Introduction

Human history indicates that all industrial revolutions, from the first to the fourth, are related to the adoption of creative and destructive technologies for efficient production and then, for more responsible consumption. The ongoing Fourth Industrial Revolution, characterized by advanced manufacturing and distributing models, has been changing the world dramatically (Cheng et al. 2021; Ikram et al. 2021). In the environmental aspect, such technological advancements are expected to transform the traditional methods of natural resource exploitation and emissions management to more efficient and cleaner ones (An et al. 2021; Dhanwani et al. 2021; Khan et al. 2021; Zhang and Li 2020). Therefore, the world has made a strong effort to promote the achievements of the Fourth Industrial Revolution to realize higher economic growth while still sustaining environmental quality (Bilgili et al. 2021).

At the same time, environmental degradation has been one of the biggest challenges for humans. Nature is being changed and destroyed at an unprecedented in history, threatening human well-being. On the one hand, many side effects of the industrial revolution have been proved unpreventable and irreversible. On the other hand, the integration of the industrial revolution into environmental management and protection is potential for promoting de-carbonization, bio-diversity conservation, food and water security, to name a few (Shao et al. 2021). On an empirical aspect, while several studies indicate the detrimental effects of innovation on environmental quality (Danish et al. 2018; Kocak and Ulucak 2019; Park et al. 2018), most claim that innovation minimizes the harmful environmental impacts of human activities. However, the majority of studies attempt to explore this relationship by relying on the overall innovation (Chu and Hoang 2021; Khattak et al. 2020). To have a precise conclusion, the innovation used in such research should be environmentally specific. According to Hussain et al. (2020), environmental sustainability could be realized with environmental-related technologies. There are some channels through which environmental technologies can minimize the pollution generated from production and consumption (Ahmed 2020). For example, advanced technologies can increase the efficiency of existing energy sources (significantly reduce the amount of energy needed for producing a unit of good) or they can promote the development and adoption of new cleaner energy sources (solar, wind, hydro, and tidal). In addition, some environmental-related technologies can deal effectively with wastes discharged by human activities.

However, the technological advance may act as a double-edged sword where the progress in technology increases environmental degradation through the rebound effect (Bentzen 2004; Bessec and Fouquau 2008). While energy efficiency improvement is achieved at the micro-level, it may lead to more energy consumption at the macro level (Yang and Li 2017). If the rebound effect exists, the concentration of environmental-related technologies should be shifted from energy efficiency to reducing ecosystem damage such as capture, storage, disposal of polluted air or wastes. In addition, the discovery and exploitation of cleaner and more available energies reduce energy prices, encouraging both production and consumption activities. Consequently, they lead to greater pollution and faster depletion of natural resources (Ali et al. 2021). Wang and Wei (2020) indicate that the high levels of technological progress when combined with the heterogeneous characteristics of OECD countries make the effect of many environmental policies inter-wined and complex. Thus, an important question to be answered is whether higher environmental-related technologies (or green technologies for short) lead to environmental sustainability or not.

Recently, the ecological footprint has been introduced as a more inclusive benchmark of environmental externalities. In essence, it quantifies the amount of biologically productive land and water area needed to supply the resources that a community consumes and to dissolve the waste generated. Since the calculation and publication of data on ecological footprint by the Global Footprint Network, there have been numerous studies employing this indicator to proxy for environmental quality (Altintas and Kassouri 2020; Chu and Le 2021; Danish and Wang 2019; Danish et al. 2020).

Based on the premises above, this study’s primary motive is to analyze the potential impact of green technologies on ecological footprint while considering the role of energy intensity, renewable energy, trade openness, and real income. The reviewed literature indicates that the role of green technologies in the protecting environment, especially in OCED countries, is still limited. The scope of this research is bounded within 20 OECD countries. We decide to focus on OECD countries due to several reasons. First, OECD countries have being exposed to environmental hazards. For example, nearly two-thirds of people are facing dangerous levels of air pollution, according to OECD (2020). In addition, renewable energies still account for a minor proportion of the total primary energy supply although the share is still higher than the rest of the world. The data show that the average carbon footprint per capita has been on a declining trend since 2010 (OECD 2020). The second reason relies on the fact that advanced countries, like OECD members, are witnessing the increasing role of knowledge as a key factor of production (Chu 2021; Wang and Wei 2020). Because of huge investment in research and development, the production and diffusion of technologies have become prominent in OECD countries. The third reason is the data availability of green technologies from the OECD statistics database.

