A Final Paper co-authored with Ben Meyers and Amelia Batson (December 2023)

Abstract

In this paper, we examine the effect of carbon dioxide emissions on GDP growth across 129 countries over a period of 20 years: 2000 to 2019. We hypothesize that as carbon emissions increase, the associated increases in production will ultimately lead to measurable GDP growth. In the sections that follow, we introduce the relationship we examine and explain its empirical and political significance. This relationship is further contextualized through a summary and analysis of academic literature as it relates to our research question. Extensive data exists on similar topics; however, our analysis is conducted on a larger scale than most prior analyses. Then, we introduce and contextualize our data being mindful of its source. Next, we provide the multiple linear regression model and a theoretical justification for why the independent variables are chosen, as well as the expected signs of the slope coefficients. Our multivariable linear regression models found a significant and positive relationship between carbon emissions and economic growth after adjusting for trade, population growth rate, total natural resource rents, and primary school enrollment when examined globally, among high income countries, and within the region of Europe & Central Asia. Surprisingly, the opposite pattern was found among low income countries. We conclude by discussing the limitations and policy recommendations associated with our research. To reduce emissions while maintaining GDP growth, policy-makers must carefully consider the economic impacts of decarbonization to ensure continued growth and development. 

I. Introduction

When confronted about his controversial views on climate change, future president Donald Trump proclaimed, “Global warming is an expensive hoax!” that is used to justify high taxes (Trump, 2014). While such inflammatory rhetoric is certainly not the norm, the prospect of economic harm is consistently used to counter policy proposals aimed at mitigating climate change. Indeed, decarbonization will be expensive. Estimates suggest that attaining net-zero carbon emissions within industrial fields alone will cost between $11 and $21 trillion globally by 2050 (de Pee et al. 2018). Even on a global scale, these costs are nothing to scoff at. Environmental advocates and progressive politicians, in contrast, turn to research that claims “unchecked climate change could cost the global economy US$178 trillion in net present value terms from 2021-2070” (The Turning Point, 2022). No one figure for costs or damages is universally trusted, but climate change, and its prevention, is undoubtedly going to alter economic growth across the world. Calculating future costs of climate change damages and pricing sector-wide decarbonization is difficult to do; however, looking to the past to see the growth attributed to carbon emissions is more feasible. It is crucial to understand the economic effects of carbon emissions before we decide whether to fully turn away from them. This situation precipitates our study, prompting the research question: How do carbon emissions impact economic growth?

Decarbonization has become an increasingly salient issue in 21st century politics. Carbon emissions are widely recognized as detrimental to the environment, playing a significant role in exacerbating climate change and its associated consequences. Despite these environmental concerns, economic growth remains a critical objective for nations, prompting a debate on the necessity of carbon emissions within a country’s economy. In the United States, economic uncertainty is a primary reason that comprehensive environmental legislation has yet to pass Congress (International Monetary Fund, 2022). These debates will likely continue for the foreseeable future, even as other nations develop concrete plans to phase out fossil fuels and transition to an economy powered by renewable energy. Historically, discussions surrounding economic growth have often sidelined environmental considerations, leading to a laissez-faire approach that disregards the ecological cost of developmental pursuits. The political, economic, and environmental future of the world will hinge upon the results of these debates. 

Conventional framing suggests that carbon emissions have propelled economic growth over the past two centuries. Since the beginning of the industrial revolution, which was fueled by easily attainable deposits of coal and petroleum, the world has transformed in previously unimaginable ways. Largely agrarian societies like the United States became industrial powerhouses, interlinked with the world by vast transportation and satellite networks, all powered by fossil fuels. Modern economies could not have developed without carbon emissions (Zhang, 2020). On the other hand, however, new technologies offer viable substitutes for fossil fuels and other pollutants, perhaps limiting the economic consequences of reducing carbon emissions. Additionally, transitioning to green energy could generate revenue in new sectors while promoting job growth (International Labour Organization, 2022). Until recently, clean power sources were limited in availability and practicality (Holechek, 2022). As a result, the causal relationship between carbon emissions and economic growth is unclear. The purpose of this paper is to uncover if there is a statistically significant and positive relationship between these two variables. 

