Introduction

The increased demand for fossil fuels after the industrial revolution has led to a dramatic increase in global greenhouse gas emissions (GHGs) from human activities over the past few decades, nearly doubling environmental degradation compared to the pre-industrial revolution (Chhabra et al. 2023; Işık et al. 2023, 2022; Ongan et al. 2023; Ongan et al. 2022; Ongan et al. 2021; Sharif et al. 2023). Rapid action is required to stop environmental degradation and reach sustainable goals since rising fossil fuel consumption has accelerated environmental degradation (IEA 2023). Many international organizations, governments, decision-makers, and policymakers collaborate to overcome these environmental challenges and find solutions using clean energies to create a sustainable future (Işık et al. 2021a, 2021b; Sharif et al. 2022).

Governments have taken many regional and global meetings and decisions about combating environmental pollution. In this respect, the United Nations Framework Convention on Climate Change is an important decision taken within the scope of combating climate change (Bulut et al. 2023; Işık et al. 2018). In this context, the two most essential agreements made under the leadership of the United Nations are the Kyoto Protocol and the Paris Climate Agreement (Bulut et al. 2023; Işık et al. 2018). In the Kyoto Protocol, countries were obliged to reduce GHGs for the first time. In other words, the countries that took responsibility were limited to environmental protection (Ahmad and Jabeen 2023, Ahmad et al. 2023b, Ahmad et al. 2022a, Ahmad et al. 2022b, Ahmad et al. 2021; Chien et al. 2023; Dagar and Malik 2023; Dagher and Hasanov 2023; Alvarado et al. 2022a; Alvarado et al. 2022b; Alvarado et al. 2022c; Alvarado et al. 2021; Belaid et al. 2021; Işık et el. 2020, Işık et al. 2021b; Dagher et al. 2020; Dagher 2012). Climate scientists stress the significance of achieving 1.5 °C as part of the Paris climate agreement, which aims to keep the long-term increase in global temperature below 2 °C in comparison to the preindustrial era. Future forecasts at the COP26 meeting in Glasgow seek to cut emissions by 2030 and achieve net zero emissions by 2050 (Kirikkaleli et al. 2023). Despite all the efforts to combat climate change and global warming, the rapid rise in global production has increased energy and natural resources usage, expanding the ecological footprint (Bulut et al. 2023; Işık et al. 2018).

Environmental issues have regularly been discussed in research in recent years. Researchers have used many indicators to date as indicators of environmental degradation. These indicators include total GHGs, CO2 emissions, EFP, carbon footprint, resource rent, nitrogen dioxide, sulfur dioxide, PM2.5, and PM10. Among these variables, CO2 emissions and GHGs were used most frequently. However, since these emissions focus only on-air pollution, soil and water pollution are neglected. In this percentage, ecological pollution has been examined frequently in recent studies. The ecological footprint indicator, used as ecological pollution in these studies, is preferred because it is a more comprehensive indicator than CO2 emissions (Pata and Işık 2021).

The EFP, which measures the amount of food, water, energy, transportation, and other materials consumed by an individual or community and their contribution to GHGs, falls short of a sustainable environment. Because water and soil pollution cause environmental problems at least as much as air pollution. In addition, analyzing the EFP alone when assessing environmental sustainability neglects the supply side of nature, and we can only calculate environmental pollution (Işık et al. 2021b). In this context, all pollution must be tackled to achieve the SDGs. For example, SDG-13 is about air pollution, SDG-14 is about water pollution, and SDG-15 is about soil pollution. In this respect, biocapacity is as vital as EFP in environmental sustainability. Biocapacity, which refers to the capacity of a region to use its natural resources sustainably, plays an essential role in balancing the ecosystem. When a region's EFP is larger than its biocapacity, it means that the region is overusing its natural resources, and its consumption is not sustainable.

Researchers, who argue that examining the EFP alone is insufficient for environmental sustainability assessment, used the load capacity factor derived by Siche et al. (2010) as an indicator of environmental quality (Shang et al. 2022).

