How economic growth affected from technological innovation, CO2 emissions, and renewable energy consumption? Empirical analysis in G7 countries

Today’s economically developed nations are also among the most advanced in energy production and consumption. In particular, the widespread implementation of renewable energy sources and the plethora of technological advancements supporting long-term prosperity stand out. The present research examines how carbon dioxide (CO2) emissions, technological advancements, and renewable energy sources affect economic expansion. Research and development (R&D) expenses are considered a proxy for technological progress. The analysis quantified the interplay between the factors using annual data for the G7 countries from 1996 through 2020. We examine the association between our variables using panel unit root tests, Pedroni cointegration tests, ARDL coefficient estimations, and Dumitrescu and Hurlin causality tests. The Pedroni cointegration test indicated that the variables are cointegrated. According to the ARDL method of computation, increasing levels of CO2 emissions are beneficial to long-term economic growth. However, improvements in renewable energy and technology dampen economic expansion. The economy’s expansion and increased carbon dioxide emissions have a reciprocal relationship. The Dumitrescu and Hurlin causality test shows a uni-directional chain of events from CO2 emissions to technological improvements, from economic growth to the use of renewable energy, and from consumption of renewable energy to technological advances. Based on our research results, investing in renewable energy consumption is still suggested for long-term sustainable development and environmental protection. Also, directing technological innovations to renewable energy resources and facilities to reduce costs and improve efficiency is suggested.


Introduction
Academic researchers and policymakers have long focused on economic growth, especially long-term sustainability. Economic growth is closely related to research and development activities, energy consumption, and environmental pollution. Energy use, one of the foundations of economic growth, is of immense importance in terms of the limited resources used. In addition, for the production factor, which is an essential factor for economic development, countries must constantly engage in innovative activities to ensure Communicated by Arshian Sharif production techniques, product diversity, and product continuity. These creative activities are provided by technological innovations, patent studies, and research and development activities. As a result of all these activities, the concept of environmental pollution, one of the most critical issues of our century, appears before us. Here, the close relationship between innovation activities, economic growth, energy use, and air pollution emerges.
Energy is crucial for economic growth and activities. Still, environmental hazards arising from fossil fuels (oil, coal, gas) and the possibility of depletion of these resources have triggered the search for alternative energy sources (Bildirici and Gökmenoğlu 2017). Researchers' attention to energy consumption, environmental safeguarding, and economic growth has increased remarkably to cope with the common threat (Baz et al. 2019;Akadiri et al. 2019). At this point, renewable energy sources' importance comes forward. Renewable energy is an environment-friendly energy source known as harmless to public health. In addition, renewable energy resources are a continuous resource that can increase energy security by reducing a country's dependence on fossil fuels (Al-Mulali et al. 2015). Considering specially developed countries' energy needs, economic growth and renewable energy consumption relationship gain tremendous importance. Increments in energy demand and climate problems resulting from fossil fuel consumption have forced most countries to rely on renewable resources to improve air quality and reduce their ecological footprint (Nasrullah et al. 2021). In addition to renewable energy sources, technological progress and industrial innovation also play an essential role in ensuring long-term economic growth (Grossman and Helpman 1994). Schumpeter (2010), one of the first thinkers on the relationship between industrial innovation/technological progress and economic growth, mentioned that these two factors have an essential role in economic development (Hasan and Tucci 2010).
The greatest challenge to achieving global sustainable development today is climate change. Destructive effects of global warming include declining biodiversity, rising ocean levels, reduced food supply, and rising disease mortality. The reasons behind these disasters are mainly vast amounts of CO 2 and greenhouse gases released by fossil fuel consumption and deforestation Liu et al. 2022a, b). When using coal, oil, and natural gas resources, economic growth was maintained at the expense of the emission of greenhouse gases, principally carbon dioxide (Martins et al. 2021). Since CO 2 emissions-accounting for around 76% of all greenhouse gas emissions-are viewed as the leading contributor of GHGs, it is plausible to infer that ecological deterioration has a substantial economic impact because of the enormous development of the industrial sector in industrialized countries (Shahzad et al. 2021;Adebayo et al. 2022a, b). The recent conference in the UK (COP26) in 2021 deals with climate change by making joint decisions agreed on reducing CO 2 emissions by reducing fossil energy consumption and deforestation activities with innovative technologies and intensive forestation activities (UN, 2021).
