India's Green Growth Puzzle: Decoding the interplay between carbon dioxide emissions, energy mix and GDP

doi:10.21203/rs.3.rs-4498621/v1

We show the dynamic relation between environmental degradation, energy composition, and economic growth in India, utilizing annual data from 1985 to 2022. Our research illuminates a robust, long-term equilibrium relationship among these pivotal variables. Notably, our findings underscore the pivotal role of renewable energy adoption and strategic enhancements in energy efficiency as potent strategies for curtailing CO2 emissions. In addition, we emphasize the imperative of diversifying the energy portfolio and reducing dependence on fossil fuels. Moreover, our Granger causality analysis unveils dynamic interdependencies among these variables, enriching our comprehension of their temporal dynamics. Building upon these compelling results, we proffer a comprehensive set of evidence-based policy recommendations, encompassing the incentivization of renewable energy sources, the augmentation of energy efficiency measures, the introduction of carbon pricing mechanisms, and the steadfast pursuit of sustainable development objectives. This synthesis of findings, intricately coupled with the robust VECM methodology, equips policymakers with invaluable insights to navigate the multifaceted landscape of sustainable development in the dynamic Indian context.

Energy mix

Economic growth

Time series analysis

Vector Error Correction Model

CO2 emissions

Fossil fuels

Granger causality analysis

Reducing the emissions of dangerous gases increasing Earth's temperature is urgently needed, as climate change and global warming are the most pressing concerns facing modern society (Metz, Davidson & Meyer, 2007; Owusu & Asumadu, 2016). The United Nations Framework Convention on Climate Change (UNFCCC), established in 1992, the Kyoto Protocol, adopted in 1997 and coming into effect in 2005, the Intergovernmental Panel on Climate Change (IPCC), which has been in operation since 1988, and the Paris Agreement and Sustainable Development Goals (SDG), both adopted in 2015, are some of the major international efforts to mitigate the disastrous effects of climate change. All of these programmes, meanwhile, have not been able to reduce world emissions significantly. Evidence strongly suggests that if current levels of pollution are not properly dealt with, they will not only cause a drastic effect on the health of living beings around the world but also cause more frequent and erratic natural disasters, ultimately hurting the aspirations of national as well as global economic growth (Azam, Khan, Abdullah, & Qureshi, 2016).

There are two schools of thought on solving the global emission problem. According to the first perspective, carbon emissions should be reduced, which means affluent nations should slow down their revenue growth while underdeveloped nations should limit their expansion (Dinda & Coondoo, 2006). On the other hand, the second viewpoint puts forth the environmental Kuznets curve (EKC) theory, which states that when the economic expansion is in its early stages, pollution and environmental degradation will rise in tandem with income per capita. However, economic growth and the elimination of pollution and environmental degradation slow down at a certain income level. According to Peng and Sun (2010), the EKC theory suggests that economic growth resolves the problems posed by environmental degradation. In the age of rapid growth and development, countries aim to achieve both economic expansion and environmental sustainability simultaneously. Hence, they must identify the important developmental and environmental variables, as understanding the complex relationship between such variables can help develop policies that support both goals in tandem. In light of this, this study looks into how GDP, total energy consumption, and power generation from various sources affect carbon dioxide emissions to comprehend the connection between India's development, energy mix, and environmental degradation.

As per the 2007 World Bank Report, about 60% of greenhouse gas (GHG) emissions are attributed to carbon dioxide (CO2), making it one of the primary pollutants. As a result, governments everywhere have recognised the urgent need to reduce the negative effects of CO2 emissions and, by SDG 13, have implemented national policy decisions to establish a low-carbon development strategy (Küfeoğlu, S., 2022). With 2.7 billion metric tons of CO2 emissions in 2022, India is the world’s third largest CO2 emitter, with a 7% increase in emissions in 2022 relative to 2021 (Liu, Z., Deng, Z. et al., 2022). Moreover, as energy consumption drives industrial production, transportation, and other essential economic activities, it is a key driver of economic progress. At the same time, energy consumption negatively impacts the environment, especially when producing toxic CO2 emissions. In India, coal accounts for about 58% of energy consumption, followed by hydrocarbons (38%), nuclear energy, and hydroelectric power for the remaining 4%. India produced 1.85 billion metric tonnes of carbon dioxide from coal burning in 2022. As a result, decarbonizing the energy sector is urgently needed since it is now a major source of CO2 emissions in India. Energy production, as well as consumption, is shown to be a major contributor to greenhouse gas emissions, accounting for about 25% of total GHG emissions in the United States alone (M.M. Rahman, 2020; D. Nong, P. Simshauser, D.B. Nguyen, 2021). Thus, this study adds to the body of literature in the Indian context by including power production from various sources (nuclear, renewable, and fossil fuels) as independent variables influencing CO2 emissions. This methodology aligns with the approach taken in previous studies, which have underscored the importance of differentiating between energy production and consumption in assessing environmental and economic impacts (Apergis and Payne, 2014; Ozturk and Acaravci, 2013). The GDP has a significant effect on CO2 emissions. From 1970 to 2021, while India’s per capita GDP has grown by 1899%, per capita CO2 emissions have also grown by 491%, indicating a close correlation between the two variables — a typical feature of a growing, industrializing economy.