If a country’s EFP exceeds the biocapacity of its space (EFP > biocapacity), they consume more resources and produce more waste and pollution than the ecosystem can sustainably support. Figure 1 shows that in 1971, the EFP exceeded the biocapacity (EFP biocapacity ≥ 0), creating an ecological deficit that grew rapidly daily and increased the ecological deficit.

Fig. 1
figure 1

Source: Global Footprint Network

World’s Biocapacity and Ecological Footprint.

Full size image

For example, while the ecological deficit was 0.30 in 1982, it increased to 1.19 in 2018, and the ecological overcapacity exceeded four times.

The EFP measures the land and water needed to produce the goods and services consumed in a country. An ecological deficit occurs when the EFP exceeds the biocapacity, the amount of land and water available to provide these resources and services. Figure 2 shows the world's ecological deficit and reserves. In most developed countries, the EFP is 150% larger than the biocapacity.

Fig. 2
figure 2

Source: Global Footprint Network

World Ecological Deficit and Reserves.

Full size image

For example, according to the Global Footprint Network 2018 data for OECD countries, the EFP in Israel, South Korea, and Belgium were 25.8, 9.8, and 9 times the biocapacity, respectively. The fact that the EFP in developed countries is much higher than the biocapacity (BC) makes the load capacity factor values between 0 and 1. When the logarithm of these numbers is taken for analysis, the analysis results are biased or cannot be calculated. For this reason, instead of the load capacity factor (LCF) obtained from the BC x EFP-1 formula, the inverted load capacity factor (ILCF) obtained from the EFPxBC-1 formula gives more effective results in the analysis. LCF shows the load capacity, while ILCF shows the load capacity excess. Used for the first time in the analysis as an environmental indicator, the ILCF measures the pollution dimension of a sustainable environment.

Figure 3 shows the logarithmic reverse load capacity factor values of 30 OECD countries in 2018. In OECD countries, 8 countries (green column) with an ILFC below "0" have ecological reserves, and 22 countries (red column) with an ILFC above "0" have ecological deficits.

Fig. 3
figure 3

Source: Global Footprint Network

Inverted Load Capacity Factor of 30 OECD Countries In 2018.

Full size image

Since the LCF value is greater than 1 in countries with more ecological reserves, effective results can be obtained in the analysis. Siche et al. (2010), who first introduced the LCF as a proxy for a sustainable environment, analyzed the principle of Peru, which has a large ecological reserve as a sample. In addition, Pata et al. (2023a) took LCF as a proxy for a sustainable environment for Latin American and Caribbean countries and found similar results.

However, obtaining the same effective results with the LCF variable may not be possible in countries with ecological deficits. Since most countries have ecological deficits, more efficient results can be obtained with the ILCF variable. In this framework, this study explores the effects of renewable energy and institutional quality on a sustainable environment by including 22 countries of the OECD with ecological deficits.

Do existing environmental pollution proxies represent environmental sustainability? Environmental pollution proxies have focused on the demand side of environmental pollution but neglected the supply side of the ecosystem. In this context, the ILCF variable used in this study considers both the supply and demand side of the ecosystem to ensure environmental sustainability. This study examines the effect of institutional quality, economic growth, and renewable energy on environmental sustainability with annual data from 1999 to 2018 in OECD countries using ARDL and DHPC methods. This study differs from the literature for the following reasons: First, we created a new proxy called ILFC to measure Environmental sustainability. This variable is more comprehensive than the previous environmental pollution variables as it captures environmental quality's supply and demand characteristics. Second, The ILFC variable gives more comprehensive and consistent results than the LCF variable. This is because most countries in the world have an ecological deficit, and the size of this deficit is large. Third, it is the first study to test the EKC hypothesis for ILFC according to the approach of Narayan and Narayan (2010). In this study, the EKC hypothesis is valid in OECD countries. Fourthly, the impact of institutional quality (IQ), economic growth (GDP), renewable energy consumption (REC), and population (POP) on ILFC in OECD countries using annual data from 1999 to 2018 is analyzed by the ARDL cointegration method. Finally, this study will help policymakers and decision-makers formulate appropriate environmental and energy policies by considering the effects of REC, IQ, GDP, and POP. For these reasons, research of this nature, in which sustainable environment (SDG-13) is modeled as the dependent variable and IQ (SDG-16), renewable energy consumption (SDG-7), and economic growth (SDG-8) as independent variables, which require urgent action to combat climate change, is timely and valuable.

