per capita minus any subsidies not included in the value of the products. It is calculated without deductions for the depreciation of fabricated assets or the depletion and degradation of natural resources. 1960-2020 org/search/dataset/0037712/ World-Development-Indicators
A Data envelopment analysis
Technology fractional score is estimated to capture the technologies in use for efficiency for energy decoupling. See GDP per capita and CO2 Emissions.
Source: Andrew, Peters, 2021; Project, 2021. Retrieved 24 December, 2023, from www.bp.com; https:// datacatalog.worldbank.org/search/dataset/0037712/World-Development-Indicators
Econometric estimation in the extended IPAT model
The below specified empirical models calculated the extended CO2 emissions utilising the modified IPAT identity for gas consumption. This method permits the model to include other variables dealing with econometric problems, i.e., residuals and lagged dependent variables. The equations below illustrate the extended IPAT identity relationship, and the definitions of the variables are in Appendix.
Gas consumption model — extended IPAT model:
lnCO2emissions = residuals + lntotalpopulation + lngasconsumption +
+ lndecouplete + Lag.lnCO emissions. (2)
Regression analysis frequently uses lagged dependent variables, as is common in many research fields. To properly characterise the underlying relationship between the independent and dependent variables, researchers can consider time-varying autocorrelation and persistence in the data by using lag-dependent variables, which capture the historical effect of the dependent variable on itself (Gujarati, 2004). By accounting for previous values of the dependent variable, which may be correlated with the independent variables as well as the error term, a lagged dependent variable can be included in the regression analysis to improve the accuracy of the coefficient estimates and the overall goodness of fit of the model (Wooldridge, 2010).
The natural logarithms of the original IPAT identity variables, the estimated residual to account for anticipated endogeneity, and the dependent variable's lagged variable are all used in the second econometric estimation. Natural logarithms help change the expression of the multiplicative form of the IPAT and allow for the introduction of other variables in the model. Endogeneity issues will likely arise due to the nature of the dependent and independent variables. The former measures CO2 Emissions, and the latter estimates technology using CO2 Emissions. Variables that introduce endogeneity are Endogenous Explanatory Variables (EEVs). A Control Function (CF) solves the problem of such EEVs in both linear and nonlinear models (Wooldridge, 2015). A lagged value or a previous value of a dependent variable (i.e., Lag.ln CO2emissions) is reliable as a proxy variable. An outcome of the dependent variable from an earlier period can be a valuable proxy for various omitted factors or variables.
The Markov switching dynamic regression modelling
Markov chains are statistical models that use a set of probabilities determined by past events to represent a series of future events. The Markov switching dynamic regression model (MSDRM) quantifies these stages or states in the data, which also calculates the likelihood that the series will change states (Hamilton, 1989). For example, p11 represents the probability that the series is in State 1 at time t and remains in State 1 at time t + 1. P12 represents the probability that the series is in State
1 at time t and transitions to State 2 at time t+1. Dynamic regression models are used to model high-frequency data and allow for quick adjustment after the process changes state.
Results
Determining structural breaks in CO2 emissions
The swald test determined if there were any structural breaks in the CO2 emissions data. The null hypothesis of this test states that there is no structural break. A statistically significant p-value of 0.0000 was obtained, showing a break in 1990 that resulted in two states of CO2 emissions. With the results of this estimation test for structural breaks, the MSDRM seeks to quantify these two states. The following graphical illustrations show the structural breaks in CO2 over time (Figures 1, 2).
_л
/ А
к/ г
V V7V
I960 1980 2000 2020
vear
Figure 1. Carbon dioxide emissions in South Africa, 1960-2020 Source: compiled by the authors.
Figure 2. Carbon Dioxide emissions plotted using the natural logarithm of the data Source: compiled by the authors.
The IPAT summary data are shown in Table 2. Input and output variables for calculating the energy decoupling technical efficiency are also displayed in the table. The previously established pattern for CO2 emissions indicates a fundamental break, with rising emissions in the second regime. The political situation in South Africa, which affected industrial activities before and after independence in 1994, may cause the pattern.