Much of the existing research on the relationship between carbon emissions and economic growth is at the country or regional level, often specifically looking at rapidly developing nations like India or China. As detailed in the next section, most data suggests that economic growth rises with carbon emissions; however, the scale is often limited. Against this backdrop, our research will delve into empirical analyses, drawing upon a diverse range of indicators. We aim to increase the scale of the analysis by looking at 129 countries from around the globe. We include several useful control variables and implement fixed effects within our regression to maximize the statistical soundness of our research. We also run multiple regressions with different parameters to avoid missing important conclusions that are not apparent when examining this relationship at a global scale. As the world grapples with the imperative of mitigating climate change and achieving sustainable development goals, understanding the intricate relationship between carbon emissions and GDP growth is paramount. While our research only scratches the surface of what it will take to sufficiently analyze the impacts of decarbonization, it is a starting point for further analyses. In this paper, we provide an overview of the existing literature on this issue (Section II), describe our data (Section III) and model (Section IV), report our results (Section V), discuss experimental limitations and opportunities for future research (Section VI), and offer conclusions (Section VII).

II. Literature Review

There is a wide range of literature on the topic of carbon emissions and economic growth, some of which focuses on a specific country and some of which focuses more globally.

On a country-wide level, Akalpler and Hove (2019) examine how the Indian economy is affected by carbon emissions, energy use, gross fixed capital formation, real GDP per capita, exports, and imports. By employing an autoregressive distributed lag (ARDL) model on annual data from 1971 to 2014, their goal is to determine the driving factors behind Indian economic growth. They find that the growth of the Indian economy has been encouraged by high energy consumption levels over their period of study. In the long term, however, gross fixed capital formation and exports play a pivotal role in determining growth, as indicated by the cointegration relationship identified within the error correction model. Additionally, Zhang and Cheng (2009) studied China and implemented a multivariate model of economic growth, with control variables such as energy use, capital, and population to enrich the model. Ultimately, they found that neither carbon emissions nor energy consumption were driving forces behind economic growth between 1960 to 2007. Consequently, they recommended that the Chinese government could adopt a prudent energy policy and carbon emissions reduction strategy over the long term, all while promoting sustained economic growth.

Farhani and Ozturk (2015) examined the causal relationship between CO2 emissions and real GDP in Tunisia from 1971 to 2012 and found a positive monotonic relationship between the two variables. The two researchers used an unrestricted VAR model and ultimately concluded that the research does not support the validity of the Environmental Kuznets Curve (EKC). The EKC, often cited by environmental economists, posits an inverted-U relationship between pollution and economic development (Dasgupta et al., 2002). Recent evidence suggests that the shape of the curve is flattening and shifting to the left, perhaps due to new strategies for pollution regulation and advancing technology. This finding challenges the commonly held notion that economic growth will eventually lead to higher carbon emissions, and instead suggests a more nuanced relationship, emphasizing that these relationships may not be as simply captured as a curve of a graph. In addition, they pointed to another variable, financial development, that deeply affects the relationship between CO2 emissions and real GDP. This variable enables investment into cleaner technologies and provides capital market solutions to environmentally friendly projects. 

Moving beyond the country-level, Barassi and Spagnolo (2012) sought to determine the relationship between GDP per capita and carbon dioxide emissions for six countries. The researchers concluded that, in the long run, emissions follow an upward trend linked to increased consumption, in line with the EKC. For example, they found the United Kingdom exhibited a feedback effect between emissions and output in the long run, and Italy experienced the phenomenon of reverse causality. Lastly, their multivariate generalized autoregressive conditional heteroskedasticity (GARCH) analysis revealed spillovers, suggesting policymakers should consider the link between emissions and economic volatility. Additionally, Olubusoye and Musa (2020) examined 43 African countries and found that in the majority of countries (79%), carbon emissions rose in tandem with economic growth, with only a limited number of nations (21%) experiencing a decline in emissions due to economic growth. The research underscores the need for proactive measures in most African countries where economic growth tends to lead to higher emissions, such as the widespread adoption of renewable energy, the implementation of carbon taxes, and the establishment of carbon emissions trading systems to mitigate the increase in carbon emissions. 

