1. Introduction
In the following article, we analyzed the value of renewable energy in the context of the circular economy considering the need to create greater sustainability and resilience of global energy systems. The dataset used for the analysis is the “Environmental, Social and Governance-ESG” of the World Bank [
1]. However, as part of the large dataset, the variables relating to energy production in 193 countries for the period 2011–2020 were mainly analyzed. The circular economy paradigm requires attention to all phases of production and consumption in all countries globally. However, notoriously, the possibility of triggering economic growth processes, economic development, and the achievement of potential productivity, in all countries, requires energy consumption. In fact, the consumption of energy worldwide always tends to grow despite crises and regardless of the application of technological innovation to energy production methods.
However, the need to grow in income could push countries to use highly polluting energy sources such as non-renewable carbon-based energy sources. In this sense, the question of the energy sustainability of economic growth becomes an objective of economic policy both nationally and internationally. In fact, governments are in the need to guarantee, on the one hand, growth and prosperity to their nation, and on the other hand, they must respect the international cooperation agreements that are oriented to the reduction of CO
2 emissions to reduce overheating globally. In this sense, the so-called Paris Agreements [
2] and the Conference of Parties 27-COP27 [
3] recently held in Egypt, try to impose ambitious environmental policy at country level to combat climate change.
One of the methods that can increase the propensity to the circular economy in the context of the sustainability of economic growth processes is renewable energy. In fact, renewable energy tends to be positively associated with the reduction of CO2 emissions and the reduction of temperature. Furthermore, since the renewable energy production methodologies require the use of a sort of production, decentralization also derives the possibility of offering energy at reasonable prices even in regional, local, and rural areas that in the past would not have had the opportunity to participate in the energy production.
However, the use of renewable energy is certainly not free from criticism or economic and technological limitations. In fact, as reported in paragraph two relating to the analysis of literature, there are empirical studies that have shown how economic growth is more compatible with the use of non-renewable energy sources rather than renewable energy sources. This dynamic can be understood because for a country it is certainly easier and faster to grow in GDP through pollution especially if it is a developing country that intends to fit into the ranks of medium and high-income countries.
However, these methodologies of economic growth are very risky as they generate pollution and relevant negative externalities at the environmental level that weigh both on the national population and on the global energy balance. It, therefore, derives the need to find new solutions, and new economic policies, which keep together economic growth, respect for the environment, need to grow energy consumption, with the reduction of CO2 emissions. One of the methodologies to achieve compatibility between the sustainability of the circular economy and economic growth consists of the growth of renewable energy production. Renewable energy can also have further positive effects at the country level by changing the mentality either of producers or consumers relating to the impact of productive and consumption choices on the environment. This change of mentality and public sensitivity can have an impact especially in developing countries and in new industrialization countries to facilitate an overall passage to the circular economy, operating as a cultural and value reform supported by the positive experience relating to the production and renewable energy consumption. Although there are certainly models of economic growth that have hypothesized the total passage also of medium-high income economies to the circular economy through the exclusive use of renewable energy, it is known that renewable energy still has limitations in the ability to be a useful tool to support, alone, the productivity of an industrialized country.
However, it is very probable that new technologies that can effectively use the forces of nature to generate energy without the emission of CO
2, in a compatible way for the environment are very probable through the investment of research and development. R&D can promote new technology to support renewable energy production for cities, the manufacturing industry, as well as means of transport [
4].
The development of renewable energy, therefore, becomes the essential element to guarantee the sustainability of the economic growth of countries and to allow a full transition to the circular economy.
The article continues as follows: In the second paragraph, a brief analysis of the literature is presented, which analyzes the relationship between the use of renewable energy and the sustainability of economic growth. In the third paragraph, an econometric model is presented for the esteem of the determinants of the production of renewable energy. In the fourth paragraph, the clusterization with the k-Means algorithm optimized with the Elbow Method to investigate the presence of groupings among the 193 countries analyzed in terms of renewable energy production. The fifth paragraph presents a prediction with algorithms of Machine Learning-ML to identify a path of possible development of renewable energy production, the sixth paragraph contains a Deep Learning Analysis with Convolutional Neural Network-CNN, and the seventh paragraph concludes.
2. Literature Review
A brief analysis of the literature is presented below, which, above all, takes into consideration the relationship between the production of renewable energy and the sustainability of economic growth processes.