Table 2
Summary statistics of IPAT variables
Variable Obs Mean Std. Dev. Min Max
CO2 Emissions 59 7.497467 1.136301 5.727223 9.922518
Total Population 61 3.35e+07 1.55e+07 2391309 5.86e+07
Gas Consumption (TWh) 56 15.78599 16.58234 0 44.10787
GDP per capita 61 3195.873 2119.326 443.0099 8007.477
Energy decoupling TE 59 0.3345763 0.2769669 0.076 1
ln CO2em 59 2.00366 0.1478537 1.745231 2.294807
lnGDP per capita 61 7.784901 0.8385268 6.093592 8.988131
lnTotPop 61 17.13301 0.7704493 14.68735 17.88553
Notes: In CO2em = natural logarithm of CO2 emissions; InGDP per capita = natural logarithm of GDP per capita; InTotPop = natural logarithm of the total population. Source: authors' calculation.
Residuals prediction
According to the study, a good model for residual prediction is the Kaya identity. Using the MSDRM to estimate the parameter estimates of the natural logarithms for the identity, residual prediction was the next step in the data-generating process for the residuals (Table 3). The predicted values are the same metric as the model's dependent variable. Before obtaining these estimation results, it was necessary to remedy multicollinearity. Estimating a linear regression model allowed for multicollinearity diagnostics and the variables with high variance inflation factors (VIFs), thereby excluding variables with these high VIFs. The predicted residuals serve as a control for endogeneity in the final model estimating CO2 emissions. An increase of 1 % in LNGDPPERCAPITA and LNPOPHISTORICALESTIMATES variables causes a -0.1948 % decrease and a 1.3170 % increase in emissions.
Table 3
Markov switching dynamic regression model on Kaya identity
Indicator Coef. Std. Err.
LNANNUAL CO2 EMISSION
LNGDPPERCAPITA -0.1948*** 0.0453
LNPOPHISTORICALESTIMATES 1.3170*** 0.0319
p11 0.9492 0.0361
p21 0.1420 0.0878
Legend: * p < 1; ** p < 0.05; *** p < 0.01
Sample: 1965-2018; No. of obs = 54; Number of states = 2; Unconditional probabilities: transition AIC = -2.6307; HQIC = -2.5312; SBIC = -2.3728 Log-likelihood = 78.028207
LNANNUAL COMMISSIONS = natural logarithm of annual CO2 emissions LNGDPPERCAPITA = natural logarithm of GDP per capita
LNPOPHISTORICALESTIMATES = natural logarithm of the historical population estimates Source: compiled by the authors.
The table's results display the Markov switching dynamic regression model applied to the Kaya identity. The model examines the relationships between three independent variables and logarithmic annual CO2 emissions: logarithmic historical population estimates, logarithmic GDP per capita, and residuals from the MSDRM. The results indicate that the coefficients for logarithmic Gross Domestic Product per capita (-0.1948) and logarithmic historical population estimates (1.3170) are statistically significant (p < 0.01). The coefficients measure how much logarithmic annual CO2 emissions would change given a unit change in the independent variables. The coefficients show that logarithmic Gross Domestic Product per capita has a negative relationship with logarithmic annual CO2 emissions, and logarithmic historical population estimates have a positive relationship with logarithmic annual CO2 emissions. The table also shows the probabilities of the two states (p11 and p21), which are the transition probabilities from state 1 to state 2 and state 2 to state 1, respectively. The sample used in the analysis covers the period 1965 to 2018, and the number of observations is 54. The model used 2 states, and the evaluation criteria (AIC, HQIC, SBIC) suggest that the model is a good fit. The log-likelihood of the model is 78.028207.