In a European context, Acaravci and Ozturk (2010) examined the relationship between CO2 emissions, energy consumption, and economic growth using an ARDL approach for 19 European countries from 1960 to 2005. They concluded that Denmark, Germany, Greece, Iceland, Italy, Portugal, and Switzerland showcased a positive relationship between the two variables. Moreover, Denmark and Italy confirmed the validity of the EKC curve. Overall, the study arrived at a conclusion which suggested that energy conservation policies are unlikely to adversely affect economic growth in the countries examined. Looking globally, Yang, Hao, and Feng (2021) acknowledge that for much of history, economic growth and carbon emissions have been intensely interlinked. They analyze the decoupling process between these two that is occurring for the sake of low-carbon economic development. They determine that decoupling is possible and effective, as evidenced by Europe and North America, but requires global participation and can be impacted by recessions or financial crises.

Our paper aims to add to the existing literature in providing a global perspective of the relationship between carbon emissions and economic growth. While much of the literature suggests that economic growth increases with carbon emissions, it usually focuses on a single country or region and different studies produce conflicting results. Against this backdrop, our research will delve into empirical analyses, drawing upon a diverse range of indicators. We aim to increase the scale of the analysis by looking at 129 countries from around the globe. We investigate this relationship on a global level and by income level and region. By segmenting the world this way, this paper is able to recognize trends within each subcontext and parse through the complex relationship to arrive at a universal regression coefficient. Furthermore, we control for four key variables, which serves to limit omitted variable bias and lead to a more efficient estimator. 

III. Data Description

Our research hinges on data sourced from the World Bank’s World Development Indicators. The World Bank sources data from member countries, national statistics offices, and surveys. Our key variables of analysis are carbon emissions (independent variable) and GDP per capita growth (dependent variable). Additionally, we incorporate four control variables that are known to impact economic growth: trade, population growth rate, total natural resource rents, and primary school enrollment.

As defined by the World Bank, GDP per capita growth is the annual percentage growth rate of GDP per capita (gross domestic product divided by midyear population) based on constant local currency (World Bank, GDP per capita growth, 2022). This metric provides a more compelling measure of economic growth by going beyond aggregate economic output. It offers insights into the tangible impact of growth on citizens’ daily lives, reflecting income, living standards, and overall quality of life — an approach that aligns with a more relevant and people-centric assessment of economic success. Carbon emissions (metric tons per capita) are those stemming from the burning of fossil fuels and the manufacture of cement, including carbon dioxide produced during consumption of solid, liquid, and gas fuels and gas flaring (World Bank, CO2 emissions, 2022). The World Bank sources carbon emissions data from Climate Watch Historical CO2 Emissions. Notably, it excludes greenhouse gasses (GHG) like methane, water vapor, CFCs, and nitrous oxide, as they show less direct correlation to economic growth compared to CO2. Trade is the sum of exports and imports of goods and services measured as a share of gross domestic product (World Bank, Trade, 2022). Annual population growth rate for year t is the exponential rate of growth of midyear population from year t-1 to t, expressed as a percentage, and population is based on the de facto definition of population, which counts all residents regardless of legal status or citizenship (World Bank, Population growth, 2022). Total natural resources rents are the sum of oil rents, natural gas rents, coal rents (hard and soft), mineral rents, and forest rents (World Bank, Total natural resources rents, 2022). Primary school enrollment is the gross enrollment ratio (the ratio of total enrollment, regardless of age to the population of the age group that officially corresponds to primary education) (World Bank, School enrollment, 2022). 

Using individual data for each of these variables, we constructed a panel dataset for 129 countries from 2000 to 2019. We chose to only focus on countries that had data for the majority of variables during the time period and included countries that either had complete data or were missing 5 or fewer data points. The exception to this was the United States, which was missing 12 data points but was still included in order to have more than one country in the North America region represented. This time period was deliberately selected to encompass the globalization wave post the turn of the century, while excluding any skewed data due to COVID-19 which significantly impacted economic growth. Table 1 shows the descriptive statistics for the variables, including the mean, standard deviation, and minimum and maximum values of the observations.

IV. Regression Model

The model includes one dependent variable (GDP per capita growth), one main independent variable (carbon emissions), and four other control variables (trade, population growth rate, total natural resource rents, and primary school enrollment). Additional control variables were included to reduce bias and make such variables independent of each other. 