Ref. [
5] verify the existence of a positive relationship between the value of financial inclusion, globalization, the reduction of carbon dioxide emissions, and the presence of energy from renewable sources. Ref. [
6] establishes a positive relationship between economic development and renewable electricity consumption in the Baltic region using data for the period 1992–2011. Ref. [
7] analyzes the presence of a causal link between renewable energy, non-renewable energy, and economic growth for 26 OECD countries. The author concludes by arguing that economic growth policies should encourage renewable energy to facilitate GDP growth and increase energy independence at the country level. Ref. [
8] consider the relationship between renewable energy production and Gross Domestic Product-GDP in 36 OECD countries over the period 2001–2015. The authors verify the existence of a negative relationship between GDP growth and an increase in the production of renewable energy. This report suggests that economic growth and development of renewables do not necessarily go hand in hand, especially in countries with medium-high per capita income.
Ref. [
9] investigate the existing relationships between CO
2 emissions, renewable electricity consumption, non-renewable electricity consumption, and economic growth in Algeria in the period between 1980 and 2012. The results show that in the long run, economic growth and the consumption of energy that is not renewable have a negative effect on the quality of the environment, while the use of renewable energy positively impacts the quality of the environment. Energy consumption is produced by economic growth. However, to improve the environmental condition in Algeria, it is necessary to invest in renewable energy by generating positive effects also in terms of reducing CO
2 emissions. Ref. [
10] analyze the relationship between renewable energy consumption, non-renewable energy consumption, economic growth, and CO
2 emissions in Mediterranean countries in the period between 1980 and 2014. The authors verify the presence of a negative relationship between the use of renewable energy, CO
2 emissions and economic growth. It follows that if on the one hand, the use of renewable energy can reduce CO
2 emissions, on the other hand it is not certain that it is compatible with a program of economic growth in the short term.
Ref. [
11] analyze the relationship between renewable energy production, GDP growth, and CO
2 emissions in Pakistan. The results show that in the long run CO
2 emissions, renewable energy production, and GDP growth increase together. The authors suggest that economic policies be applied to increase the production of renewable electricity and reduce CO
2 emissions in Pakistan. Ref. [
12] consider the positive impact that the ecological transition and the switch to a fully renewable energy production system could have on the Dutch economy. The authors estimate that the transition to renewable energy could create 50,000 new jobs by 2030 and increase Gross Domestic Product by 1%. Ref. [
13] confirm the presence of a positive relationship between both renewable and non-renewable energy production in terms of economic growth for 174 countries between 1980 and 2012.
Ref. [
14] underline the need to strengthen and facilitate the transition to fully sustainable and renewable energy-based production energy systems globally. The energy transition to a system completely free from carbon sources is considered a necessary objective to respect the limits of growth of global temperatures established in international agreements and treaties. Ref. [
15] analyzes the positive relationship between the production of renewable energy and the reduction of CO
2 emissions in South America over the period 1980–2010. Ref. [
16] verify the presence of a positive relationship between renewable energy production, CO
2 reduction and gross domestic product growth in a panel of 84 countries between 1991 and 2012. Ref. [
17] consider the perverse effects that the political economic incentive for renewable energy installations has had on households in Germany. In fact, federal government subsidies for renewable energy production have pushed up prices. Poor households have been faced with higher energy prices. The authors suggest solutions to combine the growth in the price of renewable energy with the need to offer subsidies to households.
Ref. [
18] considers the impact of the energy transition to renewable energy from a regional perspective. The method of investigation is bibliographic. The analysis starts from the idea that since renewable energies have a degree of reduced energy density, a significant change in the areas in which renewable energy is produced above all, at a regional, local, and rural level, is derived. Effects in terms of employment, supply chains, and value are thus detectable and are achieved at the regional and local levels. Ref. [
19] analyze the impact of renewable energy in terms of economic growth in 38 countries in the period between 1990 and 2018. The results show the presence of a positive relationship between renewable energy production and economic growth. The authors suggest implementing actions at the international, governmental, and intergovernmental levels to increase the production and consumption of renewable energy.
Ref. [
20] show the presence of a positive relationship between the usage of renewable energy and the ability to apply the principles of the circular economy. A circular economy promotes renewable energy consumption and production [
21]. Ref. [
22] shows how biomass energy if a tool to enforce circular economy principles [
23,
24]. Ref. [
25] consider the role of renewable energy consumption and production in promoting a circular economy in the context of the Georgescu–Roegen theoretical approach [
26]. The combination between circular economy and renewable energy is a strategic tool to achieve the Sustainable Development Goals-SDG [
27]. An efficient technological design of renewable energies is essential to provide sustainable energy to communities oriented to a circular economy [
28]. Nevertheless, circular economy principles should inspire renewable energy, too. In this sense, it is necessary to calibrate the relationship between the extraction rate of renewable energy and the regeneration rate of renewable resources [
29]. Ref. [
30] promote bio-hydrogen as a new renewable energy source able to boost the passage from the traditional economic system to the circular economy model. Ref. [
31] suggest the application of wastewater treatment and technology to implement the energy transition toward a circular economy. Ref. [
32] afford the question of the shift toward renewable energy in a country like Russia, which is an exporter of oil and other pollutant natural resources. Renewable energy operates as an instrument to improve the transition toward a circular economy in China [
33] reducing emissions in the short and long run [
34].