Technological innovations are seen as a key to reducing environmental pollution. Therefore, fostering technological innovation has become a widely accepted method of addressing ecological issues, such as CO 2 , in the target nations (Chen and Lee 2020;Khan et al. 2020). Many researchers believe in technological advancements' power to improve environmental quality and reduce CO 2 emissions (Yang and Li 2017;Henriques and Borowiecki 2017;Churchill et al. 2019). On the other hand, according to some studies, technological innovation may negatively impact environmental quality (Gu and Wang 2018;Cheng et al. 2019). The logic is that although new technologies might increase resource usage efficiency, their marginal impact is decreasing. An improving economy may necessitate more significant natural resource investment (Chen and Lee 2020).
This research uses panel data analysis to examine the interplay between factors such as CO 2 emissions, technological advancement, the adoption of renewable energy sources, and the long-term growth of economies. We find that the variables have been persistently related using data on the G7 nations from 1996 to 2020 and the Pedroni cointegration test. The G7 countries, the seven most prosperous and most developed countries in the world, were included because they regard themselves as a "community of values," which provides for freedom and human rights, democracy and the rule of law, prosperity, and sustainable development. According to the ARDL model, increasing R&D investment and renewable energy use will harm economic growth, whereas increasing CO 2 emissions would have a favorable impact. According to the Dumitrescu and Hurlin causality test, there is only a one-way connection between renewable energy use and technical advancement, economic development and renewable energy, CO 2 emissions and technological progress, and renewable energy use and CO 2 emissions. The correlation between CO 2 emissions and economic expansion is demonstrated.
In conclusion, the influence of technological innovation on CO 2 emissions is debatable, indicating that it merits additional investigation. Additionally, the presence of heterogeneity shows that it may be researched. This study was based on the following question: how are the variables that affect carbon dioxide emissions affecting economic growth, and in which direction in the long and short term? The study's main contributions are the following: (1) it is a first effort to look at the connections between the use of renewable energy, research, and development spending, carbon dioxide emissions, and economic growth among the G7 countries; (2) one of the different values that the study will add to the literature is to expand the research line by assuming that innovation, which is thought to affect economic growth, makes a significant difference not only in the number of patents but also in R&D expenditures so the research and development expenditures used as a proxy for technological innovations in the study; and (3) it is investigated whether there is a correlation between the economic might of nations with a high level of R&D spending and the benefits of innovation on economic growth through developed economies. Many studies investigate the effect of renewable energy consumption and the number of patents as a proxy of technological innovation on carbon dioxide emissions. As a novelty, we studied renewable energy consumption, research, and development expenses as a proxy for technological innovations and carbon dioxide emissions' effects on economic growth. Additionally, the effects of renewable energy consumption, research and development expenses as a proxy of technological innovations, and economic growth on carbon dioxide emissions are investigated to see both results.
Here is how the rest of the study is laid out. The theoretical underpinnings of the investigation were assessed in the "Literature review" section. Datasets, econometric models, methods, and findings are all described in detail in the "Data, methodology, and results" sections. The "Conclusion and policy implications" section reveals the findings of the research.

Literature review
The findings of a comprehensive investigation of the link between technological progress, economic expansion, and greenhouse gas emissions might have far-reaching policy ramifications and provide potential answers to several related issues. A wealth of research in the literature looks at how rising CO 2 levels are linked to thriving economies. Some academics argue that these two issues need to be addressed together for the reliability of the study (Acheampong 2018). Even while there is some published research on the issue of innovation and economic growth, the bulk of this material focuses on the connection between carbon dioxide emissions and GDP growth. Acheampong (2018) analyzed the GDP growth, carbon emissions, and energy usage of 116 countries from 1990 to 2014 using the PVAR and System-GMM. When economies expand, carbon emissions decrease globally, but there is no evidence that this is true for the Caribbean or Latin America. The research concluded that carbon emissions are helpful for economic expansion. Heidari et al. (2015) used a state-of-the-art PSTR model to test the environmental Kuznets curve (EKC) by analyzing the relationship between energy consumption, CO 2 emissions, and economic development in five ASEAN countries (Thailand, Indonesia, Singapore, Philippines, and Malaysia). The study validated the environmental Kuznets curve for five ASEAN nations and revealed an inverse nonlinear relationship between CO 2 emissions and economic development. Saidi and Hammami (2015) used data from 58 states to analyze the connection between GDP growth, carbon