The research paper aims to analyze the correlation among the previously mentioned variables: GDP, energy consumption, CO2 emissions, electricity generation from fossil fuels, nuclear resources, and renewable sources. The objective is to gain insight into the intricate connection between environmental degradation, energy mix, and economic growth in the context of India from 1985 to 2022. The study begins with an analysis of descriptive statistics, followed by unit root tests and co-integration testing. The study utilises the Vector Error Correction model, Granger-causality, variance decomposition analysis, and model diagnostic tests to ensure the reliability of the research findings. The final section of the text presents India's viewpoint on the global discourse surrounding climate change. It contributes to the current body of knowledge on energy policy and climate change by offering significant policy recommendations.

The subsequent paper is structured in the following manner. Section 2 focuses on the "Literature Review"; Section 3 examines the "Methodology" and Section 4 gives the "Results and Discussion". The final part of the study focuses on the "Conclusions" and "Policy Recommendations".

Numerous studies have been conducted that have sought the relationship between energy, GDP, and carbon dioxide (CO2) emissions. Vladislav Marjanovica, Miloš Milovancevic, and Igor Mladenovic conducted a study to develop and implement the Extreme Learning Machine (ELM) to forecast the GDP growth rate using carbon dioxide emissions as a basis. Regarding the coefficients of determination, their research achieved better results than previous studies, including those that utilised genetic programming (GP) and artificial neural networks (ANN). The intricate correlation between economic growth and carbon dioxide emissions, shaped by technical progress and regulatory laws, amplifies the significance of this research. The ELM model's findings have significant policy implications as policymakers can utilise it to formulate sustainable climate change policies that promote economic growth while considering environmental concerns. This involves actively creating regulatory policies that promote cleaner and more ecologically sustainable technologies.

Tommaso Luzzati and Marco Orsini examined the correlation between CO2 emissions, real GDP, and energy use in nine developed economies from 1870 to 2008. The researchers employ parametric and semi-parametric additive models to examine the long-term changes in these crucial elements. Ultimately, their findings validate the EKC hypothesis, which posits a curvilinear relationship between economic progress and environmental effects. This offers vital perspectives on the pivotal overlap between economic growth and environmental deterioration. Tiwari's study uses a structural vector autoregression (VAR) methodology to investigate the complex relationship between India's carbon dioxide (CO2) emissions, economic growth, and renewable energy consumption. The study's results demonstrate that the utilisation of renewable energy sources has a beneficial impact on both GDP growth and the decrease of CO2 emissions. Positive changes in renewable energy consumption positively correlate with economic growth, but negative correlations exist between these changes and carbon dioxide emissions. Nevertheless, a decrease in GDP has a substantial and adverse effect on CO2 emissions. Moreover, the variance decomposition analysis reveals that renewable energy consumption significantly contributes to explaining prediction inaccuracies in GDP, but its impact on explaining forecast errors in CO2 emissions is minor. The results of this study strongly suggest that governments should give high importance to promoting the use of renewable energy to attain sustainable economic growth and reduce CO2 emissions. To do this, it is necessary to enact policies that promote the utilisation of renewable energy sources, such as tax incentives, financial support, and legislation that fosters the advancement and use of renewable energy technologies. Moreover, implementing measures that support energy efficiency and conservation can effectively decrease CO2 emissions and stimulate economic development.

Marco Mele and Nicolas Schneider examined the causal connections between the production of solar and wind energy, the usage of coal, economic development, and CO2 emissions in China, India, and the United States, the three world's largest economies. Their research, employing sophisticated machine learning methods, unveiled substantial findings, indicating that China and the United States are making progress in decreasing carbon emissions by adopting renewable energy sources. However, the trajectory of India's emissions is worrisome as it shows a considerable growth in CO2 emissions. These findings emphasise the urgent need for a comprehensive shift away from fossil fuels and towards renewable energy sources to reduce carbon emissions efficiently. China and the United States are highlighted as countries that offer more encouraging models of sustainable behaviour. As an emerging leader in renewable energy, India should prioritize and accelerate the implementation of low-carbon energy sources in its energy portfolio. Consequently, the suggested policy measures involve promoting renewable energy sources, decreasing reliance on coal, and fostering economic expansion while restricting carbon emissions. Patrcia Belfiore, Marcos Roberto Luppe, and Marcos Severo comprehensively analyzed the intricate interplay between global economic growth and CO2 emissions. A multilevel regression model was employed to analyse data from 1800 to 2016. The findings indicate a positive association between CO2 emissions and economic growth across 187 countries. Nevertheless, this link lacks consistency across many countries and periods. The report also identifies key factors that impact this correlation, including energy efficiency, the utilisation of renewable energy, and environmental regulations. Consequently, the paper suggests that policies aimed at fostering sustainable economic development and mitigating carbon emissions should prioritize initiatives that improve energy efficiency, encourage the utilisation of renewable energy sources, and implement robust environmental regulations. Emmanouil Hatzigeorgiou, Heracles Polatidis, and Dias Haralambopoulos conducted a study on the intricacies of CO2 emissions, GDP, and energy intensity in Greece over a span of three decades (1977-2007). They tried to unravel the intricate cause-and-effect connections between these variables using state-of-the-art time series analytical methods. The researchers systematically collected yearly data from multiple sources, enabling them to establish definitive conclusions. Their research demonstrates a robust correlation between CO2 emissions, GDP, and energy intensity, supported by a stable long-term equilibrium connection. The data reveals that Greece has successfully separated the increase in CO2 emissions from economic development in recent years, a phenomenon that may be attributed to significant improvements in energy efficiency. Nevertheless, it emphasises a concerning trend of increasing energy intensity, suggesting a potential decline in overall energy efficiency throughout the duration of the study. The paper highlights the crucial policy implications for Greece, emphasising the pressing necessity to decrease energy intensity while simultaneously enhancing economic growth in order to attain sustainable development. Furthermore, it underscores the importance of comprehending the intricate correlation between energy consumption and GDP to effectively address the task of maintaining economic growth amidst fluctuating energy supply circumstances.