There is skepticism about the possibility to apply a circular economy, even with renewable energies, in developing countries such as Bangladesh [
35]. The negative relationship between the price of non-renewable energy and renewable energy has a relevant role in promoting the energetic transition of China toward a circular economy [
36]. Renewable energy induces a reduction in CO
2 emission even in more advanced countries in terms of complexity index suggesting the existence of a positive relationship among circular economy, environmental sustainability, and knowledge-oriented economic systems [
37].
Countries that have a greater level of renewable energy production suffer less from energy vulnerability [
38]. Policymakers play an essential role in creating the institutional framework and the economic incentives to enforce the passage from a traditional and pollutant economy to a circular and more sustainable economy through the promotion of renewable energy [
39]. New forms of renewable energy can boost the passage toward a more circular economy such as the bioconversion of biomass into biofuels and bio commodities [
40]. Many OECD countries, such as Ireland, Luxembourg, and Norway, are investing in renewable energy and in the reduction of municipal waste and CO
2 emissions to promote the shift toward sustainability and circular economy [
41]. Ref. [
42] find the presence of a positive association among renewable energy output, GDP growth, and the reduction of CO
2 emissions in Switzerland, a case that confirm the validity of the Environmental Kuznets Curve-EKC. Renewable energy is positively associated with economic development and the transition toward a more sustainable economy in Switzerland [
43]. The development of the renewable energy sector, especially the bioenergy industry, requires an increase in governance indicators especially in Western Europe [
44].
The possibility to accomplish a Renewable Energy Transition-RET depends on a set of variables that are strongly related to the social, economic, institutional, and human capital at the country level and that requires a more conscious and green-oriented political economy [
45]. Investment in Research and Development can promote the transition toward a renewable energy output system [
46]. Even high-pollutant countries, such as China, have targeted carbon neutrality thanks to renewable energy sources [
47,
48]. However, the energy transition of China toward sustainability and a circular economy not only implies the change from non-renewable to renewable energy, but it also requires the change from a centralized to a de-centralized form of energy production [
49]. Renewable energy can foster economic growth and financial development even in low-per-capita-income countries [
50]. There is a positive relationship among renewable energy consumption, economic growth, and financial development [
51]. The ability of renewable energy output to sustain the energy demand of advanced countries can be estimated with appropriate probabilistic functions and algorithms [
52].
To promote energy transition it is necessary to respect cultural and institutional values at the country level. For example, the application of an Anglo-Saxon definition of energy transition for the Chinese economy is not efficient while the orientation toward a “guanxi” method could generate better effects [
53]. The negative relationship between economic growth and sustainability persists even in Europe, despite the European green political economy that has financed renewable energy production and has created the conditions for the application of the circular economy [
54]. The shift toward a circular economy in an ecomodernist political environment also requires financial incentives to improve the profitability of private investments [
55]. The empirical analysis of the relationship among economic growth, renewable energy production, and urbanization shows that EKC holds for China [
56]. The energy transition requires also political economies able to change the structure of incentives and can start with a price guarantee for renewable energy providers in order to sustain initial investments in the sector, as in the case of Germany [
57]. The development of a more efficient renewable energy sector at the country level also requires investment in smart grids [
58,
59]. However, the intermittence of renewable energy production can produce inefficiency in the grid at the national and local levels [
60]. Different Energy Storage Technologies-ESTs can solve the question of fluctuations in the production of renewable energy, such as Li-ion batteries, even if the high cost of Li-ion batteries impedes their widespread application [
61]. Furthermore, energy storage improves the usage of renewable energy output and consumption [
62]. Another solution to promote a better energy transition is to invest in micro-grid based [
63] on multi-energy sources [
64]. In any case, the risk of intermittence of renewable energy output should be computed adequately in energy planning even at the local level [
65]. Smart grid based on multi-energy sources can either reduce the level of either reduce the cost of energy [
66]. The increase in renewable energy output and consumption requires a more flexible power system [
67].
The reduction in total energy consumption is essential to connect renewable energy output to economic growth [
68]. The price of wind power has a negative impact on wind power generation, while the local demand can strengthen wind energy production [
69]. However, in countries with low per capita income, i.e., Eastern African countries, there is a negative relationship between economic growth and non-renewable energy production [
70]. Furthermore, the shift to a 100% renewable energy system can be too costly to implement [
71]. The Integrated Energy System-IES can solve the question of uncertainty in the production and pricing of renewable energy [
72]. The substitution of non-renewable energy with renewable energy consumption can happen only in the long run, even in the case of an increase in the price of oil, coal, and natural gas [
73]. A probabilistic approach can solve the question of renewable energy output variability and the issues related to the stability of the distribution system [
74].