The study proceeds as follows: “Literature review” section presents the literature review, “Data, model description, and methodology” section presents the data, model identification, and methodology, “Results and discussion” section presents the results and discussion, and “Conclusions and policy recommendation” section presents the conclusions and policy recommendations.

Literature review

REC is a critical component of environmental sustainability. Renewable energy sources reduce dependence on non-renewable energy (NREC) sources such as fossil fuels. This, in turn, reduces GHGs and helps mitigate the effects of climate change. Furthermore, renewable energy technologies are becoming increasingly affordable and accessible, making them an attractive option for individuals, businesses, and governments. In addition to reducing carbon emissions, renewable energy sources have other environmental benefits, such as reducing air and water pollution and reducing the environmental impacts of energy production.

Environmental sustainability and IQ are closely linked because effective institutions are necessary to promote sustainable practices and address environmental challenges. Countries with strong IQs are better equipped to implement and enforce environmental regulations, promote sustainable development policies, and manage natural resources effectively. In the fight against environmental pollution, it is crucial to improve IQ and REC. The relationship between IQ and REC and environmental sustainability is discussed separately.

Institutional quality and environmental sustainability

Many factors drive the quality of a country's institutions. These factors include public concerns, the rule of law, strict government regulation, regulatory quality, political stability, and control of corruption (Khan et al. 2021). Studies on the environment-IQ relationship have increased in the last decade. There is a close relationship between a country's environment and the quality of its institutions because countries with weak environmental laws will have increased environmental pollution (Egbetokun et al. 2020).

Studies in the literature show that IQ is an essential determinant of environmental degradation. Therefore, countries with stronger institutions can better enforce environmental regulations, protect property rights, promote sustainable development, and ensure public participation in environmental decision-making.

Karimi Alavijeh et al. (2023) investigated the effect of IQ on environmental pollution in EU countries between 2000–2019 and concluded that IQ reduces environmental pollution. However, Le and Ozturk (2020) examined the effect of IQ on CO2 emissions in developing countries from 1990–2014 and concluded that IQ has an increasing effect on CO2 emissions. Safi et al. (2022) investigated the effects of IQ on environmental pollution in E-7 economies for the period 1995–2019 with CS-ARDL and AMG methods. They found a positive relationship with the CS-ARDL method and a negative relationship with the AMG method.

Ni et al. (2022) analyzed the impact of IQ, GDP, natural resources, and digitalization on LCF for the period 1965–2018 for high-resource-consuming economies with the CS-ARDL method, and because of the analysis, natural resources and GDP decreased LCF, while digitalization and good governance improved LCF.

While Le and Ozturk (2020) found a bidirectional causality relationship between IQ and environmental pollution, Udeagha and Ngepah (2022) and Yang et al. (2022), on the other hand, a unidirectional causality relationship was determined. Table 10 in the appendix summarizes selected studies linking IQ and environmental degradation.

Renewable energy consumption and environmental sustainability

Energy use, the most essential input for producing goods in a country, is increasing daily. Since the need for energy is very high, both NREC and REC are increasing. Energy from fossil fuels is unsustainable because it is more accessible and cheaper. In contrast, renewable energy sources are expensive but sustainable (Alvarado et al. 2022b). Countries encourage using alternative energies due to energy scarcity, population growth, energy efficiency, global warming, and climate change (Karimi Alavijeh et al. 2023).

In previous studies in the literature, total energy consumption was included in the model as a proxy for energy consumption (Ozturk and Acaravci 2010; Pao and Tsai 2010). However, in recent studies, REC reduces environmental pollution, while NREC has an increasing effect on environmental pollution (Habiba et al. 2022; Obobisa et al. 2022). Therefore, research on REC is rapidly growing among researchers (Alvarado et al. 2022b). REC, which plays an essential role in sustainable development, is indispensable in countries that seek to improve the quality of the environment as part of sustainable development policies. Many studies have suggested that REC helps reduce environmental degradation (Liu et al. 2022).