The contribution of this paper to the current literature is three folds. We first examine the long-term connection between ecological footprint and its determining factors, including environmental-related technologies, trade openness, energy intensity, renewable energy, and real income. The reviewed literature indicates that while there have been many studies on the environmental effects of technologies in OECD context, the number of studies exploring the role of environmental-related technologies in the ecological footprint is limited and should be further investigated. Moreover, previous literature often uses carbon emissions for measuring environmental degradation while the ecological footprint is considered a more comprehensive measure (Danish and Wang 2019; Destek and Sarkodie 2019; Chu and Le 2021; Ulucak and Bilgili 2018). Accordingly, this paper connects a series of sustainable development goals (SDG) such as SDG 7 (affordable and clean energy) and SDG 9 (industry, innovation, and infrastructure) to other aspects of the ecological system such as SDG 6 (clean water and sanitation), SDG 13 (climate change), SDG 14 (life below water), and SDG 15 (life on land).

Second, the latest econometric techniques such as second-generation econometric approaches (Pesaran (2015)’s unit root, Westerlund (2005)’s co-integration) and cross-sectional augmented autoregressive distributive lag (Chudik and Pesaran 2015) are adopted to discover the aforementioned nexus. These techniques deal rigorously with the potential existence of cross-sectional dependence and heterogeneity in panel data. It is because the socio-economic activities of OECD countries are integrated due to increasing cross-border trade, investment, and environmental commitments.

Third, the beneficial impact of environmental-related technologies on footprint over the long-run is firmly identified by empirical results. However, such a desirable effect of green technologies on ecological footprint is not found in the short-run. The empirical results also posit that rather than addressing the issues related to environment, technology, and economic growth separately, a co-integrated approach that cover these interrelated factors is necessary to achieve a sustainable environment. Based on these findings, some policy implications are suggested to facilitate the development and deployment of green technologies to achieve sustainable development goals.

The paper’s structure is divided into five parts. The next section provides a review of the literature. The “Methodology” section describes the data, model specification, and methodological issues. The “Results and discussion” section reports and discusses empirical results. The last section wraps up the paper and suggests recommendations.

Literature review

The topic of potential determinants of environmental quality, especially ecological footprint among other indicators, and the role of environmental-related technologies have been extensively examined in the literature. The existing literature, while is inconsistent in the empirical findings, could be classified into three broad categories.

The first collection focuses on the association between environmental-related technologies and the popular environmental quality indicators such as carbon and greenhouse gas emissions (hereafter GHG). Ahmed (2020) studies the nexus between environmental policy stringency, green technologies, and carbon emissions in 20 OECD nations from 1999 to 2015. The empirical results indicate that both environmental regulation and innovation reduce CO2 in the long-run. Similarly, Mensah et al. (2019) find that eco-patents and trademarks alleviate carbon emissions in OECD economies from 1990 to 2015. Alvarez-Herranz et al. (2017) and Balsalobre et al., (2015) explore the impact of research and development on energy on GHG emissions for OECD economies over. The results suggest the beneficial environmental effect of energy innovation and technical innovation can correct the harmful effect of energy intensity. In a smaller context, the study by Wang et al. (2020) analyzes the impact of eco-innovation and export diversification on CO2 emissions in G7 countries from 1990 to 2017. The empirical results show that while ecological innovation reduces CO2 emissions, export diversification magnifies CO2 emissions. Nevertheless, the drawback effect of export diversification gets weakened with the increase in the level of innovation. Other studies indicate insignificant or harmful effects of technological innovations on environmental sustainability. Wang and Wei (2020) find evidence that OECD countries are at a level of excessive technological progress, which causes a rebound effect on carbon emissions. A similar result is discovered by Ahmad et al. (2020a, b) who criticize innovation as the main culprit of carbon emissions in 24 OECD countries. Sohag et al. (2019) analyze the role of scale, composite, and technology effects in OECD countries and find that technological progress has a marginal role in limiting carbon emissions through energy efficiency.