Wenwen Wang, Man Li, and Ming Zhang analysed the changes in the decoupling indicator between energy-related CO2 emissions and GDP in China from 1996 to 2013. They employed the Tapio decoupling method and the LMDI method for their examination. The figures indicate that China's energy-related carbon dioxide (CO2) emissions have been consistently rising, seeing an average annual growth rate of 6.28%. Remarkably, its growth entailed an initial gradual phase followed by a rapid surge. Although emissions have increased, the decoupling index has shown significant improvement, suggesting a decrease in the Chinese economy's dependence on energy consumption and CO2 emissions. Based on the research, it is recommended that China should persistently implement policies that encourage energy saving, enhance energy efficiency, and promote the development of renewable energy sources. Implementing such measures is crucial for achieving sustainable economic growth while mitigating environmental impacts. Moreover, the study highlights the significance of doing further investigation to gain a deeper understanding of the underlying causes of fluctuations in the decoupling indicator. This understanding will have an impact on future policy-making endeavours.

Using the autoregressive distributed lag method, Sulaiman Chindo and Abdulsamad Abdulrahim looked into the cointegration between Nigeria's GDP, CO2 emissions, and energy consumption. Based on empirical data, there appears to be a sustained correlation between GDP, CO2 emissions, and energy usage. The study has demonstrated that CO2 emissions substantially positively impact GDP over the long and short run. On the other hand, it has been discovered that energy use significantly lowers GDP in the short term. The report suggests the utilisation of renewable energy sources such as solar and wind to diminish CO2 emissions and maintain sustainable long-term economic development. The paper proposes the implementation of environmentally-friendly energy policies and the exploration of alternative energy sources that have reduced greenhouse gas emissions. These measures are recommended as part of Nigeria's ongoing transformation programme. Sebastián Lozano and Ester Gutiérrez utilised a non-parametric frontier technique called Data Envelopment Analysis (DEA) to construct a model that examines the connections between population, GDP, energy consumption, and CO2 emissions. The methodology is applied using the US scenario. To determine the lowest level of greenhouse gas (GHG) emissions that can be sustained by a specific population, gross domestic product (GDP), energy intensity, or carbonisation index, as well as the highest GDP that can be sustained by a specific population, energy intensity, and carbonisation intensity, several Linear Programming (LP) models are created. The results suggest that the suggested methodology can offer valuable understanding of the connections between these factors and can aid policymakers in formulating efficient strategies to decrease CO2 emissions while maintaining economic growth. The policy recommendations encompass promoting energy efficiency, investment in renewable energy sources, and establishing carbon pricing mechanisms to encourage reductions in emissions.

2.1. Data & Period of Analysis

The goal of this research is to comprehend the intricate association that exists between economic growth, the energy mix, and environmental degradation. The variables used in this work are presented in Table A.

[Insert Table A Here]

The time series data for these variables has been collected from multiple sources, including the World Bank - World Development Indicators (2014), the U.S. Energy Information Administration (EIA), and the Energy Institute Statistical Review of World Energy (2023). The data covers the period from 1985 to 2022 and pertains to India.

2.2. Methodology

The relationship between the variables included in the study is modelled as follows with CO2 emissions as the dependent variable:

In time series analysis, it is common practice to apply a logarithmic transformation to the data. Taking the log of variables, particularly in the case of economic time series, can stabilize the variance of a series which is critical for meeting the assumption of homoscedasticity in time series models (Gujarati and Porter, 2009), linearize relationships between variables, simplifying modeling and interpretation in linear regression frameworks (Wooldridge, 2012) reduce skewness in series showing exponential growth or high skewness (Box et al., 2015) and, finally, enable the interpretation of results in terms of percentage changes, often more intuitive and relevant for policy implications (Stock and Watson, 2015). In the present study, all the variables are first log-transformed and then the Vector Error Correction Model is specified as follows:

The variables lnCO2, lnTEC, lnEPF, lnEPN, lnEPR, and lnGDP represent the log transformations of CO2 emissions, total energy consumption, electricity generation from fossil fuels, electricity generation from nuclear sources, electricity generation from renewable sources, and GDP, respectively. In the context of using annual data, 𝑡 represents the year. The difference operator is represented by the symbol ∆, the error correction term is 𝐸𝐶𝑇, ε represents the serially independent error 𝑡−1 terms, and α, β, and δ are the parameters that need to be estimated.

4.1. Descriptive Statistics Analysis

We have started our analysis with descriptive analysis. The results can be inferred from Table 2. The positive skewness values for all the variables indicate that an increased trend is present in all six-time series. The high mean value for EPF shows that India relies on fossil fuels for a majority of its electricity production followed by renewable sources and nuclear sources. In the next step, we conducted the stationary test to avoid any spurious relationship among them.