Theoretical and conceptual framework

In the twenty-first century, population, technology, and economic growth have grown enormously, but with them have come many challenges that threaten the world, notably environmental problems. Sustainable development is a structural reform in solving these problems (Fukuda-Parr 2016). The term "sustainable development" was first used in the 1980s by environmental activists and scientists who drew attention to environmental issues worldwide (Hák et al. 2016). This expression focuses on the importance of intergenerational resource equality through resource efficiency. At the same time, sustainability is a crucial part of sustainable development, and its implementation is necessary to preserve and improve the quality of life of future generations.

The Millennium Development Goals (MDGs) and the SDGs are two sets of development goals that the United Nations has established. While the MDGs were introduced in 2000 and had a deadline of 2015, the SDGs were introduced in 2015 and are set to be achieved by 2030 (Sachs 2012). The MDGs consist of 8 goals: hunger, education, women's rights, child mortality, infectious diseases, sustainable environment, and global cooperation. In 2005, the United Nations committed to reviewing progress toward the MDGs and set a deadline for achieving the MDGs by 2015 (Sachs and McArthur 2005).

The SDGs are based on the MDGs and consist of 17 goals and 169 targets covering a wide range of issues such as climate change, sustainable consumption and production, and peace and justice. The SDGs are broader and more ambitious than the MDGs and aim to achieve a sustainable future for all people and the planet (Kumar et al. 2016). A buzzword these days, the SDGs have become a research topic for scholars and a problem-solving guide for policymakers (Wahab et al. 2022). While there are similarities between SDGs and MDGs, they differ from MDGs in terms of goals, concepts, and policies (Fukuda-Parr 2016). When the world's current load capacity is analyzed, environmental issues in implementing the United Nations SDGs have attracted attention. Economic development, environmental sustainability, and socio-cultural factors are the three broad categories of the SDGs (Sachs 2012; Rehman et al. 2021a). Particularly economic growth (SDG-8), renewable energy consumption (SDG-7), and institutional quality (SDG-16) play an essential role in collectively achieving global climate change targets and protecting the environment (SDG-13).

Effective management of environmental resources is a major problem for the sustainable development of OECD countries. In this context, one of the critical problems of OECD countries is overconsumption. Therefore, to reduce the negative environmental effects of human activities, increasing the efficiency of resource consumption is crucial (Azam et al. 2021). In this context, for OECD countries to achieve SDG-7 by 2030, decision-makers should encourage and support renewable energy investments, especially wind and solar, increase clean energy efficiency, and support technology-based clean and efficient energy sources as determinants of economic growth in all OECD countries (Cao et al. 2022). Furthermore, OECD governments should strengthen IQ and ensure the active participation of their members in formulating, monitoring, and evaluating environmental sustainability policies (Christoforidis and Katrakilidis 2021).

The SDG framework, especially the indicators, must be conceptually and methodologically well-designed and tested before adoption (Hák et al. 2016). In this context, in the empirical analysis conducted on OECD countries between 1999–2018, the following hypotheses are tested by including the ILCF variable (SDG-13), which is an indicator of environmental quality, institutional quality (SDG-16), economic growth (SDG-8) and renewable energy consumption (SDG-7) as independent variables in the model.

  • H1. According to the general view in the literature, the increase in real income in OECD countries increases ILCF.

  • H2. Renewable energy consumption in OECD countries positively affects environmental quality in line with expectations.

  • H3. The quality of institutions in governments affects environmental sustainability.

  • H4. Population growth negatively impacts environmental quality.

This study tests the effects of independent variables on ILCF using the PMG-ARDL approach to test the above four hypotheses.