dioxide emissions, and energy consumption. The panel data dynamics were calculated in this study using the Generalized Method of Moments (GMM) from 1990 to 2012. According to the reports of four international committees, a rise in carbon dioxide emissions is shown to have a positive effect on energy efficiency. Bouznit and del Pablo-Romero (2016) analyzed Algeria's CO 2 emissions and economic development. By employing an autoregressive distributed lag model, we verified the validity of Algeria's environmental Kuznets curve (EKC) between 1970 and 2010. Economic growth has been linked to lower carbon emissions, and this trend is likely to continue. Lotfalipour et al. (2010) used the Toda-Yamamoto technique, a novel time series methodology, to analyze the connection between GDP growth, fossil fuel use, and carbon dioxide emissions in Iran from 1967 to 2007. According to experimental evidence, the correlation between economic growth and carbon emissions appears to be uni-directional. A recent study found that emitting carbon dioxide had no positive effect on GDP growth. Using data from 1980 to 2009, Al-Mulali (2011) examined how MENA oil consumption affected GDP growth. Using a panel model, the authors found that CO 2 emissions, oil consumption, and economic growth had a long-term relationship. A cointegration study was performed by Mikayilov et al. (2018) over the years 1992-2013 to look at how GDP growth and CO 2 emissions interacted in Azerbaijan. Long-term coefficients were predicted using various techniques, including the cointegration test, CCR, ARDLBT, FMOLS, DOLS, and the Johansen approach. The studies showed that economic expansion had a statistically significant beneficial effect on carbon dioxide emissions. Ozturk and Acaravci (2010) used the autoregressive distributed lag limit test to look for cointegration between GDPs per capita and carbon emissions per capita in Türkiye between 1968 and 2005 and found none. Employment, GDP growth, carbon dioxide (CO 2 ) emissions, and energy use were all investigated to determine their respective causes and effects. However, linking per capita carbon emissions to GDP has not been verified. Changes in the driving forces of the economy were shown to have an impact on CO 2 emissions, which was investigated by Robalino-López et al. in 2015. The environmental Kuznets curve was used to verify the study's findings, and the results revealed that Venezuela did not satisfy the necessary parameters for the environmental Kuznets curve throughout the test period of 1980-2025. CO 2 emissions and economic development in EU member states were analyzed by Bengochea-Morancho et al. (2001). The correlation between GDP expansion and CO 2 emissions from 1981 to 1995 in 10 European countries was analyzed using a panel data technique. According to the data, there is a clear divide between the most and least industrialized nations. Those nations with incomes above the EU average were found to have higher carbon emissions than those below the average. According to the research of Bhattacharyya and Ghoshal (2010), who focused on the 25 nations responsible for the most considerable proportion of global carbon emissions, the rate at which individual countries' economies expand is positively correlated with the rate at which their carbon emissions rise. Magazzino (2015) utilizes a VAR model, a unit root test, and a Granger causality test to investigate the correlation between CO 2 emissions, GDP growth, and electricity consumption in Israel from 1971 to 2006. The findings demonstrated a cause-and-effect partnership between GDP and CO 2 output. Using the cointegration test, Bozkurt and Akan (2014) examine the relationship between energy consumption, CO 2 emissions, and economic growth in Türkiye. Annual data on carbon dioxide emissions, GDP, and energy use were compiled from 1960 to 2010. The study found that carbon dioxide emissions slow economic expansion. Maclaurin (1953) wrote about how technological innovation contributes to economic expansion. He defined innovation as securing patents and discussed how this process correlates with a thriving economy. Timmons and Bygrave (1986) analyzed the effects of investments in technological advances and the initiatives that resulted from these investments on economic growth from 1967 to 1982; the two factors interact positively. A dynamic stochastic general equilibrium (DSGE) model of economic expansion was developed by Segerstrom (1991). Some companies invest in developing high-quality items while others duplicate them, and this cycle repeats in the model's steady-state equilibrium. Bilbao-Osorio and Rodríguez-Pose (2004) examined the effect of research and development (R&D) activities and investments on the economy and society at large. He claims that discovering novel items through R&D procedures will boost national economies. In recent years, rising rates of innovation have come to be viewed as one of the most critical factors in fostering economic expansion. Promoting innovative technology to pursue better profits through entrepreneurs is crucial to economic progress, and innovation is the vehicle to do just that. Using cross-sectional data for 37 countries that joined the GEM in 2002, Wong et al. (2005) employ a Cobb-Douglas production function to identify technological innovation and company creation as independent determinants of growth. The authors state that even if technical entrepreneurship were to increase or new businesses were to pop up more frequently because of the study, it would not necessarily lead to better economic performance or quicker economic growth rates. Galindo and Méndez-Picazo (2013) examined the connection between economic growth and innovation using a Schumpeterian entrepreneurial