[Insert Table 1 Here]

4.2. ADF Unit Root Test

Before performing the Augmented Dickey-Fuller (ADF) unit root tests, the most suitable lag duration was calculated using the Akaike Information Criterion (AIC). The lag length that yielded the best results was determined to be eight. Subsequently, the Augmented Dickey-Fuller (ADF) tests were applied to each of the variables that had been converted using logarithmic functions in a systematic manner. The findings of this study are displayed in Table 2.

[Insert Table 2 Here]

At the level (I(0)), all variables exhibit non-stationarity. This observation is aligned with the common characteristics of economic and environmental time series data, where non-stationarity at the level is frequently observed (Enders, 2014). Upon differencing the variables, they all show strong evidence of stationarity. The presence of non-stationarity at the level and stationarity after the first difference indicates that each of these variables is integrated into order one, denoted as I(1). This attribute is crucial in determining the selection of econometric modelling methods and is consistent with the ideas described in Hamilton (1994). It implies that while the levels of these series exhibit random walks, their changes over time are mean-reverting, making them suitable for analysis using techniques that accommodate integrated variables, such as error correction models or cointegration analysis, as discussed in Engle and Granger (1987), and reinforced by Johansen (1995).

4.3. Johansen Cointegration Test

The Johansen test is pivotal in time series analysis, particularly in the context of multivariate systems where long-term relationships among variables are of interest. It allows for the determination of the number of cointegrating vectors in a non-stationary time series dataset, indicating long-term equilibrium relationships among the variables. The importance of this step is underscored by its ability to discern whether individual non-stationary time series move together over time, a critical aspect in understanding the dynamics of economic and environmental systems (Johansen, 1991; Engle & Granger, 1987).

[Insert Table 3 Here]

[Insert Table 4 Here]

The Trace test results indicate the rejection of the null hypothesis for the presence of no cointegrating equation and suggest two cointegrating equations at the 5% significance level. Similarly, the Max-Eigenvalue test suggests the presence of one cointegrating equation at the 5% significance level. These results are indicative of at least one, possibly two, long-term stable relationships among the variables thus confirming the existence of cointegration among these variables which suggests that any short-term deviations among these variables are corrected over time, bringing the variables back to a long-term equilibrium state.

4.4. Vector Error Correction Model

All variables under consideration became stationary at the first difference. The Johansen cointegration test indicated the presence of cointegrating relationships among these variables. Given these two conditions - stationarity at first differences and cointegration - the Vector Error Correction Model (VECM) becomes an appropriate econometric approach. VECM is specifically designed for use with multiple, cointegrated time series data. It captures both the short-term dynamics and the long-term relationships among the variables. In a VECM framework, the short-term adjustments towards the long-term equilibrium are explicitly modeled through the error correction term (Banerjee et al., 1993). Besides, multivariate causality can easily be inferred.

[Insert Figure 1 Here]

4.4.1. Error Correction Term

The results of the Vector Error Correction Model developed in this study state that the error correction term [ECT(-1)] is -0.27, i.e. it is negative. Besides, the t-statistic also shows that the ECT(-1) is statistically significant at 95% confidence level. This implies a long-run equilibrium relationship among lnCO2, lnTEC, lnEPF, lnEPN, lnEPR and lnGDP.

4.4.2. Long-Run Cointegrating Relationship

The following table shows the coefficients of the variables in the cointegrating equation which is represented as:

[Insert Table 5 Here]

All the coefficients are found to be statistically significant at the 95% confidence level. This indicates that there each variable on the right-hand side of the cointegrating equation has a long-run equilibrium relationship with lnCO2. Moreover, these coefficients represent that a 1% increase in total energy consumption results in a 0.18% rise in CO2 emissions.; a 1% increase in electricity production from fossil fuels results in a 0.71% increase in CO2 emissions; a 1% increase in electricity production from nuclear sources results in a 0.03% decrease in CO2 emissions; a 1% increase in electricity production from renewable sources results in a 0.07% increase in CO2 emissions; and a 1% increase in GDP results in a 0.19% increase in CO2.

4.5. Granger Causality Test

[Insert Table 6 Here]

In our analysis of Granger causality at a 95% significance level, we observed notable insights into the interplay between various economic and potentially environmental variables. Specifically, the results indicated that CO2 levels or environmental factors (LNCO2) do not precede changes in a financial indicator (LNF), suggesting that environmental changes do not have a short-term predictive impact on this particular financial variable. Additionally, fluctuations in a financial variable (LNF) were found not to be predictors of changes in another economic variable (LNN), indicating a lack of immediate influence of financial changes on this economic parameter. Furthermore, our findings revealed that changes in an economic or social variable (LNN) do not lead to short-term impacts on GDP growth (LNGDP), highlighting the complexity in the relationship between social/economic factors and macroeconomic growth. Finally, the analysis showed that GDP growth (LNGDP) does not predict short-term changes in a resource-related or environmental variable (LNR), which could be significant in understanding the lagged effects of economic growth on environmental or resource parameters. These results underscore the multifaceted nature of the relationships between environmental, financial, and economic variables, emphasizing the necessity for long-term planning and comprehensive policy formulation to address the intricate dynamics between these factors.