Data, model description, and methodology

Data description, and model construction

In the analysis of the study, by applying yearly panel data between 1999–2018, the effects of IQ, GDP, REC, and POP on environmental sustainability are investigated in 22 OECD countries with reverse LCF values greater than 0 shown in Fig. 2. In the study, ILFC was used to measure environmental sustainability and inspired by the LFC developed by Siche et al. (2010), is used as the dependent variable. This indicator used as the dependent variable is captured as EF/BC. Various studies have been conducted to identify the key factors contributing to environmental pollution. However, research has ignored the role of IQ in achieving the SDGs. IQ refers to the overall quality of a country's institutions, including the legal system, government, regulatory framework, and other institutions that shape the country's economic and social policies. Some indicators often used to measure IQ is corruption control, government effectiveness, the rule of law, supervision quality, suitability to comment, and political stability. The values of these indicators are close to each other. In this study, government effectiveness is preferred as an indicator of IQ. Table 1 provides detailed information on the variables.

Table 1 Data description and sources
Full size table

We converted all variables to natural logarithms in order to obtain reliable and consistent results in the analysis results. Table 2 reports the correlation matrix and descriptive statistics.

Table 2 Descriptive statistics and correlation matrix
Full size table

In Table 2, POP is the variable with the highest mean among the series in which we used annual data for 1999–2018 in OECD countries. However, the ILFC has the lowest mean. POP has the highest maximum value, while REC has the lowest maximum value. When we analyze the minimum values of the series, only REC has negative values, while all other variables have positive values. When we examine the correlation matrix, REC affects ILCF negatively, while other independent variables affect ILCF positively. Variance inflation factor (VIF) analysis is applied to check whether there is a multicollinearity problem among the independent variables used in the analysis of our study. VIF analysis results are presented in Table 3.

Table 3 Multi-collinearity test results
Full size table

According to the test results in Table 3, the VIF value is less than 5, indicating no multicollinearity problem among the series. This empirical study investigates the impact of REC, IQ, GDP, and POP on environmental sustainability in OECD countries. The mathematical function and econometric model of the empirical analysis are as follows (Eqns. 1 and 2):

$${lnILCF}_{it}=f({lnIQ}_{it}^{{\beta }_{1i}}, {lnGDP}_{it}^{{\beta }_{2i}},{lnREC}_{it}^{{\beta }_{3i}}, {lnPOP}_{it}^{{\beta }_{4i}})$$
(1)
$$ln{ILCF}_{it}={\mathrm{\alpha }}_{0}+{lnIQ}_{it}^{{\beta }_{1i}}+ {lnGDP}_{it}^{{\beta }_{2i}}+{lnREC}_{it}^{{\beta }_{3i}}+ {lnPOP}_{it}^{{\beta }_{4i}}+ {\upvarepsilon }_{it}$$
(2)

In Eq. 2, \({\mathrm{\alpha }}_{0}\); symbolizes the constant term, \({\upvarepsilon }_{it}\); displays the simultaneous error term, i; the cross-section and t; the time periods. \({\beta }_{1}\)→ \({\beta }_{4}\) show the elasticity of ILFC in response to the concerned explanatory variables. The analysis period of this study covers the period from 1999–2018.

Methodology process

In panel data analysis, a cross-sectional dependence (CSD) test should be performed first to select the appropriate unit root and estimator. If a shock to one variable affects other units, a CSD problem arises between series. Therefore, in this case, 2nd generation unit root tests should be chosen instead of 1st generation unit root tests (Pesaran 2007; Pesaran and Yamagata 2008). In this study, we test for CSD between series using Breusch-Pagan LM, bias-corrected scale LM, and Pesaran's CD tests (Breusch and Pagan 1980; Baltagi et al. 2012; Pesaran 2021). Then, appropriate unit root tests were selected according to the results of the CSD test.

We used the PMG-ARDL approach developed by Pesaran et al. (1999) to evaluate the short and long-term relationship. The PMG-ARDL estimator indicates whether the series are integrated with I(0)/I(1) or a mixture of both. In addition, we tested whether there is causality between the series or the direction of causality. In the picture, Fig. 4 maps the econometric methodology of this study.

Fig. 4
figure 4

Econometric framework

Full size image

Panel estimators

Since the series are stationary at different levels in the unit root test results, the long- and short-term estimates were analyzed using the PMG-ARDL method found by Pesaran et al. (1999). In addition, the PMG-ARDL method provides more efficient and robust results due to the correlation problems between the independent variables and the white noise term in the ARDL method.