The results in the estimated model
The t-test parameter estimates in the MSDRM show statistically significant relationships. The null hypothesis is that the coefficient estimate of each of the variables equals zero. Therefore, the null hypothesis of these variables is rejected, and it is concluded that the estimated coefficients are not equal to zero.
The proposed results light the relationship between CO2 emissions and independent factors, such as gas consumption, decoupling efficiency, and lagged CO2 emissions. The Markov switching dynamic regression model was applied to the extended IPAT framework (Table 4).
Table 4
Markov switching dynamic regression model on extended IPAT: Gas Consumption model
Natural logarithm of CO2 emissions
Estimated residuals 0.149***
Natural logarithm of the total population 0.000515
Natural logarithm of decoupling technical efficiency variable 0.332**
Lagged natural logarithm of CO2 emissions 0.349***
Natural logarithm of gas consumption -0.0461***
State1 constant -2.580***
State2 constant 2.774***
p11 0.8475
p21 0.1349
Observations 48
p-values in brackets * p < 0.10, ** p < 0.05, *** p < 0.01 Source: compiled by the authors.
The estimated residuals represent unexplained factors affecting gas consumption. After accounting for other model variables, a positive estimate of 0.149*** indicates a statistically significant positive association between gas consumption and unexplained factors. This suggests that the model underestimates emissions when these unaccounted-for elements are present.
The statistically significant positive influence of historical CO2 emissions on current gas consumption is indicated by the Lagged CO2 Emissions (0.349*) coefficient. Given that past emissions typically positively impact current consumption, this conclusion emphasises the significance of historical emissions levels when trying to reduce emissions in the future.
Assuming all other factors stay constant, the positive coefficient for decoupling efficiency (0.332)** shows that a rise in the technical efficiency of decoupling is linked to a statistically significant positive influence on gas consumption. This emphasises how crucial it is to cut emissions by disrupting the link between resource consumption and economic growth. Nevertheless, it is important to remember that increasing the usage of decoupling could cause energy to rebound and raise CO2 emissions.
There is a statistically significant negative correlation between gas consumption and CO2 emissions, as indicated by the coefficient for the natural logarithm of gas consumption (-0.0461***). In particular, it implies that lower CO2 emissions are
linked to higher gas consumption and vice versa. This inverse relationship suggests a tendency for CO2 emissions to rise with increasing gas use and decrease with decreasing gas usage.
In the Markov switching model, transition probabilities ("p11" and "p21") represent the chance of changing between two states or regimes. If the system is already in State 1, then "p11" (0.8475) indicates an 84.75 per cent probability of remaining in that state. "p21" (0.1349) means that there is a 13.49 per cent probability that State 2 will change to State 1. These transition probabilities describe the dynamics and switching behaviour of the model, providing insight into how changing regimes affect the link between independent variables and emissions as well as changes in the factors influencing CO2 emissions.
These findings highlight that factors such as historical levels, past CO2 emissions, and decoupling technical efficiency affect gas usage. The model identifies two separate states with different levels of gas consumption. By enabling the investigation of various states, transitions, and their corresponding impacts on CO2 emissions, the transition probabilities provide important insights into the dynamics of the model. These results can help formulate practical strategies for emissions reduction and provide insight into the intricate relationship between gas consumption and CO2 emissions in various circumstances.
Discussion
Reducing the frequency of on-grid power outages is often the main reason for implementing renewable energy technology and other alternative electrical power generation sources. An unstable grid can cause everyday disruptions, negative effects on enterprises, and impediments to socioeconomic progress in many locations (StanCin et al., 2020, Jacal et al., 2022). By investing in alternative power production systems, people and businesses can lessen their need for the grid and offer a more consistent and dependable source of electricity.
Nevertheless, the comparatively greater cost of alternative power generation than on-grid electricity supply is an issue, particularly when fossil fuel-based generators, such as gasoline generators, are utilised (Jacal et al., 2022). The energy production cost can rise dramatically due to these alternative power generation technologies' fuel, maintenance, and operating costs (Stancin et al., 2020). Because of this, many people and companies might find it financially unfeasible to completely switch to these alternatives, which would keep them dependent on fossil fuels to generate electricity.