Fixed Effects Model: GDP Growth_it = β₀ + β₁ (CO2 Emissions)_it + β₂ (Trade)_it + β₃ (Population Growth)_it  + β₄ (Resource Rents)_it  + β₅ (School Enrollment)_it + α_i + γ_t + u_it

Random Effects Model: GDP Growth_it = β₀ + β₁ (CO2 Emissions)_it + β₂ (Trade)_it + β₃ (Population Growth)_it  + β₄ (Resource Rents)_it  + β₅ (School Enrollment)_it + Z_iγ_i + u_it

Our models are shown in the equations above. In the fixed effects model, the term α_i represents country fixed effects, γ_t represents time fixed effects, and u_it is the error term. In the random effects model, the term Z_iγ_i assumes the country-specific effect for all entities is randomly distributed around a common mean value and that the common mean is constant across all time periods. We chose the functional form linear-linear, instead of a model that includes logged variables. This is because both CO2 emissions and GDP growth are relative linear estimations and do not require percentage adjustments. Before running the regression, we also determined the following hypothesis testing benchmarks. The null hypothesis is that carbon emissions have no effect on economic growth. The alternative hypothesis is that carbon emissions have an effect on economic growth. Based on the confidence level we choose, if the p-value is greater than 0.05, which is the critical value of a 95% confidence interval, then the null hypothesis fails to be rejected. Otherwise, we reject the null hypothesis, and conclude that our data is consistent with the alternative hypothesis.

We used the Hausman test to determine if a fixed effects or random effects model would be the best fit for our data. To run the Hausman test, we first ran both fixed effects and random effects regressions. If the test returned a significant p-value (less than 0.05), we could reject the random effects model and use the fixed effects model. For the majority of our data, the Hausman test found that the fixed effects model was the best fit, although there were two regressions at the regional level where the random effects model was found to be the best fit. Fixed effects in regression models account for individual-specific characteristics by including dummy variables for each entity or group, treating them as constants, or by adding a single variable to capture all of the individual-specific characteristics, rather than capturing them separately for each entity. On the other hand, random effects models treat individual-specific effects as random variables, allowing them to vary across entities and capturing unobserved heterogeneity. 

In addition to looking at global impacts of carbon emissions on economic growth, we investigate this impact by income level and region to identify clear relationships between CO2 emissions and GDP growth. Thus, we separated our results into eleven different regressions. This included regressions for four different income levels and seven different regions. First, the four categories of income analyzed were high income, upper middle income, lower middle income, and low income. The categorization was based on the World Bank’s income group that bases these measures on Gross National Income (GNI) from previous years, after taking into account inflation by using a Special Drawing Rights (SDR) deflator. The World Bank uses GNI as a cutoff, as opposed to GDP, due to GNI’s inclusion of factor incomes earned by its foreign residents and its subtraction of income earned by nonresidents. Thus, in order to be qualified as a high income country in 2023, one country’s GNI must exceed $13,205 and for a country to qualify as low income, their GNI must be below $1,085. Second, the seven regions analyzed were Latin America & Caribbean, South Asia, Sub-Saharan Africa, Europe & Central Asia, Middle East & North Africa, East Asia & Pacific, and North America. In doing these regressions for smaller, more similar groups of countries, we hope to uncover the impact wealth and geopolitics play in our regression, with the aim of identifying clear trends between these two variables and how they manifest themselves in different parts of the world. With this knowledge in hand, policymakers in developing countries who are keen on implementing decarbonization strategies can draw from the lessons of wealthier, culturally-similar countries. 

V. Results

World Results

As illustrated in Table 2, the results on global data found that carbon emissions were associated with GDP per capita growth across our sample of 129 countries, such that as carbon emissions increased, GDP growth increased. This result is consistent with our alternative hypothesis. We find that when carbon emissions increased by 1 metric ton per capita, GDP growth increased by 0.247 percentage points. All of the variables included in our regression had a statistically significant effect on GDP growth at the 99% confidence level. 

Income-Level Results: The relationship between CO2 emissions and GDP growth by country income level