[Insert Table 7 Here]

In our exploration of the Granger causality relationships at a 90% significance level, we discerned several intriguing dynamics among different economic indicators. Firstly, we found that economic conditions, as indicated by lnTEC, do not have a short-term predictive relationship with CO2 levels or environmental factors represented by lnCO2. This observation suggests a degree of independence in the short-term evolution of these two variables. Intriguingly, CO2 levels or environmental factors (lnCO2) were observed not to be predictors of short-term changes in another economic or social variable (lnEPN), indicating a potential disconnect or lag in their influence on these factors. Additionally, our results showed that GDP growth (lnGDP) has a significant short-term influence on CO2 levels (lnCO2), implying a closer, more immediate relationship between economic growth and environmental changes. The reverse relationship, where economic conditions (lnTEC) predict changes in a financial variable (lnEPF), further highlights the nuanced interplay between broader economic conditions and specific financial indicators. Moreover, the finding that GDP growth (lnGDP) does not predict short-term changes in lnTEC, while lnTEC can predict changes in lnGDP, suggests a complex, bidirectional relationship between overall economic health and GDP growth. These insights are crucial for understanding the immediate impacts and feedback loops within the economic system, particularly in the context of environmental, financial, and broader economic variables, and emphasize the need for policies that consider these intricate relationships in both the short and long term.

4.6. Variance Decomposition Analysis

The above analysis represents the variance decomposition results derived from the Vector Error Correction Model which provides insights into the relative importance of the variables in explaining the forecast error variance of CO2 emissions (lnCO2) over different time horizons. The increasing standard error (S.E.) across periods reflects the diminishing predictability of the model with the extension of the forecast horizon.

[Insert Table 8 Here]

Initially, at period 1, the forecast error variance of lnCO2 is entirely explained by its own shocks, indicating the predominant influence of the variable's immediate past values. This observation aligns with the expectation that in the short run, a variable is largely influenced by its own lags. As the analysis progresses through subsequent periods, a gradual decline in the percentage contribution of lnCO2’s own shocks to its variance is observed. This decline is accompanied by an incremental increase in the contribution of other variables, notably lnTEC (total energy consumption), to the forecast error variance of lnCO2. This shift suggests that as we move further away from the immediate past, the influence of total energy consumption on CO2 emissions becomes more pronounced. Furthermore, the contributions of lnEPF (electricity production from fossil fuels), lnEPN (electricity production from nuclear sources), lnEPR (electricity production from renewable sources), and lnGDP (Gross Domestic Product) to the variance of LNCO2, though relatively modest in comparison to LNEC, are not negligible. These contributions gradually increase over time, highlighting the growing importance of various types of energy consumption and economic activity in explaining long-term CO2 emissions trends.

4.7. Model Diagnostics

The research examines the independence of the residuals by conducting tests for VEC residual serial correlation, VEC residual heteroskedasticity, and VEC residual normality. These tests aim to determine the reliability and strength of the VEC model used in the study. The subsequent tables provide a concise overview of the outcomes of all these examinations. At a 5% level of significance, the absence of serial correlation in the residuals at the given lag of the Breusch-Godfrey test, that is, the null hypothesis, cannot be rejected. Similarly, the null hypothesis that the residuals exhibit a consistent variance, as tested by the Breusch-Pagan-Godfrey test, cannot be rejected at a significance level of 5%. The null hypothesis of the Jarque-Bera test, which asserts that the residuals have a normal distribution, cannot be rejected at a significance level of 5%. The aforementioned results collectively demonstrate the robustness of the created model.

The research we have conducted has led to a set of policy proposals that attempt to address the complex relationship that exists between environmental deterioration, energy mix, and economic growth in India. India should give first priority to promoting renewable energy sources due to their significant influence on lowering CO2 emissions. This involves implementing incentives and subsidies to stimulate investments in solar, wind, and other environmentally friendly energy technologies. Furthermore, in order to address the increasing levels of CO2 emissions linked to economic expansion, India should prioritise the improvement of energy efficiency in all sectors. The adoption of energy-efficient technology and practices in industry, transportation, and households can result in a decrease in energy consumption per person. Furthermore, it is imperative to diversify the energy mix in order to decrease reliance on fossil fuels for the generation of power. Policymakers must to contemplate a progressive decrease in subsidies for fossil fuels and the advancement of cleaner alternatives such as nuclear energy and renewables. Nuclear energy, specifically, has shown promise in reducing CO2 emissions.

Moreover, to acknowledge the delayed impacts of economic growth on environmental factors, India must embrace a forward-looking approach that incorporates sustainable development objectives into its economic strategy. This entails endeavors to decrease the carbon intensity of economic activities. It is advisable to introduce carbon pricing mechanisms, such as carbon taxes or cap-and-trade systems, to create economic motivations for the reduction of CO2 emissions. The proceeds from these methods can be reinvested in renewable energy infrastructure and projects promoting sustainable development. Public awareness and education efforts play a crucial role in cultivating a culture of sustainability. Disseminating information to the public about the upsides of energy preservation and the environmental effects of energy-related decisions can lead to more environmentally conscious energy consumption practices. Furthermore, India should promote and support research and development endeavours in sustainable energy technology, carbon capture and storage, and inventive approaches to decrease energy intensity. The partnership among the government, industry, and academia has the potential to propel technological progress in various fields. It is essential to create a thorough system for monitoring and reporting environmental indicators, energy consumption, and economic growth. Periodic evaluations can offer significant perspectives for modifying policies and monitoring advancements towards sustainability objectives. It is advisable to participate in international collaboration to gain access to the most effective methods and technology for achieving sustainable development. Additionally, joining global climate agreements can help secure finance for clean energy projects. Finally, it is important to acknowledge the ever-changing nature of environmental and economic issues. Therefore, policies should be created to be flexible and responsive, with frequent evaluations and revisions to guarantee ongoing advancement toward sustainable development.