Pesaran et al. (1999) developed the PMG-ARDL (p, q) model, which includes the long-term relationship between the series, as in the Eq. (3):

$${\Delta Y}_{1it}={\alpha }_{1i}+{\beta }_{1i}{Y}_{1it-1i}+{\sum }_{l=2}^{k}{\beta }_{1i}{X}_{1it-1}+{\sum }_{j=1}^{p-1}{Y}_{1ij}{\Delta Y}_{1it-j}+{\sum }_{j=0}^{p-1}{\sum }_{l=2}^{k}{Y}_{1ij}{\Delta X}_{1it-j}+{\varepsilon }_{1it}$$
(3)

Here, Y1; the dependent variable, X1; independent variable, Δ: the first difference operator and εit: the error term. If the equation is rearranged by adding the dependent and independent variables in the study to Eq. (3):

$$\begin{array}{c}{\Delta lnILFC}_{it}={\alpha }_{1i}+{\beta }_{1i}{lnILFC}_{it-1i}+{{\beta }_{2i}{lnIQ}_{it-1}+{\beta }_{3i}{lnGDP}_{it-1}+{\beta }_{4i}{lnREC}_{it-1}+{\beta }_{5i}{lnPOP}_{it-1}+\sum }_{j=1}^{p}\\ {Y}_{1i}\Delta {lnILFC}_{it-j}+{\sum }_{i=0}^{q}{Y}_{2i}{\Delta lnIQ}_{it-j}+{\sum }_{i=0}^{q}{Y}_{3i}{\Delta lnGDP}_{it-j}+{\sum }_{i=0}^{q}{Y}_{4i}{\Delta lnREC}_{it-j}+{\sum }_{i=0}^{q}{Y}_{5i}{\Delta lnPOP}_{it-j}+{\varepsilon }_{1it}\end{array}$$
(4)

Panel causality test

Many causality tests are used in the literature to question the causal relationship between variables. The Dumitrescu Hurlin Panel Causality Test (DHPC), frequently preferred recently, is applied in this study. In the DHCP test, the Zbar test represents the normal distribution, and the Wbar test represents the mean (Dumitrescu and Hurlin 2012). The following equation expresses the DHPC test:

$${Z}_{it}={\alpha }_{i}+{\sum }_{j=1}^{p}{\beta }_{i}^{j}{Z}_{it-j}+{\sum }_{j=1}^{p}{\gamma }_{i}^{j}{T}_{it-j}$$
(5)

where j represents the lag length and \({\beta }_{i}^{j}\) represents the autoregressive parameters.

The H0 and H1 hypotheses of this test are as follows:

$$\begin{array}{ll}H_0:\beta_i^{(k)}=0\;for\;i&\mathrm{No}\;\mathrm{causality},\\H_1:\begin{array}{c}\beta_i^{(k)}=0,i=1,2,\;\dots\;.\;N_1\\\beta_i^{(k)}\neq0,i=N_1+1,N_1+2,N_1,\;\dots\;.\;N\end{array}&\mathrm{Unidirectional}\;\mathrm{causality}.\end{array}$$

Results and discussion

Pre-estimation test results

CSD and homogeneity tests are performed between the series before performing the unit roots test in panel data analysis, and appropriate unit root tests are determined. A slope homogeneity test is applied in this study, and the results are presented in Table 4.

Table 4 Slope homogeneity test results
Full size table

According to the results in Table 4, delta (Δ) and corrected delta (Δadj) values confirmed that the slope coefficients were not homogeneous, and it was decided to apply CSD tests that consider heterogeneity.

In panel data analysis, the CSD between the series is examined in the first stage with different tests. If there is CSD between sets, methods that consider this should be used because if not, the results may be biased and inconsistent (Pata et al. 2023b). Therefore, we used the Breusch-Pagan LM (BP-LM), bias-corrected scaled LM (BCS-LM), and Pesaran CD (P-CD) tests developed by Breusch and Pagan (1980) to test for horizontal CSD of the series Baltagi et al. (2012) and Pesaran (2021).

The results presented in Table 5 confirm the CSD between series. The H0 hypothesis is strongly rejected for all series in all three tests. This shows that a possible shock in the electoral states also affects others.