After examining the global effect of carbon emissions on GDP growth, we ran regressions for four income levels: low income, lower middle income, upper middle income, and high income. As Table 3 shows, in low income countries, a 1 metric ton per capita increase in carbon emissions led to a 13.81 percentage point decrease in GDP growth, significant at the 95% confidence level and strikingly different from the result for global data. In high income countries, a 1 metric ton per capita increase in carbon emissions led to a 0.321 percentage point increase in GDP growth, significant at the 99% confidence level and similar to the result for global data. In lower middle income countries and upper middle income countries, carbon emissions were found to have no significant effect on GDP growth, consistent with the null hypothesis. After running these regressions, it is clear that countries in each of the four income levels are drastically different from one another, confirming our belief that GDP growth is expressed differently in countries with different income levels. While carbon emissions have no effect on economic growth in lower or upper middle income countries, they have a large negative effect on economic growth in low income countries and a small positive effect on economic growth in high income countries. It is important to note that the number of countries in the low income category is much smaller than the number of countries in the high income group, due to limited data availability. However, these contrasting results beg us to consider why carbon emissions harm GDP growth so much in low income countries yet appear to facilitate economic growth in high income countries. The results also push us to consider why middle income countries appear unaffected by carbon emissions and to discover the channels by which these countries are insulated from carbon emission shocks.

Regional Results: The relationship between CO2 emissions and GDP growth by region

We also conducted regressions for seven geographic regions: East Asia & Pacific, Europe & Central Asia, Latin America & Caribbean, Middle East & North Africa, North America, South Asia, and Sub-Saharan Africa. It is worth noting that only two countries were included in North America and only four countries were included in South Asia due to the small size of these regions and limited data availability. As shown in Table 4, regression results found that carbon emissions had no significant effect on GDP growth for all regions except Europe & Central Asia. In the Europe & Central Asia region, a 1 metric ton per capita increase in carbon emissions led to a 0.362 percentage point increase in GDP growth. The results question us to think further about the validity of the data as well as what factors may be in play that stabilize regions of the world to carbon emissions output. It also makes us question if, regionally, carbon emissions play any role in impacting GDP growth.

VI. Discussion

Overall, the differences in the relationship between CO2 emissions and GDP growth across regional, income-level, and global (aggregate) data illustrate the complexity of the relationship between carbon emissions and economic growth and help explain why different studies in the past have come across different results. As the regression results for the world as a whole (and specifically noted in high-income countries and Europe & Central Asia) found a positive effect of carbon emissions on economic growth, it will be important for policy-makers to account for potential economic harm when making future plans for decarbonization. 

In contrast, our results suggest that within low income countries, CO2 emissions had a negative effect on GDP growth. The reasons for this surprising finding are not clear. It should be noted that among low income countries, the rate of CO2 emissions was extremely low and restricted in range (0.022 to 0.55 metric tons per capita) in comparison to high income countries (1.36 to 47.66 metric tons per capita). There was also significant variability in GDP growth among low income countries (ranging from -14.96% to 27.83%, in comparison to high income countries ranging from -14.46% to 18.91%). The combination of the limited range of carbon emissions and large range of GDP growth among low income countries may have contributed to this paradoxical result. Further research with larger sample sizes will be needed to determine whether this finding is reproducible.

Overall, we recommend the implementation of carbon taxation and emission trading systems to efficiently reduce carbon emissions while preserving economic growth, a strategy that may be particularly important for high-income countries. This approach not only incentivizes reductions in carbon emissions but also generates revenue that can be strategically reinvested in sustainable practices and economic growth initiatives. Unlike command and control mechanisms, these are known to limit economic harm and work within the free market. The policy should be designed to minimize economic disruption, particularly in vulnerable sectors, ensuring a balanced approach between environmental sustainability and economic growth. Successful systems of tradable permits, like the United States sulfur trading market, can bring about positive environmental changes. Throughout this process of decarbonization, international cooperation is key. Aiding developing countries to transition toward greener practices is especially important. These economies often face the dual challenge of reducing emissions while sustaining growth. The international community must provide financial aid and technological support to ensure equitable and effective decarbonization. This combination of internal policy shifts and external support structures offers a comprehensive strategy for achieving sustainable development goals while considering economic implications.

Moreover, there are a number of experimental limitations we faced in our work. First, reverse causality could pose significant challenges to the usefulness and applicability of the research. Reverse causality refers to a situation where a cause-and-effect relationship between two variables is incorrectly identified because the assumed cause is actually the effect. Sometimes, like in this situation, causes and effects work in both directions, leading to additional complexities, including endogeneity and threats to internal validity. If the true causal direction is not deciphered, then the study’s results may be unreliable. This false correlation can lead to biased parameter estimates that would make it difficult to isolate the true causal impact. In investigating the correlation between a country’s pollution levels and its economic development, the risk of reverse causality arises when economic growth influences carbon emissions rather than the other way around. For example, a country experiencing rapid economic growth might simultaneously witness an increase in industrial activities and aggregate energy consumption. This would result in higher carbon emissions. In this scenario, it would be wrong to wholly attribute the environmental impact solely to economic growth, given the presence of other positive feedback loops regarding emissions. 