Ultimately, this study provides useful insights into the complex interplay between environmental degradation, energy composition, and economic development in India. Our analysis revealed strong evidence of stable correlations between important factors across a period of time spanning from 1985 to 2022. By utilising a rigorous methodology that incorporated time series analysis, cointegration tests, and Granger causality assessments, we discovered valuable insights regarding these linkages. The results of our study highlight the presence of stable and balanced connections over a prolonged period of time between CO2 emissions, overall energy usage, power generation from different sources, and real GDP per capita. This implies that temporary changes in these variables are gradually rectified, in accordance with economic and environmental principles.

Our investigation found that environmental changes do not have an immediate effect on financial indicators. However, economic growth has a more direct impact on environmental parameters, including CO2 emissions. Moreover, the interdependent connections between the general economic well-being and the rise of Gross Domestic Product (GDP) underscore the intricate nature of these dynamics. The variance decomposition study clarified the way in which the importance of distinct components fluctuates throughout different time periods. In the near term, the levels of CO2 emissions are mostly impacted by their recent values. However, as we progress into the future, the significance of total energy consumption becomes more prominent. The findings of our research have significant consequences for policy. In India's pursuit of sustainable development, it is crucial to prioritise energy efficiency, decrease dependence on fossil fuels, and advocate for the adoption of renewable energy sources. An all-encompassing, enduring strategy is crucial to tackle the complex interplay among environmental, energy, and economic factors. This study provides vital insights into the complex relationships that support sustainable development in India. The statement underscores the importance of comprehensive policy approaches that consider both immediate changes and long-term stability in the quest for a more environmentally friendly and economically productive future.

Ethical Approval: we declare that the study is original and follows all ethical aspects.

Consent to Participate: We declare that this manuscript is original, has not been published before, and is not currently being considered for publication elsewhere. We confirm that the manuscript has been read and approved by all named authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that all have approved the order of authors listed in the manuscript.

Consent to Publish: The authors hereby consent to the above manuscript in all forms to publish in this esteemed journal.

Authors Contributions: Rohan Kunwar and Krish Vora have written the importance of the study, literature section, written methodology, and result analysis section. Rohit Raj and Kashika Sharma have written the introduction section, overall correction of the manuscripts, and references. Debi Prasad Bal developed the idea, analyzed econometrics, and interpreted the results section.

Funding: The authors have not received any funding from any agency to develop this research paper.

Competing Interests: We declare that this manuscript has neither been published nor submitted elsewhere for publication. We further declare no conflict of interest related to this manuscript.

Availability of data and materials: The data and other materials will be shared as per the request.