As evident, the analysis of the literature tends to confirm the presence of a positive relationship between the growth of renewable energy production and the reduction of CO
2 emissions. However, the relationship between the growth of the production of renewable energy and growth of the gross domestic product is ambiguous, at least in the long run. It should be considered that almost all the articles analyzed apply analytical methodologies aimed at investigating the presence of causal relationships between the variables. Furthermore, the ambiguity of the metric results about the relationship between renewable energy production and economic growth rate depends on the fact that the analyzed data fail to incorporate that the implementation of the new economic policies arranged internationally to combat climate change is likely to be incorrectly and reduce CO
2 emissions.
Table 1 synthesizes the references by main themes.
4. Clusterization with the k-Means Algorithm Optimized with the Elbow Method
In the following analysis, a method is applied using the k-Means algorithm optimized with the Elbow Method (
Figure 1). The application of a method of clusterization is necessary for a database consisting of 193 countries characterized by enormous heterogeneity from an economic, social, institutional, and geographical point of view. In this way, clusterization allows the presence of groupings that can be due to a composite set of environmental economic policy choices that put together economic, geographical, financial, and technological elements. Generally, the Silhouette coefficient is used to optimize the K-means algorithm. However, due to the excessive polarization of the data, the use of the silhouette coefficient appeared excessively simplifying, manifestation of the existence of only two Clusters in a context of 193 countries. This analysis initially attempted would have simply represented a contrast between two blocks of countries, namely rich countries and poor countries. To grasp the greatest heterogeneity in the dataset, it was chosen to use the Elbow method that highlights four clusters (
Figure 2). Specifically, the following clusters were identified, namely:
Cluster 1: Angola, Belize, Cambodia, Canada, Congo Rep., Croatia, Denmark, Ecuador, El Salvador, Eswatini, Fiji, Gabon, Ghana, Guatemala, Guinea, Honduras, Latvia, Madagascar, Mali, Montenegro, Myanmar, Nicaragua, Panama, Peru, Portugal, Rwanda, Sierra Leone, Sri Lanka, Suriname, Sweden, Switzerland, Venezuela, Vietnam, Zimbabwe;
Cluster 2: Algeria, Antigua and Barbuda, Australia, Azerbaijan, Bahamas, Bahrain. Bangladesh, Barbados, Belarus, Benin, Botswana, Brunei Darussalam, Bulgaria, Burkina Faso, Chad, Comoros, Cuba, Cyprus, Czech Republic, Djibouti, Dominican Republic, Egypt, Eritrea, Estonia, Finland, Gambia, Grenada, Guinea-Bissau, Guyana, Haiti, Hungary, Indonesia, Iran, Iraq, Israel, Jamaica, Japan, Jordan, Kazakhstan, Kiribati, Korea Rep., Kuwait, Lebanon, Liberia, Libya, Malaysia, Maldives, Malta, Marshall Islands, Mauritania, Micronesia, Moldova, Monaco, Mongolia, Morocco, Nauru, Netherlands, Niger, Oman, Palau, Poland, Qatar, San Marino, Sao Tome and Principe, Saudi Arabia, Senegal, Seychelles, Singapore, Solomon Islands, Somalia, South Africa, South Sudan, St. Kittis and Nevis, St. Lucia, Syrian Arab Republic, Thailand, Timor-Leste, Tonga, Trinidad and Tobago, Tunisia, Turkmenistan, Tuvalu, Ukraine, United Arab Emirates, United States, Yemen Rep;
Cluster 3: Afghanistan, Albania, Andorra, Austria, Bhutan, Brazil, Burundi, Cameroon, Central African Republic, Colombia, Congo Dem. Rep., Ethiopia, Georgia, Iceland, Kenya, Kyrgyz Republic, Lao PDR, Lesotho, Liechtenstein, Malawi, Mozambique, Namibia, Nepal, New Zealand, Norway, Paraguay, Sudan, Tajikistan, Togo, Uganda, Uruguay, Zambia;
Cluster 4: Argentina, Armenia, Belgium, Bolivia, Bosnia and Herzegovina, Cabo Verde, Chile, China, Cote d’Ivoire, Dominica, Equatorial Guinea, France, Germany, Greece, India, Ireland, Italy, Korea, Dem. People’s Rep., Lithuania, Luxembourg, Mauritius, Mexico, Nigeria, North Macedonia, Pakistan, Papua New Guinea, Philippines, Romania, Russian Federation, Samoa, Serbia, Slovak Republic, Slovenia, Spain, St. Vincent and the Grenadines, Tanzania, Turkey, United Kingdom, Uzbekistan, Vanuatu.