framework. The three equations have been used to test hypotheses in this setting for industrialized nations. Estimating equations for the years 2001-2009 employs GLS and PLS techniques. With their research, Galindo and Méndez (2014) hope to conclude the connection between economic development, technical innovation, and entrepreneurialism. The report offers a comparative analysis of entrepreneurial endeavors across advanced economies. For the years 2002-2007, the results reveal a positive association between entrepreneurial and innovative activity and economic growth and a cyclical relationship between the three variables. Pece et al. (2015) investigated the potential impact of innovations on GDP growth over time. Central and Eastern European (CEE) nations are analyzed using multiple regression models. A good correlation between innovation and GDP growth was found when the variable was analyzed using various tools (patents, number of brands, R&D expenditures, etc.). According to Adenle et al. (2017), novel company models and agricultural practices contribute to economic expansion. Thompson (2018) found that the growth rate of internal innovations was crucial to the economy's growth rates and all other macroeconomic variables. To contribute to the research on the link between institutions, innovation, and economic performance in developing nations, Bekana (2020) used an empirical panel dataset to assess 37 sub-Saharan African countries. According to the findings, a more stable democracy can boost economic growth and development by creating a more conducive environment for innovation. Using data from 1990 to 2019, Zhang et al. (2022) investigated the effects of technological innovation on CO 2 emissions in China. Other factors, including economic expansion, renewable energy, and nonrenewable energy, were also included in the study. Economic expansion, foreign direct investment, and energy (renewable and nonrenewable) were independent factors; CO 2 emission was a dependent variable. The study's findings indicated that although foreign direct investment and the use of renewable energy improved the quality of the environment, economic expansion and nonrenewable energy harmed it. Additionally, both positive and negative shocks to technology advancements decreased and enhanced environmental quality.
The effects of financial development and renewable energy on CO 2 emissions in Nordic countries from 1980 and 2020 were studied by Wu et al. (2022). The long-term relationships between CO 2 , renewable energy, and financial development were discovered due to the Westerlund cointegration. In addition, the CS-ARDL results revealed that financial progress and the use of renewable energy increased environmental sustainability in Nordic nations. Wen et al. (2022) used data covering 1990 to 2018 to assess the unequal influence of information and communication technologies, renewable energy consumption, economic growth, financial development, and population on CO 2 emissions in the MINT nations. The study's findings demonstrated a favorable association between population increase, economic expansion, and financial development but a negative relationship between CO 2 emissions, information and communication technologies, and renewable energy sources. Adebayo et al. (2022a, b) used quarterly data between 1992 and 2018 in Russia to study the impact of energy, economic development, and trade openness on the ecological footprint. The study's conclusions showed that while trade openness and renewable energy contribute to the environment's sustainability, nonrenewable energy and economic expansion increase ecological footprint. Using quarterly data from 1985 to 2019, Akadiri et al. (2022) evaluated the impact of economic complexity, energy consumption (renewable and nonrenewable), and environmental footprint in China. The ecological footprint was the study's dependent variable, whereas economic growth, economic complexity, and nonrenewable and renewable consumption were independent factors. The study's findings demonstrated that while nonrenewable energy contributed to environmental disruption in China, renewable energy benefited the ecological footprint. The ecological footprint was also greatly influenced by economic complexity. Anwar et al. (2022a) examined the effects of technical advancements, the use of renewable energy, institutional quality, economic development, and CO 2 emissions in the E-7 nations from 1996 to 2020. The study's findings showed that while population expansion and economic growth increased environmental disturbance, the use of renewable energy, technical advancement, and institutional quality lowered CO 2 emissions. Ibrahim et al. (2022) used data from 1990 to 202 in Germany to examine the effects of technical innovation, transportation services, renewable energy, and trade openness on CO 2 emissions. Results indicated that there were long-term relationships between the factors. Anwar et al. (2022b) used a panel data set of E7 nations from 1996 to 2018 to investigate the relationship between institutional quality, technological innovation, and CO 2 emissions. The study's findings indicated that while technical advancement and institutional quality harmed CO 2 emissions, economic development, trade openness, and the population had a favorable influence. To analyze the effects of financial development, renewable energy consumption, and green technology advancements on CO 2 emissions, Habiba et al. (2022) examined data from the period between 1991 and 2018. The study concluded that while green technological innovation and renewable energy cut carbon emissions, financial developments increased them. To determine how renewable energy consumption, economic development, technical