Table 5 CSD test results
Full size table

Panel unit root test results

According to the slope homogeneity and CSD test results, it was decided to apply the 2nd generation unit root test. Therefore, the unit root CIPS was chosen for this analysis.

In Table 6, while ILFC, IQ and GDP series are stationary at I(1) level, REC and POP series are stationary at I(0) level in OECD countries. The fact that the series are stationary at different levels helps the application of the PMG-ARDL method.

Table 6 CIPS unit root test results
Full size table

Panel estimator results

The PMG-ARDL method was used to determine the results of the long-term and short-term estimates of the panel data analysis, and the results are shown in Table 7.

Table 7 PMG-ARDL estimation results
Full size table

In Table 7, REC is negative and significant at the 1% level of ILCF in the long run. Namely, a 1% increase in the REC reduces the ILCF by 0.195% in the long run. There is substantial evidence that REC positively impacts environmental quality in OECD countries. In the literature, Anwar et al. (2021), Liu et al. (2022), and Warsame et al. (2022) studies identified a similar effect of REC on environmental degradation. The results show that IQ positively affects ILFC in OECD countries in the long run at the 1% significance level. In other words, a 1% increase in IQ increases ILCF by 0.250% in the long run. A similar increasing effect of IQ on environmental degradation is concluded by Le and Ozturk (2020), Obobisa et al. (2022), and Wenlong et al. (2022). We then conclude that the effect of GDP and POP on environmental degradation is positive and statistically significant. A 1% increase in GDP increases ILCF by 0.250% and 0.741% in the long and short run, respectively. A 1% increase in POP increases ILCF by 1.089% in the long run. Anwar et al. (2021), Liu et al. (2022), and Warsame et al. (2022) support the results of our analysis.

If we look at the short-term results in Table 7, we can see that ECT is negative and significant, meaning there is a symmetrical cointegrating relationship. These results support studies in the literature and show that the system converges back to equilibrium in the long run after any shock. Table 7 also shows that since the long-run income elasticity is lower than the short-run income elasticity, the EKC hypothesis is valid in OECD countries, according to Narayan and Narayan (2010). In other words, GDP growth reduces the ILCF in OECD countries.

In this study, FMOLS and DOLS tests were analyzed to check the consistency and robustness of the prediction result. The FMOLS and DOLS results are presented in Table 8.

Table 8 FMOLS and DOLS robustness test results
Full size table

When the FMOLS and DOLS test results in Table 8 are examined, it is seen that they are consistent with the PMG-ARDL test results. Only in the FMOLS test is the effect of POP on ILCF as positive as in the PMG-ARDL test, but the effect is weaker.

Panel causality test results

In this study, we use the DHPC heterogeneous panel causality test to investigate whether there is a causal relationship between the dependent and independent variables, as in other related literature. Table 9 shows the results of the causality test.

Table 9 DHPC test results
Full size table

The results indicate a bidirectional causal relationship between REC and ILFC. REC positively promotes environmental sustainability. Similar to the results obtained, Chu (2022) finds a bidirectional causal relationship between REC and environmental pollution in OECD countries. In addition, the analysis shows a unidirectional causal relationship between IQ and ILCF. Ahmad et al. (2022a) obtained similar results to our analysis of developing countries. Finally, although a unidirectional causal relationship was found between POP and ILFC, no causal relationship was found between GDP and ILFC. A graphical representation of the causality of the long-term results and our analysis results can be seen in Fig. 5.

Fig. 5
figure 5

Graphical representation of long run and causality results

Full size image

Conclusions and policy recommendation

Previous studies on environmental sustainability have usually used variables such as CO2 emissions, GHGs, and EFP as indicators of environmental quality. However, these variables neglect the supply side of nature when calculating environmental pollution. In this context, the contribution of environmental resources should be addressed to demonstrate true environmental sustainability. Recently, to solve these problems, researchers such as Shang et al. (2022) used the LCF developed by Siche et al. (2010) as the proxy for environmental sustainability. However, the increasing ecological deficit of the countries decreased the LCF value, and problems were experienced in these values whose logarithms were taken for analysis. To solve these problems and as the proxy of environmental sustainability, we created the ILCF variable. The ILCF variable, like the LCF variable, takes nature's supply and demand aspects. This variable is calculated as EFP divided by BC. An ILCF value greater than 1 indicates that the environmental quality of the country/countries is unsustainable.