This reliance impacts the reduction of emissions on fossil fuels. The principal cause of climate change and a major contributor to greenhouse gas emissions is power generation that uses fossil fuels. Even with the environmental risks linked to fossil fuels, some people and companies find it challenging to switch from fossil fuel-based power due to the greater cost of electricity produced by other sources. As a result, there is still a good chance that emissions from the use of fossil fuels will decline.
To get beyond these obstacles and hasten the switch to cleaner energy sources, it is imperative to address the cost and dependability aspects of alternative power
generation. Improving the overall reliability and reducing the problem of power outages can be achieved by increasing the stability and efficiency of renewable energy systems, such as wind turbines and solar panels. Governments and legislators can also contribute significantly by putting supportive laws, incentives, and subsidies in place to lessen the financial strain of switching to alternative energy sources.
Reducing reliance on fossil fuels and attaining significant emissions reductions are more likely when these issues are resolved and adopting dependable, affordable alternative power production technologies is encouraged. Switching to cleaner energy sources promotes energy security, resilience, and sustainable economic growth while also assisting in mitigating environmental effects.
Determinants of fossil fuels
The study demonstrates that historical emissions of CO2 have a large positive impact on current emissions of CO2 from gas, indicating that the CO2 buildup in the atmosphere from prior emissions continues to contribute to current emissions. This can probably be explained in multiple ways. The long residence time of CO2 in the atmosphere of CO2 in the atmosphere (Granshaw, 2020), the feedback loops associated with climate change (Ripple et al., 2023), and the inertia in transitioning to cleaner energy sources (Gielen et al., 2019) are some of the reasons. Moreover, one factor contributing to the impact of legacy emissions is the resistance to switching to cleaner energy sources. It takes time to replace or phase out fossil fuel-dependent infrastructure, industrial processes, and energy systems, which results in ongoing emissions (Gielen et al., 2019).
A potential way to cut emissions is through energy decoupling, as demonstrated by the positive correlation between energy decoupling and gas CO2 emissions levels in the estimated empirical model. Findings from earlier studies are consistent with the energy decoupling from CO2 emissions seen in this investigation (Wang and Su, 2020; Wang, Zhang, 2020; Neves, Marques, 2021; Chovancova, Tej, 2020; Wang, Zhang, 2021). According to the correlation analysis, energy decoupling and CO2 emissions have a statistically significant positive link. This finding demonstrates how high emissions may be addressed by energy decoupling or the separation of energy consumption from economic growth. Several writers stress switching from fossil fuels to more renewable and sustainable energy sources (Raihan et al., 2022; Lin and Zhu, 2019; Stancin et al., 2020; Gielen et al., 2019). Deploying technologies that enable energy decoupling is another critical step in addressing the root cause of emissions levels.
Investing in research and developing new technologies and ideas to help reduce emissions is one possible way to achieve this goal. To lessen the influence on the environment, carbon dioxide emissions from power plants and industrial operations can be captured and stored through the development of carbon capture and storage (CCS) systems. Governments can also encourage companies to use energy-efficient technologies and renewable energy sources, and they can inform the public about ways to use less energy.
A systemic shift in how energy is produced and consumed and technological solutions are also necessary to reduce emissions. This shift may entail adjustments to laws and regulations and consumer behaviour. For instance, governments can enact laws and provide financial incentives to entice companies to use energy-efficient technologies and renewable energy sources. In order to motivate customers to use less energy, they might also offer initiatives for education and awareness.
Energy decoupling, or detaching economic growth from resource consumption and environmental degradation, is important for lowering emissions, as evidenced by the finding that positive decoupling coefficients for gas consumption models imply increased CO2 emissions.