advancements, and institutional quality affected CO 2 emissions between 1996 and 2018 in the E7 nations, Liu et al. (2022a) employed the panel quantile regression approach. The research findings showed that while institutional quality, technical innovation, and renewable energy rose, economic expansion and population growth lowered CO 2 emissions. Using population, environmental innovation (EI), and economic growth factors, Liu et al. (2022b) researched the significance of public-private partnership investment (PPP) on transportation emissions. PPP and EI reduced transport emissions, according to the ARDL approach. Population growth, however, has increased transportation emissions. Using data from 1991 to 2018 in the top 10 polluted nations, Sun et al. (2022) investigated the asymmetric influence of green innovation, economic growth, globalization, and renewable energy use on CO 2 emissions. This work computed the panel data series using the Moment Quantile Regression method (MMQR). The results of the MMQR estimator showed that globalization increased carbon emissions at all levels of quantiles. Contrarily, using renewable energy reduces carbon emissions. Using FMOLS, DOLS, and MMQR tests, Sunday et al. (2022) examined the effects of economic complexity and fragmented energy on CO 2 emissions. Additional factors like technological advancement and economic growth were added to the model. The study's data collection period ran from 1993 to 2018. The investigation revealed a long-term link between the factors and CO 2 emissions. Using FMOLS and DOLS methodologies, Anwar and Malik (2022) investigated the relationships between technical advancement, economic growth, renewable energy, institutional quality, CO 2 emissions, and population for the data period between 1996 and 2018 for the G7 nations. The research revealed that although other factors reduced CO 2 emissions, population and economic expansion increased environmental pollution. Wang et al. (2023) used significant variables such as commerce, GDP, industrialization, government stability, and human capital to examine the impact of biomass energy, hydropower, and nuclear energy on China's footprint. The most recent study used quarterly data spanning 1970Q1 to 2020Q4. The study's findings demonstrated a causal connection between these factors and footprint.
Numerous studies have looked at how economic development affects environmental quality over time. Still, few have looked at how it interacts with other factors, including renewable energy, technological innovation, and carbon dioxide emissions. Only a few studies managed these variables together in a study. Also, few studies investigated these variables' effect on economic growth, but many have about the environment. Therefore, this study will fill the current gap in the literature.
A considerable number of studies have examined the relationship between the use of renewable energy and economic expansion. The brief literature on renewable energy and economic growth is included in Table 1.

Data, methodology, and results
The primary goal of this research is to analyze the effect of G7 countries' usage of renewable energy sources, technical advances, and CO 2 emissions on GDP growth from 1996 to 2020. To do this, the panel data method was employed. The data sets, econometric models, and study procedures are all broken down here. In this section, econometric models are first constructed, and descriptive statistics are given in a table. Later, our variables' stability was checked with LLC and IPS tests. After that, the Pedroni cointegration test was applied to check the correlation between variables. ARDL model has been constructed and applied to see long-and short-term estimation results. Finally, the Dumitrescu and Hurlin causality test was applied to see causalities between variables.
One indication of technical progress included in the study is the amount spent on R&D. This is because these costs are factored into the budgets of wealthy nations. It has been possible to utilize the GDP variable as a proxy for economic growth and then to use the CO 2 emission variables to analyze the effects of that increase on the natural world. All the statistics are from the World Bank's World Development Indicators database. Table 2 displays every independent variable included in the analysis.
Similar to that of Aslan et al. (2021), the following functions are created wherein economic growth (GDP) is the response variable: renewable energy consumption (REN), CO 2 emissions (CO 2 ), and R&D expenditures (RD) are the dependent variables in the first function, and CO 2 emissions (CO 2 ) is the response variable; renewable energy consumption (REN), economic growth (GDP), and R&D expenditures (RD) are the dependent variables in the second function: The following econometric models (Eqs. (3) and (4)) are constructed in line with the variables to be used in the study according to Eqs. (1) and (2): where t denotes the times series (1996-2020) and β k (k = 1,2,3) are the coefficients on GDP, REN, CO 2 , and RD. Finally, µ t represents the estimation residual. Table 2 displays the information that was analyzed throughout the investigation. In this study, we utilize logarithmic analysis on every variable. We will use the econometric model for a panel series analysis (Eqs. (3) and (4)). Unit root tests will first determine whether the variables are stationary. Finally, we will use ARDL coefficient estimates and Granger causality tests to understand the nature of the long-term link between the variables we have identified.
Before proceeding to the analysis, variables' descriptive statistics will be presented in Table 3. The table figures the descriptive statistical information of its data.