Considering the value of the median of the clusters, it results that Cluster 3-C3 is equal to an amount of 91.63, while Cluster-C1 is equal to an amount of 54.64 units, Cluster-C4 is equal to an amount of 25.41 units and Cluster 2-C2 equal to an amount of 1.47. The following ordering of the clusters is, therefore, derived: C3 > C1 > C4 > C2. From a geographical point of view, we can see that it is above all the countries with low per capita incomes that have the greatest percentage of electricity production from renewable sources. However, this condition can be better explained by considering that countries that have low per capita incomes are in the following situation:
have probably not invested sufficiently in traditional forms of energy production due to lack of infrastructure;
being African and Asian countries, they are in the climatic conditions to optimize the advantages of renewable energies.
However, there are of course exceptions. For example, in Cluster 3-C3 which is the one that has the greatest value in terms of presence of renewable energy production, there are also Liechtenstein, Norway, Austria, New Zealand, and Iceland. It is obviously a presence that needs to be explored. In the case of Austria, most of the renewable energy is produced by hydroelectricity, followed by solar, biomass, solar and geothermal. Also, in Norway, the main source of energy is hydroelectricity, followed by solar and thermal energy. Finally, Iceland manages to produce geothermal and wind energy. However, we must consider that Austria, Norway, Iceland are small countries, and their energetic demand is smaller with respect to other more populous high-income countries.
It is also relevant to afford the question of the relationship between the demand for CO2 in the industrial sector in OECD countries and the percentage of renewable energy output as a percentage of the total energy produced. We have analyzed the relationship between CO2 emissions in the industrial sector and renewable energy output for OECD countries. Specifically, first of all, a cluster analysis is performed with the application of the k-Means algorithm optimized with the Elbow Method to verify the presence of clusters in terms of CO2 emissions in the industry sector in OECD countries. Subsequently, for each cluster, a comparative analysis was carried out with the level of Renewable Energy Output-REO to verify the presence of a general positive or negative trend through a graphical representation method.
In particular, the variable used to approximate the “CO
2 emissions embodied in domestic final demand, by source country and industry”. The variable is calculated in millions of tons. A clustering with k-Means algorithm is therefore carried out by applying the Elbow method. The analysis shows the presence of 4 clusters (
Figure 3).
The clusters are composed as follows:
Cluster 1: Taipei, Thailand, Netherlands, Argentina, Malaysia, Kazakhstan, Belgium, Greece, Vietnam, Czech Republic, Malta, Iceland, Philippines, Brunei Darussalam, Switzerland, Cambodia, Cyprus, Luxembourg, Hong Kong, Latvia, Costa Rica, Austria, Israel, Romania, Estonia, Lithuania, Slovenia, Colombia, Chile, Croatia, Tunisia, Sweden, Denmark, Slovakia, Peru, Bulgaria, Portugal, Singapore, New Zealand, Ireland, Finland, Hungary, Morocco, Norway,
Cluster 2: Japan, Russia, India, Germany;
Cluster 3: Poland, South Africa, Turkey, Spain, United Kingdom, Indonesia, Saudi Arabia, Brazil, Italy, South Korea, Canada, Australia, France, Mexico;
Cluster 4: China, USA.
The value in millions of tons, specifically it is possible to identify the following ordering of the clusters, namely: C4 = 6886.2 > C2 = 1264.25 > C3 = 460.2 > C1 = 61.6. It therefore follows that from the point of view of the emissions produced in the industrial sector, China and the USA have very high values of eliminate pollution from industry, i.e., a value equal to about 40% among the countries considered in the OECD dataset.
As can be seen from the analysis, it appears that the relationship between the percentage of renewable energy production and CO
2 emissions in the industrial sector, with reference to 2015, is negative for the first and second clusters (
Figure 4 and
Figure 5), tends to be zero for the third cluster (
Figure 6), and is positive for the fourth cluster (
Figure 7). Probably the countries that have a greater balance between the level of CO
2 emissions from the industrial sector and the level of renewable energy output are the countries of cluster 3. It follows that between renewable energy output andCO
2 emissions from industrialization, it is not necessarily a negative relationship. There are some countries, such as for example the countries of cluster 3, which manage to invest in the production of renewable energy despite the presence of CO
2 emissions from the industrial sector.
The ability of countries to invest in renewable energy output can be considered as a function of a series of socio-economic factors that comprehend also demographic and cultural elements. However, overall, the investment in renewable energy output at the global level is the consequence of the development of a green conscious and the implementation of green political economies. In this sense, it is not sufficient to consider the economic consequences and incentives for the implementation of renewable energy sources. In effect, the promotion of political movements that are centered on the fight against climate change has a positive impact on the improvement of renewable energy output at the global level. The idea of environmental sustainability and circular economy together with the application of Cost-Benefit Analysis and worst-case scenario oriented to visualize and compute the negative and adverse consequences of climate change are powerful and useful tools to promote renewable energy outputs even in high developed and pollutant countries.