Environmental sustainability has recently become the focus of researchers' attention compared to other SDG issues (Altunöz 2023; Çetin et al. 2023; Dogru et al. 2023a, 2023b; 2020, Dogru et al. 2019a; Dzafic and Omerbašić 2023; Faroog, et al. 2023; Islam et al. 2023; Jabeen et al. 2023; Jiang et al. 2023; Jie et al., 2023; Piwowar-Sulej et al. 2023; Rao et al. 2023; Razzaq et al. 2023; Rehman et al. 2023; Ahmad and Wu 2022; Deng et al., 2022; Rehman et al. 2022; Rehman et al. 2021a; Rehman et al. 2021b; Avci and Sarıgül, 2022; Işık 2010, 2013; Işık and Radulescu 2017; Işık et al. 2017, 2019a, b). In addition to energy use, economic growth, trade openness, and financial development, factors such as renewable energy and IQ also play an essential role in environmental sustainability. Since the creation of the SDGs, OECD countries trying to reduce their carbon emissions and improve environmental sustainability have struggled to meet the SDG targets. This study examines the factors affecting the ILCF in 22 OECD countries from 1999 to 2018. This study uses independent variables, including IQ, GDP, REC, and POP. The PMG-ARDL bounds test approach was used for long and short-run forecasts of the series, and the DHPC test was used for the causality relationship between the series. In the PMG-ARDL test results, it has been determined that REC reduces the ILCF in the long term. On the contrary, it has been found that GDP, IQ, and POP increase the ILFC in the long run. Therefore, the DHPC test was applied to determine the causality relationship among the series. According to the DHPC test results, a bidirectional causality relationship exists between REC and POP to ILFC. However, a unidirectional causality relationship was detected from IQ to ILFC, while no causality relationship was detected from GDP to ILFC. The validity of the EKC hypothesis in OECD countries emphasizes that environmental quality will improve over time. Therefore, essential to the world economy, OECD countries should focus on environmental regulations while increasing their income.

This study draws some important implications for policymakers. First, OECD countries must take more challenging and stringent measures against environmental degradation to achieve the SDGs. Considering that resource efficiency and conservation significantly reduce environmental pollution, resource efficiency and conservation in OECD countries should be supported, and these countries should be directed toward environmental research and development activities.

Since REC reduces the ILCF, which is the overcapacity of the LCF, OECD governments running ecological deficits should provide the necessary funds for R&D, innovation, and infrastructure spending to develop these energy types. In this context, decision-makers of OECD countries should improve the share of REC in total energy. Furthermore, these countries should ensure price stability in clean energy and reduce clean energy prices to achieve SDG-7 targets related to REC among the SDGs. Considering that REC reduces ILCF in OECD countries, it can be concluded that they both fulfill the SDG-13 target, and REC has a high share in achieving SDG-7. Therefore, these countries need to improve the share of REC in their total energy consumption.

OECD countries should adopt more challenging and stringent measures to prevent environmental pollution to achieve SDGs. Since resource efficiency and conservation has a significant impact on reducing environmental pollution, it is crucial to support REC practices in OECD countries and channel them into environmental research and development activities.

In light of the information obtained in this study, policymakers and decision-makers in OECD countries should not deviate from environmental sustainability targets while ensuring sustainable economic growth. Given that the REC has reduced the ILCF, which represents excess capacity in the LCF, OECD governments with ecological deficits should allocate the necessary funds for R&D, innovation, and infrastructure investments to promote these forms of energy. In this context, OECD countries should aim to increase the share of RECs in their total energy consumption.

This study has some innovations. First, the ILCF variable is taken as an indicator of environmental sustainability. In future research, new studies can be conducted using different methods and samples with the ILCF variable, considering both the environment's supply and demand dimensions. In this respect, this study will serve as a resource for future studies.