To address reverse causality, we employed several research and analysis techniques that prevent the worst of its issues. Firstly, we used detailed panel data that includes many countries over a twenty-year time period. This can help establish the direction of causality by examining how changes in carbon emissions precede or follow changes in economic growth over time. Secondly, we added additional variables to our regression equation in the hopes of isolating the effect of emissions on growth. These extra variables are carefully designed to account for many other factors that contribute to economic growth. Finally, we added country-fixed and time-fixed effects to our regression. These fixed effects will control all things that vary over countries but not over time and all things that vary over time but not countries. Fixed effects will also help lessen the negative impacts of omitted variable bias. Overall, we believe that our regression equation sufficiently accounts for reverse causality and allows us to draw strong and statistically sound conclusions from the research.

A second possible limitation of our research is missing data. As with any research, it is important to have a complete and sufficiently-sized dataset to perform exemplary analyses. Unfortunately, there is a lot of missing data for countries across the world during our period of study. To allow our regression analyses to run, we chose to include only countries with minimal missing data as well as control variables that had mostly complete data. However, the extent of missing data may affect generalizability of findings. 

Future research directions should explore the emerging dynamics between green technologies and environmental structures. As the full decarbonization process has yet to begin, further research is necessary to confirm or refute the findings of this paper. A key area of interest is the economic impact of transitioning to renewable energy sources in various geopolitical contexts. As different nations take advantage of unique strategies, it will be useful to see which green initiatives do or do not cause economic harm. The application of advanced econometric models to forecast long-term economic impacts of environmental policies would also be valuable, offering insights into sustainable economic planning. Predictive modeling is outside the scope of this paper, but would be useful to politicians, academics, and industry leaders alike.

VII. Conclusion

In summary, our extensive analysis spanning 129 countries over two decades reveals a significant and positive relationship between carbon emissions and GDP growth. Globally, as carbon emissions increased by 1 metric ton per capita, GDP growth rose by 0.247 percentage points. However, our study identifies nuances within the four income levels and seven regions. With rising carbon emissions, low-income countries experienced a negative impact on GDP growth, while high-income nations exhibited a positive impact. Regionally, carbon emissions significantly influenced GDP growth only in Europe & Central Asia. All these findings carry crucial implications for policymakers grappling with the dual challenge of climate change mitigation and economic development. Therefore, to address these complexities, both tailored policy approaches and international cooperation are essential. As a result, we recommend the adoption of carbon taxation and emission trading systems, both of which foster emissions reductions while generating revenue under the free market. Despite these conclusions, our study also acknowledges limitations such as potential reverse causality and data gaps, particularly in smaller economies. 

The transition to renewable energy has only truly started in the past decade and it will ramp up considerably in the future. This looming reality will continue to push researchers to examine and draw conclusions between these complex variables. Therefore, as the world barrels towards decarbonization and increased renewable energy use, it is worth noting that there is, in fact, a positive relationship between carbon emissions and economic growth, and the policy implications that result from this delicate relationship need to be exercised with the prudence and foresight of environmentally-conscious leaders. 

References

Akalpler, Ergin & Hove, Simbarashe (2019). Carbon Emissions, Energy Use, Real GDP per capita and Trade Matrix in the Indian Economy – an ARDL Approach, Energy, 168:1081-1093, https://doi.org/10.1016/j.energy.2018.12.012.

Barassi, M., & Spagnolo, N. (2012, July). Linear and non-linear causality between CO2 emissions and economic growth. The Energy Journal 33(3). DOI:10.5547/01956574.33.3.2.

Dasgupta, S., Laplante, B., Wang, H., & Wheeler, D. (2002). Confronting the Environmental Kuznets Curve. Journal of Economic Perspectives, 16(1): 147-168, https://www.aeaweb.org/articles?id=10.1257%2F0895330027157. 

Farhani, S., & Ozturk, I. (2015, May 29). Causal relationship between CO2 emissions, real GDP, energy consumption, financial development, trade openness, and urbanization in Tunisia. Environmental Science and Pollution Research, 22:15663-15676. https://doi.org/10.1007/s11356-015-4767-1. 