Remarkably, the natural logarithm of gas consumption shows a statistically significant negative correlation between CO2 emissions and gas consumption. As a fuel source, switching to gas usage can decrease CO2 emissions, generally preferred for sustainability and environmental reasons. Reduced emissions might result from using gas at a higher degree of energy efficiency or from using cleaner energy sources. Changing to less efficient or more carbon-intensive energy sources may be why CO2 emissions rise when gas use falls.
Increases in energy efficiency have decoupled CO2 emissions from energy consumption, lowering emissions even in the face of rising energy usage (Wang, Su, 2020; Wang, Zhang, 2020; Neves, Marques, 2021; Chovancova, Tej, 2020; Wang, Zhang, 2021), demonstrating that there is a complex relationship between energy use and CO2 emissions and that other factors like energy efficiency, technology, and policy affect emissions levels. Similarly, cutting CO2 emissions requires severing the link between economic expansion and CO2 emissions.
However, the increased use of decoupling to reduce emissions levels shows the presence of energy rebounding. Energy rebounding occurs when a corresponding rise in energy consumption counteracts the benefits of increased energy efficiency. Increased energy efficiency may lead to increased energy use (Li R., Li S., 2021; Sorrell et al., 2020) due to increased travel and greater comfort.
Conclusion
In conclusion, this study has shed important light on the South African energy industry using Markov-Switching Dynamic Regression Models (MSDRM) to quantify structural breaks in CO2 emission data. The research effectively expanded the use of MSDRM and included other variables by modifying the IPAT identity extension. This produced a new impact assessment formula.
Reducing present emissions and resolving the cumulative legacy emissions over time is crucial since the study's findings also demonstrate the strong positive influence of past CO2 emissions on current gas emissions. The switch to renewable energy sources and increased energy efficiency are two examples of the required ambitious mitigating measures. The long-term effects of legacy emissions on climate change must be adequately addressed, which requires these steps.
Finally, cooperation with international partners is crucial to address the global problem of reducing emissions levels. Engaging actively in global accords and initiatives, like the Paris Agreement, can offer a framework for curbing global warming and make it easier for countries to share best practices and pool their resources. Countries can effectively and sustainably cut their emissions by cooperating.
In summary, this study greatly advances scientific understanding by applying a modified extension of the IPAT identity and creating a methodology for measuring energy decoupling. The results emphasise the importance of considering historical emissions levels and supporting decoupling to cut CO2 emissions successfully. These results can be used by stakeholders and policymakers in South Africa and other nations to develop and carry out emission reduction plans that support a more ecologically conscious and sustainable future.
References
Abas, N., Kalair, A., & Khan, N. (2015). Review of fossil fuels and future energy
technologies. Futures, 69, 31-49. https://doi.org/10.10167j.futures.2015.03.003 Andrew, R., & Peters, G. (2021). The Global Carbon Project's Fossil CO2 Emissions Dataset.