With regard to emerging countries, Danish and Ulucak (2020) examine the long-run relationship between green innovations, renewable, non-renewable energy, and green growth in the context of BRIC countries. The empirical estimations based on panel data from 1992 to 2016 confirm the crucial roles of environmental technologies and renewable energy in promoting green growth. Hussain et al. (2020) evaluate the contribution of environmental-related technologies on controlling CO2 and GHG emissions in E7 countries. The findings significantly confirm the supportive roles of environmental-related technologies and renewable energy sources in achieving sustainable targets. Shao et al. (2021) address the literature gap on the influence of green technology innovations, renewable energy on carbon levels in Next-11 countries. By employing a dataset from 1980 to 2018, authors find that, in the long-run, both green technologies and renewable energies significantly lessen CO2 emissions. However, such beneficial effect of green technology innovations does not occur in the short-run. In the context of a single country, Shahbaz et al. (2020) examine the nexus between public–private partnership investment in the energy sector and CO2 emissions in China taking into consideration the role of technological innovations. The empirical results indicate that while public–private partnership investment substantially degrades environmental quality, the advancement in technologies helps to lower the environmental externalities. In contrast, Su and Moaniba (2017) investigate how innovation reacts to climate change in 70 countries from 1976 to 2014. Empirical results show that a country’s innovation and climate change technology depends on CO2 and other GHG emissions. The government interventions on factors that determining environmental quality such as energy, telecom, transport, and environmental-related projects do not necessarily lead to the development of climate technologies.

Second, the recent literature has shifted attention from traditional measures of environmental quality to a much broader measure, namely ecological footprint. Destek et al. (2018) examine the presence of the Environmental Kuznets Curve (hereafter EKC) hypothesis for 15 European countries. The empirical findings based on two econometric methods of Fully Modified Ordinary Least Squares and Dynamic Ordinary Least Squares are quite different. Specifically, the estimation results from the former approach propose a U-shaped relationship between income and ecological footprint in some countries but those from the latter one suggest that is the case of others. Altintas and Kassouri (2020) check the validity of the EKC hypothesis for a group of 14 European nations over the period from 1990 to 2014. The two authors provide a new perception regarding the relevance of ecological footprint as a useful indicator of environmental quality. Specifically, the empirical results endorse the existence of the EKC hypothesis together with the beneficial environmental effect of renewable energy.

There are several studies with the aim of exploring the causes of ecological footprint in emerging economies. Danish et al. (2020) investigate the factors influencing ecological footprint in BRICS economics. All three factors, including renewable energy, natural resources rent, and urbanization, are found to have a beneficial impact on the environment. The EKC hypothesis is also empirically documented. Similarly, Destek and Sarkodie (2019) test the validity of the EKC hypothesis in the context of 11 newly industrialized countries. The panel data from 1977 to 2013 confirms an inverted U-shaped association between economic expansion and ecological footprint. Similarly, Chardeffine and Mrabet (2017) find that real income displays an inverted U-shaped connection with ecological footprint in eight oil-exporting countries in the Middle East and North African countries. Interestingly, seven non-oil exporting countries observe a U-shaped relationship between real income and ecological footprint. Danish and Wang (2019) investigate the effect of energy consumption, urbanization, and economic growth on the ecological footprint of the Next-11 countries. While higher level of urbanization leads to environmental degradation in most countries, the moderation between income and urbanization significantly reduces environmental degradation. Danish et al. (2019) examine the relationship between economic growth, bio-capacity, and ecological footprint in Pakistan. The data reveal that both factors significantly contribute to environmental deprivation in terms of ecological footprint. In a larger sample, Ulucak and Bilgili (2018) analyze the EKC hypothesis for different groups of high-, middle-, and low-income economies. The ecological footprint is found to increase at the initial level of income and decrease through economic development.