Holechek, J. L., Geli, H. M. E., Sawalhah, M. N., & Valdez, R. (2022, April 16). A global assessment: Can renewable energy replace fossil fuels by 2050? Sustainability, 14(8): 4792. https://doi.org/10.3390/su14084792. 

International Labour Organization (2022, September 22). Renewable energy jobs hit 12.7 million globally [Press release]. https://www.ilo.org/global/about-the-ilo/newsroom/news/WCMS_856515/lang–en/index.htm.

International Monetary Fund (2022, October). Chapter 3: Near-term Macroeconomic Impact of Decarbonization Policies. https://www.imf.org/-/media/Files/Publications/WEO/2022/October/English/ch3.ashx. 

Olubusoye, Olusanya Elisa & Musa, Dasauki (2020). Carbon Emissions and Economic Growth in Africa: Are They Related?, Cogent Economics & Finance, 8:1, DOI: 10.1080/23322039.2020.1850400.

Ozturk, I., & Acaravci, A. (2010, December). On the relationship between energy consumption, CO2 emissions and economic growth in Europe. Energy, 35(12): 5412-5420. https://doi.org/10.1016/j.energy.2010.07.009. 

Philip, P., & Ibrahim, C. (2022, June 20). The turning point. Deloitte. https://www.deloitte.com/global/en/issues/climate/global-turning-point.html. 

Pee, A. de, Pinner, D., Roelofsen, O., Somers, K., Speelman, E., & Witteveen, M. (2018, July 13). How industry can move toward a low-carbon future. McKinsey & Company. https://www.mckinsey.com/capabilities/sustainability/our-insights/how-industry-can-move-toward-a-low-carbon-future. 

The World Bank, World Development Indicators (2022). CO2 emissions (metric tons per capita) [Data file]. Retrieved from https://data.worldbank.org/indicator/EN.ATM.CO2E.PC. 

The World Bank, World Development Indicators (2022). GDP per capita growth (annual %) [Data file]. Retrieved from https://data.worldbank.org/indicator/NY.GDP.PCAP.KD.ZG. 

The World Bank, World Development Indicators (2022). Population growth (annual %) [Data file]. Retrieved from https://data.worldbank.org/indicator/SP.POP.GROW. 

The World Bank, World Development Indicators (2022). School enrollment, primary (% gross) [Data file]. Retrieved from https://data.worldbank.org/indicator/SE.PRM.ENRR.

The World Bank, World Development Indicators (2022). Total natural resources rents (% of GDP) [Data file]. Retrieved from https://data.worldbank.org/indicator/NY.GDP.TOTL.RT.ZS.

The World Bank, World Development Indicators (2022). Trade (% of GDP) [Data file]. Retrieved from https://data.worldbank.org/indicator/NE.TRD.GNFS.ZS. 

Trump, D. J. (2014, January 29). Snowing in Texas and Louisiana, record setting freezing temperatures throughout the country and beyond. global warming is an expensive hoax!. Twitter. https://twitter.com/realDonaldTrump/status/428414113463955457?ref_src=twsrc%5Etfw%7Ctwcamp%5Etweetembed%7Ctwterm%5E428414113463955457%7Ctwgr%5E720632dbda4db9d94aafb9149916272f0bdcbad0%7Ctwcon%5Es1_&ref_url=https%3A%2F%2Fwww.motherjones.com%2Fenvironment%2F2016%2F12%2Ftrump-climate-timeline%2F. 

Yang, Jun & Hao, Yun & Feng, Chao (2021). A Race Between Economic Growth and Carbon Emissions: What Play Important Roles towards Global Low-Carbon Development?, Energy Economics, 100: 105327, https://doi.org/10.1016/j.eneco.2021.105327.

Zhang, Xing-Ping & Cheng, Xiao-Mei (2009). Energy Consumption, Carbon Emissions, and Economic Growth in China, Ecological Economics, 68:10:2706-2712, https://doi.org/10.1016/j.ecolecon.2009.05.011.

Zhang, S., Yang, F., Liu, C., Chen, X., Tan, X., Zhou, Y., Guo, F., & Jiang, W. (2020, February 13). Study on Global Industrialization and Industry Emission to Achieve the 2 °C Goal Based on MESSAGE Model and LMDI Approach. Energies, 13(4): 825. https://doi.org/10.3390/en13040825.