Zenodo: Geneva, Switzerland. Benli, B., Gezer, M., & Karakas, E. (2020). Smart City transformation for mid-sized cities: Case
of Canakkale, Turkey. Handbook of smart cities, 1-20. Berry, E.X. (2019). Human CO2 emissions have little effect on atmospheric CO2. International Journal of Atmospheric and Oceanic Sciences, 3(1), 13-26. https://doi.org/10.1007/978-3-030-15145-4_23-1. https:10.11648/j.ijaos.20190301.13 Caineng, Z.O. U., Xiong, B., Huaqing, X.U. E., Zheng, D., Zhixin, G.E., Ying, W.A. N.G., & Songtao, W.U. (2021). The role of new energy in carbon neutral. Petroleum exploration and development, 48(2), 480-491. https://doi.org/10.1016/S1876-3804 (21)60039-3 Chertow, M.R. (2000). The IPAT equation and its variants. Journal of industrial ecology, 4(4),
13-29. https://doi.org/10.1162/10881980052541927 Chovancova, J., & Tej, J. (2020). Decoupling economic growth from greenhouse gas emissions: The case of the energy sector in V4 countries. Equilibrium. Quarterly Journal of Economics and Economic Policy, 15(2), 235-251. Covert, T., Greenstone, M., & Knittel, C.R. (2016). Will we ever stop using fossil fuels? Journal
of Economic Perspectives, 30(1), 117-138. Du, K., Li, P., & Yan, Z. (2019). Do green technology innovations contribute to carbon dioxide emission reduction? Empirical evidence from patent data. Technological Forecasting and Social Change, 146, 297-303. https://doi.org/10.1016/j.techfore.2019.06.010 Gielen, D., Boshell, F., Saygin, D., Bazilian, M.D., Wagner, N., & Gorini, R. (2019). The role of renewable energy in the global energy transformation. Energy strategy reviews, 24, 38-50. https://doi.org/10.1016/j.esr.2019.01.006 Granshaw, F.D. (2020). 5: The Carbon Cycle. Climate Toolkit 2.0. Gujarati, D.N. (2004). Basic Econometrics, The McGraw. Hill Companies. Hamilton, J.D. (1989). A new approach to the economic analysis of nonstationary time series and the business cycle. Econometrica: Journal of the econometric society, 357-384. https://doi.org/10.2307/1912559 Hubacek, K., Chen, X., Feng, K., Wiedmann, T., & Shan, Y. (2021). Evidence of decoupling consumption-based CO2 emissions from economic growth. Advances in Applied Energy, 4, 100074. https://doi.org/10.1016/j.adapen.2021.100074
Islam, M.M., Hasanuzzaman, M., Pandey, A.K., & Rahim, N.A. (2020). Modern energy conversion technologies. In Energy for sustainable development (pp. 19-39). Academic Press. https://doi.org/10.1016/B978-0-12-814645-3.00002-X
Jacal, S., Straubinger, F.B., Benjamin, E.O., & Buchenrieder, G. (2022). Economic costs and environmental impacts of fossil fuel dependency in sub-Saharan Africa: A Nigerian dilemma. Energy for Sustainable Development, 70, 45-53. https://doi.org/10.1016/). esd.2022.07.007
Li, R., & Li, S. (2021). Carbon emission post-coronavirus: continual decline or rebound?
Structural Change and Economic Dynamics, 57, 57-67. https://doi.org/10.1016/). strueco.2021.01.008
Lin, B., & Zhu, J. (2019). The role of renewable energy technological innovation on climate change: Empirical evidence from China. Science of the Total Environment, 659, 1505-1512. https://doi.org/10.1016/j.scitotenv.2018.12.449
Martins, T., Barreto, A.C., Souza, F.M., & Souza, A.M. (2021). Fossil fuels consumption and carbon dioxide emissions in G7 countries: Empirical evidence from ARDL bounds testing approach. Environmental Pollution, 291, 118093. https://doi.org/10.1016/). envpol.2021.118093
Neves, S.A., & Marques, A.C. (2021). The substitution of fossil fuels in the US transportation energy mix: Are emissions decoupling from economic growth?. Research in Transportation Economics, 90, 101036. https://doi.org/10.1016/j.retrec.2021.101036
Pham, N.M., Huynh, T.L. D., & Nasir, M.A. (2020). Environmental consequences of population, affluence and technological progress for European countries: A Malthusian view. Journal of environmental management, 260, 110143. https://doi.org/10.1016/jjenvman.2020.110143
Project, G.C. 2021. Supplemental data of global carbon budget 2021 (version 1.0). Global Carbon Project.