We started by checking if the variables were stationary. To do so, scientists use panel unit root testing. High power makes panel unit root testing preferable to individual time series analysis. These tests are adapted versions of conventional multi-series unit root analyses for panels (Al-Mulali et al. 2014). Two examples of panel unit root tests are Levin, Lin, and Chu (LLC) and Im, Pesaran, and Shin (IPS). The Levin, Lin, and Chu (LCC) test supposes that there is a joint unit root process so that ρ i is identical across cross sections, while ρ i might alter between the cross sections in the Im, Pesaran, and Shin (IPS) unit root test because it allows an individual unit root process (Im et al. 2003;Levin et al. 2002). By taking logarithms, panel unit root tests were performed to detect the stagnations of all variables. H 0 indicates that it contains a unit root, while H 1 represents that it does not contain a unit root. As can be seen in Eq. (5) below, the  It is presumed that a standard α = ρ − 1 yet allow the lag order for different terms to diverge between cross sections. The letter y presents the GDP on Eq. (5). The hypothesis used in the LLC test are as follows: H 0 : ρ i : α = 1 (where the null hypothesis contains unit root). H 1 : ρ i : α < 1 (where the alternative hypothesis contains no unit root).
We first verified the stability of all our observables before doing unit root testing through the LLC and IPS panel techniques. The results of the unit root testing are summarized in Table 4. None of the variables are statistically significant at the level of analysis, as shown by the panel unit root tests. The panel unit root condition holds for all variables; hence, the null hypothesis cannot be rejected. As a result, the variables are not holding steady. Once past the initial level, all variables have stabilized, allowing the null hypothesis to be rejected.
The results in Table 4 show that the variables are not steady since their levels have unit roots. Since they are not constant, the analysis moved on to the first differences of all variables. After calculating the initial difference, it was determined that all variables had attained stationarity. The initial difference of no variables contains a unit root. The next step was to examine the interdependencies between the variables. The cointegration test was run after the series was rendered stationary to look for evidence of a long-term correlation between the two variables.
The cointegration test is applied according to the unit root test results. The concept of cointegration may be identified as the systematic long-term joint movement of more than one economic variable (Yoo 2006). The Pedroni cointegration test (Pedroni 1999) was applied in many studies, such as Streimikiene and Kasperowicz (2016), Acaravcı (2010), andAl-Mulali et al. (2014). The Pedroni cointegration test is applied only when the stationary at first difference occurs in variables. The Pedroni cointegration test is essential in examining if the variables are cointegrated or not (Al-Mulali et al. 2014). Also, it allows for heterogeneous intersections and trend coefficients between countries which are represented in Eqs. (6) and (7) below: where i = 1, N refers to the members of the panel (7 countries); t = 1, N represents the period (1996-2020); GDP represents the economic growth; REN represents the renewable energy consumption; CO 2 represents the CO 2 emissions; RD represents the technological innovation; and β refers to the slope of coefficients. The parameter δ i symbolizes the deterministic trend effects, and α i refers to the individual effects. At last, the estimated residual deflections from the longterm relationship are represented with ε i . Pedroni (1999Pedroni ( , 2004 presented two tests to examine whether cointegration exists: panel tests and group testing. The panel v-statistic, the panel rho-statistic, the PP-statistic, and the ADF-statistic are the four statistics used in panel tests based on the within-dimension technique. But three statistics employ the between-dimension method to generate group tests. Using group statistics such as the rho-statistic, the PP-statistic, or the ADF-statistic is recommended. There is an asymptotic normality in these seven statistics. Pedroni's (1999) research provides comprehensive information on the statistics used to test for panel cointegration. The Pedroni cointegration is employed later, recognizing that the study's variables show stationariness at the first difference. It is vital to examine whether or not the variables (lnCO 2 , lnRD, lnREN) are cointegrated with lnGDP. Table 5 briefly reviews the result of the Pedroni cointegration test. According to the results, four statistics are significant when GDP is a dependent variable, and five are significant when considering CO 2 as a dependent variable, therefore rejecting the null hypothesis. This shows a relationship between lnGDP, lnCO 2 , lnRD, and lnREN, respectively, in the long term.
After discovering the cointegration between the series, ARDL (autoregressive distribution lag) models were constructed to state the short-and long-run relationships explicitly. The ARDL framework has many benefits over alternative models (Adams et al. 2018). Data of types I(0) and I(I) and short time series, or both, are suitable for usage inside the ARDL framework, as revealed by Abbasi et al. (2021). Also, the ARDL method gives coherent and vigorous results for the long-term and short-term relationships between variables. For dependent and independent variables, different lags might be used. As calculated, the Pedroni cointegration test (Table 5) reveals a cointegration between variables. Equations (8) and (9) represent the long-run ARDL model as follows: where Δ is the first difference operator and ε t is the pure white noise term. ARDL (2, 3, 3, 3) model was estimated in the model where the maximum lag length was taken as 3. Table 6 shows the estimation results of the ARDL (2,3,3,3) model and the long-term and short-term coefficients calculated based on these results.