Antle, J. M., & Heidebrink, G. (1995). "Environment and Development: Theory and International Evidence."*Economic Development and Cultural Change, 43(3)*, 603-625.
Antle, J. M., & Heidebrink, G. (1995). "Environment and Development: Theory and International Evidence."
Apergis, N., & Payne, J. E. (2010). "Renewable energy consumption and economic growth: Evidence from a panel of OECD countries." *Energy Policy, 38(1)*, 656-660.
Apergis, N., & Payne, J. E. (2014). "Renewable energy consumption and economic growth: Evidence from a panel of OECD countries." *Journal of Policy Modeling, 36(2)*, 342-356.
Apergis, N., & Payne, J. E. (2014). "Renewable energy consumption and economic growth: Evidence from a panel of OECD countries." *Journal of Policy Modeling, 36(2)*, 342-356.
Azam, M., Khan, A. Q., Abdullah, H. B., & Qureshi, M. E. (2016). The impact of CO 2 emissions on economic growth: evidence from selected higher CO2 emissions economies. Environmental Science and Pollution Research, 23, 6376-6389.
Banerjee, A., Dolado, J. J., Galbraith, J. W., & Hendry, D. F. (1993). *Co-integration, Error Correction, and the Econometric Analysis of Non-Stationary Data*. Oxford University Press.
Banerjee, A., Dolado, J. J., Galbraith, J. W., & Hendry, D. F. (1993). *Co-integration, Error Correction, and the Econometric Analysis of Non-Stationary Data*. Oxford University Press.
Box, G. E., Jenkins, G. M., Reinsel, G. C., & Ljung, G. M. (2015). *Time Series Analysis: Forecasting and Control*. John Wiley & Sons.
Engle, R. F., & Granger, C. W. J. (1987). "Co-integration and error correction: Representation, estimation, and testing." *Econometrica, 55(2)*, 251-276.
Engle, R. F., & Granger, C. W. J. (1987). "Co-integration and error correction: Representation, estimation, and testing." *Econometrica, 55(2)*, 251-276.
Fare, H. D. D. C. (2007). Growth and CO2 Emissions: How do Different Countries Fare?.
Fearnside, P. M. (2005). "Greenhouse gas emissions from a hydroelectric reservoir (Brazil’s Tucuruí Dam) and the energy policy implications." *Water, Air, & Soil Pollution, 133(1-4)*, 69-96.
Fearnside, P. M. (2005). "Greenhouse gas emissions from a hydroelectric reservoir (Brazil’s Tucuruí Dam) and the energy policy implications." *Water, Air, & Soil Pollution, 133(1-4)*, 69-96.
Gagnon, L., & van de Vate, J. F. (1997). "Greenhouse gas emissions from hydropower - The state of research in 1996." *Energy Policy, 25(1)*, 7-13.
Gagnon, L., & van de Vate, J. F. (1997). "Greenhouse gas emissions from hydropower - The state of research in 1996." *Energy Policy, 25(1)*, 7-13.
Grossman, G. M., & Krueger, A. B. (1991). "Environmental Impacts of a North American Free Trade Agreement." *NBER Working Paper No. 3914*.
Grossman, G. M., & Krueger, A. B. (1991). "Environmental Impacts of a North American Free Trade Agreement." *NBER Working Paper No. 3914*.
Gujarati, D. N., & Porter, D. C. (2009). *Basic Econometrics*. McGraw-Hill/Irwin.
Hamilton, J. D. (1994). *Time Series Analysis*. Princeton University Press.
Johansen, S. (1988). "Statistical analysis of cointegration vectors." *Journal of Economic Dynamics and Control, 12(2-3)*, 231-254.
Johansen, S. (1988). "Statistical analysis of cointegration vectors." *Journal of Economic Dynamics and Control, 12(2-3)*, 231-254.
Johansen, S. (1995). *Likelihood-Based Inference in Cointegrated Vector Autoregressive Models*. Oxford University Press.
Küfeoğlu, S. (2022). Emerging technologies: value creation for sustainable development (p. 512). Springer Nature.
Liu, Z., Deng, Z., Davis, S. et al. Monitoring global carbon emissions in 2022. Nat Rev Earth Environ 4, 205–206 (2023). https://doi.org/10.1038/s43017-023-00406-z
Metz, B., Davidson, O., Bosch, P., Dave, R., & Meyer, L. (2007). Climate Change Mitiga‐tion. Contribution of Working Group III to the fourth assessment report of the Intergovernmental Panel on Climate Change.
Nong, D., Simshauser, P., & Nguyen, D. B. (2021). Greenhouse gas emissions vs CO2 emissions: Comparative analysis of a global carbon tax. Applied Energy, 298, 117223.World Bank - World development Indicators, 2014;
Osterwald-Lenum, M. (1992). "A note with quantiles of the asymptotic distribution of the maximum likelihood cointegration rank test statistics."
*Oxford Bulletin of Economics and Statistics, 54(3)*, 461-472.
Osterwald-Lenum, M. (1992). "A note with quantiles of the asymptotic distribution of the maximum likelihood cointegration rank test statistics." *Oxford Bulletin of Economics and Statistics, 54(3)*, 461-472.
Owusu, P. A., & Asumadu-Sarkodie, S. (2016). A review of renewable energy sources, sustainability issues and climate change mitigation. Cogent Engineering, 3(1), 1167990.
Panayotou, T. (1993). "Empirical Tests and Policy Analysis of Environmental Degradation at Different Stages of Economic Development." *ILO Working Papers*.
Panayotou, T. (1993). "Empirical Tests and Policy Analysis of Environmental Degradation at Different Stages of Economic Development." *ILO Working Papers*.
Rahman, M. M. (2020). Environmental degradation: The role of electricity consumption, economic growth and globalisation. Journal of environmental management, 253, 109742.
Shafiei, S., & Salim, R. A. (2014). Non-renewable and renewable energy consumption and CO2 emissions in OECD countries: a comparative analysis. Energy policy, 66, 547-556.
Shahbaz, M., Loganathan, N., Muzaffar, A. T., Ahmed, K., & Jabran, M. A. (2016). How urbanization affects CO2 emissions in Malaysia? The application of STIRPAT model. Renewable and Sustainable Energy Reviews, 57, 83-93.
Shahbaz, M., Loganathan, N., Muzaffar, A. T., Ahmed, K., & Jabran, M. A. (2015). "How urbanization affects CO2 emissions in Malaysia? The application of STIRPAT model." *Renewable and Sustainable Energy Reviews, 47*, 1-10.
Stern, D. I. (2004). "The Rise and Fall of the Environmental Kuznets Curve." *World Development, 32(8)*, 1419–1439.
Stern, D. I. (2004). "The Rise and Fall of the Environmental Kuznets Curve." *World Development, 32(8)*, 1419–1439.
Stock, J. H., & Watson, M. W. (2015). *Introduction to Econometrics*. Pearson Education Limited. Enders, W. (2014). *Applied Econometric Time Series*. Wiley.
Wooldridge, J. M. (2012). *Introductory Econometrics: A Modern Approach*. South-Western Cengage Learning.