5. Machine Learning and Predictions
Below is a comparison between seven different Machine Learning algorithms for the prediction of the future value of renewable energy production. The algorithms were evaluated based on their statistical capacity measured based on the maximization of the R-Squared and the minimization of the following statistical errors or: Mean Absolute Error-MAE, Mean Squared Error-MSE, Root Mean Squad Error-RMSE.
Table 4 shows the statistical results.
The algorithms were trained with 70% of the data while the remaining 30% were used for the actual prediction. The following order of Machine Learning algorithms for predictive performance was therefore obtained or:
Linear Regression with a payoff equal to 4;
Polynomial Regression with a payoff equal to 7;
Gradient Boosted Trees Regression with a payoff value of 11;
Tree Regression Ensemble with a payoff value of 12;
Simple Regression Tree with a payoff value of 17;
ANN-Artificial Neural Network with a payoff value of 21;
PNN-Probabilistic Neural Network with a value of 25.
Table 5 contains the ranking of machine learning algorithms based on predictive performance.
Therefore, using the best performing algorithm or Linear Regression (
Figure 8 and
Figure 9) it is possible to predict the following values for the countries indicated, i.e.:
Antigua and Barbuda with an increasing variation from an amount of 0.10 up to a value of 0.11 units or equal to 0.01 equivalent to 10.00%;
Armenia with a change of an amount of 28.20 units up to a value of 28.18 units or equal to a change of −0.02 units equal to −0.07%;
Bahrain and Barbados which in 2020 recorded 0.00% of renewable energy production remain in their condition without further investing in the growth of energy produced through renewables;
Belarus with a diminutive variation from an amount of 0.74 units up to a value of 0.72 units or a variation equal to −0.02 units equal to −3.303%;
Botswana with a change of an amount of 0.04 units up to a value of 0.07 units or equal to a value of 0.03 units equal to a value of 64.33%;
Brazil with a variation from an amount of 76.24 units up to a value of 76.21 units or equal to a variation of −0.03 units equal to a value of −0.04%;
Brunei Darussalam with a variation from an amount of 0.05 units up to a value of 0.06 units or equal to a value of 0.01 units or equal to a variation equal to a value of 28.00%;
Burkina Faso with a variation from an amount of 11.66 units up to a value of 11.65 units or equal to a value of −0.01 units equal to a value of −0.11%;
Burundi with a value of 85.53 units up to a value of 85.40 units equal to a value of −0.13 units equal to a value of −0.15%;
Cabo Verde with a variation from an amount of 18.48 units up to a value of 18.49 units or equal to a variation of 0.01 units equal to a value of 0.03%;
China with a variation from an amount of 22.01 units up to a value of 22.99 units or equal to a variation of 0.98 units equal to an amount of 4.45%;
Colombia with a variation from an amount of 72.98 units up to a value of 72.97 units or equal to a variation of −0.01 units equal to a variation of −0.01%;
Comoros with zero variation and a value of renewable energy production equal to 0;
Congo with a variation from an amount of 99.79 units up to a value of 99.76 units or equal to a value of −0.03 units equal to a value of −0.03%;
Czech with a zero change in absolute value and a marginal change in percentage value equal to 0.02 for a total value of renewable energy produced equal to 10.69%;
Dominica with a diminutive variation from an amount of 20.52 units up to a value of 20.49 units or equal to a variation of −0.03 units equal to a value of −0.16%;
Dominican Republic with a zero change in absolute value and an amount of renewable energy production equal to a value of 13.25%;
El Salvador with an increasing variation from an amount of 59.34 units up to a value of 59.35 units or equal to a value of 0.01 units equal to a variation of 0.02%;
Eritrea with a variation from an amount of 0.52 units up to a value of 5.53 units or equal to a variation of 5.01 units equal to an amount of 963.08%;
Eswatini with a variation from an amount of 46.93 units up to a value of 46.95 units or equal to a variation of 0.02 units equal to an amount of 0.04%;
Gabon with a variation from an amount of 42.93 units up to a value of 42.40 units or equal to an amount of −0.53 units in absolute value equivalent to −1.23%;
Georgia with an increasing variation from an amount of 78.78 units up to a value of 78.79 units or equal to a value of 0.01 units equal to a value of 0.01%;
Germany with a diminutive variation from an amount of 26.26 units up to a value of 26.25 units or equal to an amount of −0.01 units equal to an amount of −0.03%;
Greece with a variation from an amount of 24.54 units up to a value of 24.56 units or equal to an amount of 0.02 units equal to a value of 0.08%;
Haiti with a variation from an amount of 10.95 units up to a value of 10.96 units equal to an amount of 0.01 units equal to a value of 0.11%;
Honduras with zero variation and an amount of renewable energy production equal to a value of 44.59 units;