The results in Table 6 show that all variables significantly differ from chance at the 1%, 5%, and 10% levels. The long-term estimating model found a statistically significant positive connection between CO 2 emissions and economic growth of 0.675221%. Investment in innovative technologies and using renewable energy sources both dampen GDP growth by 2.542573 percent and 0.161304 percent, respectively, when GDP is taken as a dependent variable. When considering the bigger picture, this CO 2 emission finding agrees with what has been found by Say and Yücel (2006), Ahmad and Du (2017), Chen (2001), and Muhammad (2019). Our finding is consistent with those of Khan et al. (2020), Mohiuddin et al. (2016), and Abbasi et al. (2021), who all found that rising carbon emissions harmed shortterm economic growth. There is a statistically significant negative link between technological innovation and economic growth, suggesting that rising levels of technological innovation dampen economic growth by 2.542573% over the long run. Kahouli (2018), who discusses a similar pattern in Mediterranean nations, lends credence to this conclusion. It was also shown that technical innovation significantly predicted future GDP growth. Contrary to what Shahbaz et al. (2020) assert, renewable energy slows down economic growth. Adopting renewable energy sources may contribute to economic growth in the not-too-distant future. Our results have been confirmed by studies conducted in Türkiye and 15 nations in West Africa (Ocal and Aslan 2013;Maji et al. 2019). Also, when CO 2 is considered a dependent variable, the long-term estimating model found a statistically significant positive connection between CO 2 emissions and economic growth of 0.192707%. Investment in innovative technologies and renewable energy consumption dampen  (2022), Ahmed et al. (2022), and Karaaslan and Çamkaya (2022), our long-term results about GDP and CO 2 are supported. Also, our renewable energy consumption and CO 2 emission results are supported by Karaaslan and Çamkaya (2022), Adebayo et al. (2022a, b), and Zafar et al. (2022). The studies of Chen and Lee (2020), Zhao et al. (2021), andObobisa et al. (2022) on technological innovations and CO 2 emissions support our findings with comparable results. Causality analysis, first developed by Granger (1969), helps to understand whether a variable can obtain helpful information from other variables. The main reason for making Granger causality in panel data analysis is that it can produce more effective results. Finally, using the Dumitrescu and Hurlin causality approach, the causal association between the variables was established after estimating the ARDL coefficients. The Granger causality test was performed and summarized in Table 7. Table 7 provides the probabilities for six distinct chains of causation. Research and development (R&D) and renewable energy sources are inextricably linked. This result agrees with the causal link between R&D and renewable energy consumption expressed by Zafar et al. (2019) and Adedoyin et al. (2020). While Adedoyin et al. (2020) show a bidirectional causation correlation between renewable energy consumption and R&D, Zafar et al. (2019) find only a unidirectional relationship. In addition, there is a direct correlation between a growing economy and the use of renewable energy, consistent with the findings of Liu et al. (2021) and Ben Jebli and Ben Youssef (2015) but not with the findings of Chang et al. (2015), Saidi and Mbarek (2016), Khobai and Roux (2017), Troster et al. (2018), and Sebri and Ben-Salha (2014), who discovered a bi-directional causal relationship between economic growth and renewable energy. Our findings challenge the claims of Tugcu et al. (2012), Chiou-Wei et al. (2008), and Narayan and Doytch (2017). They argued that economic growth and renewable energy sources are unrelated. Our findings that using renewable energy sources is linked to decreased carbon dioxide emissions are in line with those of Jebli and Youssef (2015) and Liu et al. (2015Liu et al. ( , 2021. Alam et al. (2021) and Kahouli (2018) both agree that there is a direct link between rising CO 2 levels and reduced Bi-directional funding for R&D. In contrast to this conclusion, Lin (2021) claims that no causal association exists between CO 2 emissions and R&D investment. Finally, an expanding economy is directly linked to higher CO 2 emissions. Evidence from other studies, including those by Peng et al. (2016), Shahbaz et al. (2016), Saidi and Mbarek (2016), Appiah (2018), and Shahbaz et al. (2016), corroborates this result.