Raihan, A., Muhtasim, D.A., Farhana, S., Pavel, M.I., Faruk, O., Rahman, M., & Mahmood, A. (2022). Nexus between carbon emissions, economic growth, renewable energy use, urbanization, industrialization, technological innovation, and forest area towards achieving environmental sustainability in Bangladesh. Energy and Climate Change, 3, 100080. https://doi.org/10.1016/j.egycc.2022.100080
Rehman, A., Rauf, A., Ahmad, M., Chandio, A.A., & Deyuan, Z. (2019). The effect of carbon dioxide emission and the consumption of electrical energy, fossil fuel energy, and renewable energy, on economic performance: evidence from Pakistan. Environmental Science and Pollution Research, 26, 21760-21773. https://doi.org/10.1007/s11356-019-05550-y
Ripple,W.J.,Wolf,C.,Lenton,T.M., Gregg,J.W.,Natali, S.M., Duffy,P.B., & Schellnhuber,H.J. (2023). Many risky feedback loops amplify the need for climate action. One Earth, 6(2), 86-91.
Skanberg, K., & Svenfelt, A. (2022). Expanding the IPAT identity to quantify backcasting sustainability scenarios. Futures & Foresight Science, 4(2), e116. https://doi.org/10.1002/ ffo2.116
Sorrell, S., Gatersleben, B., & Druckman, A. (2020). The limits of energy sufficiency: A review of the evidence for rebound effects and negative spillovers from behavioural change. Energy Research & Social Science, 64, 101439. https://doi.org/10.1016/j.erss.2020.101439
Stancin, H., Mikulcic, H., Wang, X., & Duic, N. (2020). A review on alternative fuels in future energy system. Renewable and sustainable energy reviews, 128, 109927. https://doi.org/10.1016/j. rser.2020.109927
Wang, Q., & Su, M. (2020). Drivers of decoupling economic growth from carbon emission-an empirical analysis of 192 countries using decoupling model and decomposition method. Environmental Impact Assessment Review, 81, 106356. https://doi.org/10.1016/j. eiar.2019.106356
Wang, Q., & Zhang, F. (2020). Does increasing investment in research and development promote economic growth decoupling from carbon emission growth? An empirical analysis of BRICS countries. Journal of Cleaner Production, 252, 119853. https://doi.org/10.1016/j. jclepro.2019.119853
Wang, Q., & Zhang, F. (2021). The effects of trade openness on decoupling carbon emissions from economic growth-evidence from 182 countries. Journal of cleaner production, 279, 123838. https://doi.org/10.1016/jjclepro.2020.123838 Wooldridge, J.M. (2010). Econometric analysis of cross section and panel data. MIT press. Wooldridge, J.M. (2015). Control function methods in applied econometrics. Journal of Human
Resources, 50(2), 420-445. https://doi.org/10.3368/jhr.50.2.420 Zou, C., Zhao, Q., Zhang, G., & Xiong, B. (2016). Energy revolution: From a fossil energy era to a new energy era. Natural Gas Industry B, 3(1), 1-11. https://doi.org/10.1016/j. ngib.2016.02.001
Appendix
Definition of variables used in equation
Short name of variable Variable
lnCO2emission Natural logarithm for CO2 emissions
Intotalpopulation Natural logarithm for total population
Ingasconsumption Natural logarithm for gas consumption
Indecouplete Natural logarithm for decouplete (i.e., the technical efficiency variable)
Lag.lnCO2emission The lagged variable for the Natural logarithm for CO2 emissions
Bio notes / Сведения об авторах
Collin L. Yobe, Postdoctoral Research Fellow, University of South Africa, Graduate School of Business Leadership, Midrand, South Africa. ORCID: 0000-0001-5270-2192. E-mail: [email protected]
Binganidzo Muchara, Doctor, Senior Lecturer. University of South Africa, Graduate School of Business Leadership, Midrand, South Africa. E-mail: [email protected]. ORCID: 0000-0003-0578-576X
Йобе Коллин Л., последипломный исследователь, Университет Южной Африки, Высшая школа бизнес-лидерства, Мидранд, Южная Африка. ORCID: 0000-0001-5270-2192. E-mail: [email protected]
Мучара Бинганидзо, доктор наук, доцент, Университет Южной Африки, Высшая школа бизнес-лидерства, Мидранд, Южная Африка. E-mail: [email protected]. ORCID: 0000-0003-0578-576X