The third group combines two strands of literature by exploring the environmental impact of green technology on ecological footprint. However, to our best knowledge, the role of environmental-related technology, as an effective tool in controlling ecological footprint has not been well studied, particularly in OECD countries. Hussain and Dogan (2021) use a panel dataset of BRICS countries from 1992 to 2016 to estimate the long- and short-run effects of environmental-related technologies and institutional quality on the ecological footprint. The findings suggest that, on the one hand, the governments in BRICS countries improve institutions to moderate environmental impacts. On the other hand, they should encourage the investment in environmental-related technologies, which may facilitate the reduction in ecological footprint. Ahmad et al. (2020a, b) employ a panel dataset from 1984 to 2016 for a group of 22 emerging countries to examine the dynamic link between ecological footprint, natural resources rent, technological innovation, and economic activities. The findings imply that technological innovation is an efficient way to abate the environmental damage. In another interesting research, Destek and Manga (2021) aim at determining the impact of technological innovation on both carbon emissions and ecological footprint for big emerging markets. While technological innovation is found to curb CO2 emissions, it proves ineffective in controlling ecological footprint.

The above summary of the recent literature suggests that the studies on the role of green technologies in controlling ecological footprint are relatively scant. Furthermore, such a limited number of studies disregard advanced economies such as OECD countries. To fill this literature gap, this on-hand study focuses on the nexus between environmental-related technologies in 20 OECD countries.

Methodology

Data

The sample comprises a panel data covering 20 OECD countries. It makes use of data spanning over the period from 1990 to 2015. The choices of countries and time coverage are subjected to the availability of data (see Appendix 1). The dataset includes the variables of ecological footprint, environmental-related technologies, trade openness, energy intensity, renewable energy consumption, and real income.

This study employs ecological footprint as a measure of environmental degradation, which is different from previous studies which mostly focus on carbon emissions (Ahmed, 2020; Alvarez-Herranz et al. 2017; Danish and Ulucak 2020; Hussain et al. 2020; Shao et al. 2021; Wang et al. 2020). The ecological footprint of consumption is defined as the area used to produce the materials consumed and the area need to absorb the emissions (Rees 1992). The ecological footprint increases with the surge in production and consumption activities, which unavoidably utilize a high amount of ecological resources, especially traditional fuels. The data are expressed in global hectares per capita and borrowed from Global Footprint Network.

To measure eco-innovation, we use patents in environmental-related technologies that are filed under the Patent Co-operation Treaty. An advantage of using this definition over the use of overall patents lies in the fact that the former directly help achieve a wide range of environmental control objectives. The data on environmental-related technologies are collected from the OECD Statistics database.

Trade openness quantifies the economic integration of a country into the global economy. It is measured as the sum of export and import as a percentage of gross domestic products. Energy intensity indicates the efficiency of energy used to produce a unit of product. Renewable energy is the proportion of total final consumed energies that are generated by renewable sources such as biomass, hydropower, geothermal, wind, and solar. The data on trade openness, energy intensity, renewable energy consumption, and gross domestic products per capita are taken from the World Bank’s database. The measurement and sources of all variables are presented in Table 1.

Table 1 Variable description
Full size table

Model and method

This study seeks to explore the nexus between environmental-related technologies and ecological footprint while taking into account the effects of economic integration, energy usage efficiency, and renewable energy. Dietz and Rosa (1994, 1997) and Rosa and Dietz (1998) propose the Stochastic Impacts by Regression on Population, Affluence, and Technology model) (STIRPAT) for understanding the complex relationship between the human system and the ecosystem upon which they depend. Based on the STIRPAT model, we design a model in which ecological footprint represents ecosystem degradation, income, and international trade refer to affluence, while energy intensity, renewable energy, and environmental-related technologies account for technology. The population factor is incorporated into the model by using per capita measurement of variables such as ecological footprint and gross domestic products. The model for the relationship between interested variables is proposed as follow:

$$EFC=f\left(ERT, OPE, ENE, REN, GDP\right)$$
(1)

where EFC is ecological footprint per capita, ERT is environmental-related technologies, OPE is trade openness, ENE is energy intensity, REN is renewable energy, GDP is gross domestic product per capita.