Italy with a variation from an amount of 37.99 units up to a value of 37.00 units or equal to a value of −0.99 units equal to a value of −2.61%;
Kazakhstan with a variation from an amount of 0.14 units up to a value of 1.91 units equal to a value of 1.77 units equal to a value of 1267.14%;
South Korea with a diminutive variation from an amount of 1.68 units up to a value of 1.60 units or equal to a value of −0.09 units equivalent to −5.06%;
Lao PDR with a diminutive variation from an amount of 94.33 units up to a value of 94.31 units equal to an amount of −0.02 units equal to a value of −0.02%;
Lebanon with a variation from an amount of 3.78 units up to a value of 3.87 units equal to an amount of 0.09 units equal to a value of 2.38%;
Luxembourg with a variation from an amount of 23.48 units up to a value of 23.50 units or equal to an amount of 0.02 units equal to a value of 0.10%;
Malaysia with a variation from an amount of 9.11 units up to a value of 9.24 units or equal to an amount of 0.13 units equal to a value of 1.43%;
Malta with an increasing variation from an amount of 4.35 units up to a value of 4.52 units or equal to an amount of 0.17 units equal to a value of 3.91%;
Marshall Islands with a variation from an amount of 0.23 units up to a value of 2.23 units or equal to a value of 2.00 units equal to an amount of 868.70%;
Mauritania with a variation from an amount of 7.47 units up to a value of 7.02 units or equal to a variation of −0.45 units equal to an amount of −6.02%;
Mexico with a variation from 15.25 units up to a value of 15.23 units or equal to a variation of −0.02 units equal to −0.13%;
Moldova with a variation from an amount of 5.78 units up to a value of 5.77 units or equal to a value of −0.01 units equal to −0.17%;
Monaco with zero change in renewable energy production;
Niger with an increasing variation from an amount of 0.66 units up to a value of 6.66 units or equal to a value of 6.00 units equal to a value of 908.79%;
Nigeria with zero absolute variation and an amount of renewable energy production equal to an amount of 18.55;
Peru with a variation from an amount of 53.46 units up to a value of 53.45 units or equal to a value of −0.01 units equal to a value of −0.02%;
Poland with a variation from an amount of 12.11 units up to a value of 12.09 units or equal to a variation of −0.02 units equal to an amount of −0.17%;
Qatar with zero variation and an amount of renewable energy production equal to 0.00%;
Russian Federation with a diminutive variation from an amount of 16.20 units up to a value of 16.19 units or equal to an amount of −0.01 units equal to a value of −0.06%;
Rwanda with a variation from an amount of 48.18 units up to a value of 48.17 units or equal to a value of −0.01 units equal to −0.02%;
Sao Tome and Principe with a variation from an amount of 9.71 units up to a value of 9.17 units or equal to a value of −0.54 units equal to a value of −5.56%;
Seychelles with a variation from an amount of 1.79 units up to a value of 1.08 units or equal to a value of −0.71 units equal to an amount of −39.72%;
Sierra Leone with a variation from an amount of 64.94 units up to a value of 64.89 units or equal to a value of −0.05 units or equal to a value of −0.08%;
St. Kitts and Nevis with a variation from an amount of 4.23 units up to a value of 4.19 units equal to an amount of −0.04 units equal to a value of −0.95%;
Timor-Leste and Trinidad and Tobago with zero variation and an absolute value of renewable energy production equal to an amount of 0.00 units;
Tuvalu with a variation from an amount of 11.68 units up to a value of 11.66 units up to a value of −0.02 units equal to an amount of −0.14%;
Uganda with a variation from an amount of 90.23 units up to a value of 90.24 units or equal to an amount of 0.01 units equal to a value of 0.01%;
United Kingdom with zero variation and an amount of renewable energy production equal to an amount of 19.16%;
Uruguay with a variation from an amount of 83.14 units up to a value of 83.10 units or equal to a value of −0.04 units equal to a value of −0.05%;
Zimbabwe with a variation from an amount of 54.88 units up to a value of 54.84 units or equal to a value of −0.04 units equal to a value of −0.07%.
Table 6 shows the results of the predictions.
Therefore, considering the average values, it appears that the value of energy production from renewable sources is expected to grow by an amount equal to 0.21 units or equal to a value of 0.83%. From a strictly geographical point of view, it is foreseen that there will be a growth of renewable energies above all in the African continent, China, and in Kazakhstan. It follows that it will certainly be precisely the countries with low per capita incomes that will increase the value of the production of renewable energy. However, it is very probable that investments in scientific and technological research could lead to further growth in the production capacity of renewables such as allowing an increase in use even in countries with higher per capita incomes, which are generally characterized by greater demand for energy.