Conclusion and policy implications
Since the beginning of industrialization, energy use has significantly grown. Countries have started looking for alternate energy sources due to the progressive depletion of energy resource stocks caused by rising demand and environmental harm. Countries have shifted to using renewable energy because energy resources must be used effectively, efficiently, and sustainably to ensure sustainable development. This way, the balance between the economy and the environment will be established. With this innovative approach, technological investments in countries have increased.
Most of the studies in the literature investigate the effects of renewable energy, innovations, research and development, and economic growth on the environment, ecological footprint, greenhouse gases, and carbon dioxide emissions. The present study investigates the relationship between renewable energy, technological innovations, carbon dioxide emissions, and economic growth. First, the impact of variables on economic growth is revealed. Later, the effect of variables on carbon dioxide emissions is investigated additionally.
The link between technological innovation, the use of renewable energy sources, and CO 2 emissions for the G7 countries is investigated in this paper, as well as how these factors affect economic growth and CO 2 emissions. The panel data analysis approach examined the long-term relationships between the variables between 1996 and 2020. As an indicator of technological innovation, the variable of R&D expenditures was preferred. This is because developed countries allocate much of their budgets to R&D expenditures. In the following parts of the study, the Pedroni cointegration test, ARDL coefficient estimation, and Dumitrescu and Hurlin panel causality tests were performed, and the relationship between variables was examined.
According to the study, there is a link between G7 countries' CO 2 emissions, consumption of renewable energy, and technological innovations from 1996 to 2020. We know that the variables are fully cointegrated based on Pedroni cointegration results. The long-term coefficients were calculated using the ARDL estimation method after the Pedroni cointegration test established a long-term relationship between the variables. It turns out that the coefficients we estimated using the ARDL method are statistically significant. According to the estimated coefficients, a 1-percent rise in the variables (RD and REN) would harm economic growth. But a growing economy is linked to higher levels of carbon dioxide emissions. It was determined by a Granger causality test that there was a uni-directional relationship between REN and RD, CO 2 and RD, GDP and REN, and CO 2 and GDP. Acheampong (2018) analyzed the GDP growth, carbon emissions, and energy usage of 116 countries. The research concluded that carbon emissions are helpful for economic expansion. We detected a positive effect of carbon dioxide emissions on economic growth, which means increasing emissions have positive effects. Espoir et al. (2022) reveal that a rise in CO 2 emissions positively affects income by 0.23%. Hao and Cho (2022), İnal et al. (2022), and Gong et al. (2022) indicate that the increase in carbon dioxide emissions leads to an increase in economic growth. All these studies show results that support our findings related to economic growth and the carbon dioxide emission relationship.
We observed that renewable energy consumption affects economic growth negatively. The analysis suggests that the following describes the negative impact of the variables on economic growth. Renewable energy may dampen economic growth because it is more expensive than traditional energy sources and necessitates more costly investments. In their study, Aslan et al. (2022) reveal the equivalent results that renewable energy consumption affects economic growth negatively. Wang et al. (2022a) reached similar findings in their study working on OECD countries between 1997 and 2015. On the other hand, Muhammad et al. (2022) found opposite findings that reveal a positive relationship between renewable energy consumption and economic growth in OECD countries. Wang et al. (2022b) also indicate that renewable energy consumption positively impacts nations' economic growth. Similar to Muhammad et al. (2022) and Wang et al. (2022b), the study of Gyimah et al. (2022) covering Ghana reveals the positive relationship between renewable energy consumption and economic growth. Considering the utilization rate of renewable energy resources in the energy share of industrialized nations, the substantial expenditures made in renewable energy to minimize environmental pollution negatively influence economic growth.
There are few studies about the relationship between economic growth and technological innovations in the current literature. Most of the studies investigate the association of innovations with the environment. In addition, it is thought that investments in innovative technologies and R&D expenditures may be more public benefit investments that do not positively affect economic growth, or those technological innovations have superabundant costs to cause a long payback period. Anakpo and Oyenubi (2022) reveal that technological innovations positively affect Southern Africa's economic growth, contrary to our results. These results do not show us that investments in renewable energy sources should be reduced. On the contrary, economic growth can be significantly contributed via renewable energy investments, reducing the environmental pollution in the future for a sustainable environment. On the other hand, shifting the technological innovations to other areas with a much higher return will help turn the negative effect on economic growth positively. Also, it is suggested for countries to spend more on research and development activities related to decreasing cost of renewable energy resources and their infrastructures. These R&D activities also can lead to environmentally friendly constructions of renewable energy facilities for sustainable economic growth and environment.