Table 1. Descriptive Statistics of the variables under study

Variables	Mean	Median	Maximum	Minimu m	Std. Dev.	Skewness	Kurtosis
CO2	1.120461	0.947477	1.925088	0.509576	0.450707	0.480472	1.861150
TEC	4149.174	3583.340	6809.620	2030.476	1488.832	0.394980	1.830569
EPF	517.9725	488.0549	950.3157	166.6983	237.8040	0.366808	1.980942
EPN	16.25366	14.98708	32.65101	4.720181	9.055456	0.412236	1.778906
EPR	110.3317	85.46169	236.0106	60.02398	49.18614	1.164826	3.352066
GDP	896.2004	543.8438	2238.127	297.9993	633.5678	0.716522	2.012568

Table 2. Results of the Augmented Dickey-Fuller (ADF) Unit Root Test

Variable	ADF-statistic	P-Value
lnCO2	-2.750374	0.2245
D(LNCO2)	-5.606031	0.0003
lnTEC	-2.084451	0.5367
D(lnEC)	-5.564695	0.0003
lnEPF	-2.708249	0.2394
D(lnEPF)	-6.468702	0.0000
lnEPN	-2.970714	0.1541
D(lnEPN)	-7.704801	0.0000
lnEPR	-1.682810	0.7382
D(lnEPR)	-6.106240	0.0001
lnGDP	-1.837536	0.6654
D(lnGDP)	-5.599155	0.0003

Note: Critical Values: 1% level: -4.243644; 5% level: -3.544284; 10% level: -3.204699

Table 3. Unrestricted Cointegration Rank Test (Trace)

Hypothesized No. of CE(s)	Eigenvalue	Max-Eigen Statistic	Critical Value	p-value
None	0.870456	71.53060	40.07757	0.0000*
At most 1	0.573965	29.86318	33.87687	0.1400
At most 2	0.457095	21.37874	27.58434	0.2540
At most 3	0.299466	12.45695	21.13162	0.5032
At most 4	0.182980	7.073210	14.26460	0.4805
At most 5	0.050823	1.825600	3.841465	0.1766

Table 4. Unrestricted Cointegration Rank Test (Maximum Eigenvalue)

Hypothesized No. of CE(s)	Eigenvalue	Trace Statistic	Critical Value	p-value
None	0.870456	144.1283	95.75366	0.0000*
At most 1	0.573965	72.59769	69.81889	0.0295*
At most 2	0.457095	42.73451	47.85613	0.1392
At most 3	0.299466	21.35576	29.79707	0.3358
At most 4	0.182980	8.898810	15.49471	0.3748
At most 5	0.050823	1.825600	3.841465	0.1766

Table 5. Coefficients of the Cointegrating Equation

Variable	Coefficient	Standard Error	t-Statistic
lnCO2(-1)	1.000000
lnTEC(-1)	-0.182229	(0.12245)	[-1.98815]*
lnEPF(-1)	-0.707412	(0.07885)	[-8.97209]*
lnEPN(-1)	0.033053	(0.01330)	[2.48443]*
lnEPR(-1)	-0.067011	(0.02031)	[-3.29876]*
lnGDP(-1)	-0.193945	(0.02756)	[-7.03695]*
TREND	0.012477	(0.00310)	[4.02202]*
C	7.071498

Table 6: Granger Causality Test Results at 95% Significance Level

Null Hypothesis (Ho)	F-Statistic	Prob
LNCO2 does not Granger Cause LNF	3.74656	0.0353
LNF does not Granger Cause LNN	5.93132	0.0068
LNN does not Granger Cause LNGDP	4.66565	0.0172
LNGDP does not Granger Cause LNR	3.99624	0.0289

Table 7: Granger Causality Test Results at 90% Significance Level

Null Hypothesis (Ho)	F-Statistic	Prob
LNEC does not Granger Cause LNCO2	2.75511	0.0797
LNCO2 does not Granger Cause LNN	2.91594	0.0696
LNGDP does not Granger Cause LNCO2	3.66636	0.0376
LNEC does not Granger Cause LNF	3.95536	0.0299
LNGDP does not Granger Cause LNEC	4.16571	0.0253
LNEC does not Granger Cause LNN	2.65221	0.0870
LNN does not Granger Cause LNGDP	4.66565	0.0172
LNGDP does not Granger Cause LNR	3.99624	0.0289

Table 8. Variance Decomposition of lnCO2 over Different Periods

Period	S.E.	% Variance of lnCO2 Explained By: lnCO2	% Variance of lnCO2 Explained By: lnTEC	% Variance of lnCO2 Explained By: lnEPF	% Variance of lnCO2 Explaine d By: lnEPN	% Variance of lnCO2 Explained By: lnEPR	% Variance of lnCO2 Explaine d By: lnGDP
1	0.034457	100.0000	0.0000	0.0000	0.0000	0.0000	0.0000
2	0.049319	98.51508	0.517947	0.005071	0.566116	0.223861	0.171924
3	0.059119	96.73229	0.985702	0.315800	0.657485	0.203802	1.104923
4	0.067861	96.06283	0.993294	0.352689	1.080886	0.184776	1.325526
5	0.075010	95.57306	1.193750	0.409093	1.103669	0.185339	1.535089
6	0.081980	95.15336	1.221637	0.436200	1.247191	0.179306	1.762307
7	0.088063	94.79883	1.316081	0.471448	1.301937	0.176809	1.934896
8	0.093911	94.51805	1.350506	0.495264	1.375772	0.173311	2.087093
9	0.099312	94.28604	1.400146	0.516031	1.420499	0.171325	2.205960
10	0.104484	94.09066	1.430143	0.533112	1.464842	0.169246	2.311999

India's Green Growth Puzzle: Decoding the interplay between carbon dioxide emissions, energy mix and GDP

Status:

Version 1

Abstract

Figures

I. Introduction

II. Literature Review

III. Methodology

IV. Results and Discussion

V. Conclusion & Policy Recommendations

Declarations

References

Tables

Status:

Version 1