We used a set of different machine learning algorithms with the aim of identifying which algorithm is more efficient in prediction. In fact, it is not possible to know, without carrying out the necessary analyses, which algorithms are efficient from a predictive point of view without first making a comparison and a statistical analysis of the predictive performance. The abundance of algorithmic tools used for prediction requires a comparative approach. The comparative approach, which is based on the statistical analysis of the predictive performance, is necessary, on the one hand in order not to underestimate the contribution of the algorithms to the prediction, and on the other hand to create rankings of the most efficient algorithms for the analyzed dataset. In the case of renewable energy output, the trend of the data has essentially a linear structure, due to the fact that the percentage variations in the production of renewable energy at the country level tend to grow slowly, or in any case tend to be devoid of maximums and minimums, whether absolute or relative.
Certainly, there are limitations in the dataset used, which do not allow to connect the production of renewable energy to the energy infrastructure at the country level, and which do not even offer the possibility to check for the presence of economic policies which encourage the installation of energy production plants renewable. In fact, it would be necessary to consider the cost, including the institutional one, associated with the growth of the production of renewable energy, due to the need to design appropriate incentives, both for the plants and for the distribution. Furthermore, it would also be necessary to calculate the capacity of renewable energy to guide the country toward the achievement of environmental sustainability objectives compatible with international agreements.
7. Implication, Limitations, and Further Research
The implications, limitations, and future developments of research around the role of renewable energy as a tool for environmental sustainability and the circular economy are briefly indicated below.
Implications. The implications of our analysis are above all political. In fact, on the one hand, international policy guidelines appear very clear in proposing a significant reduction in pollutant emissions also through renewable energy and the circular economy, while on the other hand, there are many countries that still have many incentives to grow in GDP through the consumption of non-renewable energy. In this sense, it is especially important to underline the difficulties of developing countries, which, having very low per capita incomes, need to consume large quantities of energy, often non-renewable, to reach a level of per capita income of medium-high level. In this sense, the emancipation of a large part of the world’s population from poverty would require a surplus of pollution and non-renewable energy. A paradoxical condition is therefore created: that is, rich countries have a green orientation while poor countries tend to still be linked to polluting forms of industrial production. In some way, it is about the empirical application of the EKC, even with all the limitations that this model entails. In fact, the EKC cannot be generalized and must be verified on a case-by-case basis through empirical data analysis. However, to guarantee the triggering of the positive externalities associated with the EKC, it is necessary for policymakers to have a strategic vision, especially in developing countries, to prevent economic policies from failing, on the one side of lack of economic growth and industrialization, and on the other side for the destruction of the environment.
Limitations. The limitations of the proposed research primarily concern the difficulty of introducing an analysis that also considers the distinctions between Scope 1, Scope 2, and Scope 3, with reference to CO2 emissions. In fact, in our analysis, we did not exactly calculate the relationship between the use of renewable energy and the impact on Scope 1, Scope 2, and Scope 3, at the country level. A further limitation of our analysis concerns the construction of the dataset. In fact, we limited ourselves to analyzing the environmental variables present in the World Bank’s ESG dataset and did not instead consider other variables that could be very relevant, such as the availability of natural resources such as to be able to activate the production of renewable energy. Furthermore, we have not even considered the sociological, cultural, political, institutional, and value-based elements that have an impact on the orientation toward the production of renewable energy both in connection with environmental sustainability and with reference to the circular economy. The fact that a population is more or less sensitive to the environment has a very significant impact in terms of orientation toward the production of renewable energy. Finally, the fact that a country is open to adopting international climate agreements can create the conditions for an increase in investment in renewable energy. However, we must not only consider elements connected to the dimension of moral and political values. In fact, the considerations relating to the financial and institutional dimensions are also very relevant. In this sense, it is the task of policymakers to design the appropriate financial and tax incentives to allow the installation of renewable energy production plants. It is also necessary to evaluate the quality of policymakers in connection with government effectiveness and the fight against corruption. Finally, a further limitation of our analysis consists in not having analyzed the variables that highlight the efficiency of the distribution and energy accumulation systems at the regional and national levels.
Further research. The regional aspects related to the development of renewable energy need to be analyzed in more detail in further research. In fact, one of the characteristics of renewable energy is that it is produced in a de-localized, decentralized way. Furthermore, it would also be necessary to estimate the maximum production of renewable energy that can be produced nationally based on the analysis of the natural resources that are available. Finally, it is also necessary to consider the impact that investment in research and development can have in increasing the efficiency in the production of renewable energy, allowing overcoming the current technological limits. Furthermore, it is also necessary to develop models to verify how climate change also affects the feasibility of natural resources available for renewable energy production such as wind, to avoid installing plants that are unable to produce because of climate change.