The model (1) is transformed into an economic form as below:

$${EFC}_{i,t}={\alpha }_{0}+{\alpha }_{1}{ERT}_{i,t}+{\alpha }_{2}{OPE}_{i,t}+{\alpha }_{3}{ENE}_{i,t}+{\alpha }_{4}{REN}_{i,t}+{\alpha }_{5}{GDP}_{i,t}+{\varepsilon }_{i,t}$$
(2)

where ε remains the error terms. Subscript i (i = 1,…, N) denotes the country in the sample with N being equal to 20. Subscript t (t = 1990,…, 2015) denotes the time period.

The coefficient of interest is \({\alpha }_{1}\). It sign and significance indicate whether a larger number of environmental-related technologies substantially reduces environmental degradation in OECD countries or not. All variables are transformed into natural logarithm to make them consistent with the theoretical framework. The interpretation of the estimated coefficient is in terms of the percentage change.

Due to the high level of socio-economic integration among OECD economies, the possibility of cross-sectional dependence among variables should be expected. Hence, the study starts by testing cross-sectional dependence. The Pesaran (2015)’s test for cross-sectional dependency under the null hypothesis that errors are weakly cross-sectionally dependent is employed. It is also essential to verify the cross-sectional heterogeneity issue since each OECD member has its own characteristics. The on-hand study relies on Pesaran and Yamagata (2008) test to evaluate the null hypothesis of homogeneous slopes. Given the presence of cross-sectional dependence and heterogeneity, the cross-sectionally augmented Im, Pesaran and Shin unit root test is employed to check the stationary of variables (Pesaran, 2021). To conclude whether the long-run connection between variables is established, the Pedroni (1991, 2001)’s and Westerlund (2005)’s tests are utilized. The latter approach is more suitable because it circumvents the presence of cross-sectional dependence.

If the co-integration is statistically confirmed, this study relies on three economic methods of fixed effect with Driscoll-Kay standard errors (hereafter fixed effect with D-KSE), Feasible Generalized Least Squares (hereafter FGLS), Panel Corrected Standard Error (hereafter PCSE) to estimate the long-run elasticities between environmental-related technologies and ecological footprint. Moreover, due to the heterogeneity among individual OECD countries, it is suitable to apply the pooled mean group (hereafter PMG) estimation for our panel data. The PMG estimation can be applied in the case of co-integration between I(0) and I(1) variables. It relies on an ARDL model, which can be expressed in the following equation:

$${EFC}_{i,t}=\sum_{j=1}^{p}{\lambda }_{i,j}{EFC}_{i,t-j}+\sum_{j=0}^{q}{\delta }_{ij}^{^{\prime}}{X}_{i,t-j}+{\beta }_{i}+{\mu }_{i,t}$$
(3)

where λ is a scalar of coefficients of the lagged value of the ecological footprint. X is a vector of independent variables with the corresponding vector of coefficients, δ. β is the specific individual country. μ is the error term.

It is likely to be problematic to use the traditional panel ARDL in the context of high interdependency among countries, especially those that are found within a particular region or have a high level of socio-economic integration (Dimnwobi et al. 2021; Kim et al. 2016; Sharma et al. 2021). Thus, we step further by using the cross-sectional augmented distributed lag approach (hereafter CS-ARDL) proposed by Chudik and Pesaran (2015). In this approach, the cross-sectional means of dependent and explanatory variables are added into the model to control for cross-sectional dependency. The Eq. (3) can be transformed to the style of error correction model as follows:

$$\Delta EF{C}_{i,t}={\theta }_{i}\left(EF{C}_{i,t-1}-\zeta {^{\prime}}{X}_{i,t-1}\right)+{\sum }_{j=1}^{p-1}{\lambda }_{i,t}^{*}\Delta EF{C}_{i,t-j}+{\sum }_{j=0}^{q-1}{{\zeta }^{{^{\prime}}*}}_{i,t}{X}_{i,t-j}+{\sum }_{j=1}^{p}{\omega }_{j}\overline{EF{C }_{t-j}}+{\sum }_{j=0}^{q}{\varphi }_{j}\overline{{X }_{t-j}}+{\sum }_{j=1}^{p-1}{\rho }_{j}\overline{\Delta EF{C }_{t-j}}+{\sum }_{j=0}^{q-1}{\vartheta }_{j}\overline{{\Delta X }_{t-j}}+{\alpha }_{i}+{\mu }_{i,t}$$
(4)

where the cross-sectional averages of ecological footprint and all independent variables are expressed in the form of \(\overline{EFC }\) and \(\overline{X }\). \(\overline{{\Delta EFC }_{t-j}}\) and \(\overline{{\Delta X }_{t-j}}\) are the averages of the differences of the lagged dependent variable and other regressors, respectively. θ is the error-correction term, or the speed of correction from the short-term divergence to the long-term steady state. It is expected that the error-correction term is negative and statistically significant if the variables of interest show a long-run equilibrium relationship.

The CS-ARDL is favored over the PMG, mean group, common correlated mean group, and augmented mean group as it can deal rigorously with both heterogeneity and cross-sectional dependence issues. It is also able to produce both long- and short-term coefficients, which enrich our discussions on the relationship between interested variables. A variety of studies has employed this approach to exploring the determinants of environmental quality (Ahmad et al. 2020a, b; Atsu and Adams 2021; Dimnwobi et al. 2021; Hussain and Dogan 2021; Sharma et al. 2021).

Results and discussion

For a basic understanding of the variables, the descriptive statistics, cross-sectional dependence, slope homogeneity, stationary, and co-integration tests are reported in Tables 2, 3, 4 and 5. Table 2 shows that ecological footprint consumption per capita has a mean value of 1.765 and a standard deviation of 0.255. Although the United States ranks first in terms of ecological footprint, its level has been on the decrease. A similar pattern also occurs in France, Germany, Italy, Spain, Switzerland, and the UK (see Appendix 2). With regard to environmental-related technologies, most countries witness sustainable increases in the number of patents filed under the Patent Co-operation Treaty. Countries that have the highest number of environmental-related patents are Germany, Japan, and the USA (see Appendix 3). The last column in Table 2 shows that the null hypothesis of cross-sectional independence or weak cross-sectional dependence is rejected at 1% significance level. It means that all variables are inter-related. Table 3 reports the Pesaran and Yamagata (2008)’s slope homogeneity test result, which significantly rejects the null hypothesis of homogeneous slope coefficients.

Table 2 Descriptive statistics and cross-sectional dependence test
Full size table
Table 3 Slope homogeneity test
Full size table
Table 4 Unit root test
Full size table
Table 5 Co-integration test. Pedroni test
Full size table

The existence of cross-sectional dependence and slope homogeneity suggest the application of cross-sectionally augmented Im, Pesaran, and Shin unit root test. Table 4 reports the unit root test in two cases, with and without a time trend. According to the results, most variables are stationary at level, except for trade openness and gross domestic products. Nevertheless, after taking the first difference, all variables turn to be stationary at the conventional significance level. This finding supports the use of CS-ARDL in the subsequent analysis.

We step further to check the co-integration or the long-term relationship between variables. As indicated in Tables 5 and 6, both Pedroni (1991, 2001)’s and Westerlund (2005)’s tests significantly reject the null hypothesis of no co-integration. It implies that a long-run relationship between ecological footprint, environmental-related technologies, and other control variables exists.

Table 6 Co-integration test. Westerlund test
Full size table

The core findings involve the long- and short-run relationship between ecological footprint and green technologies. Table 7 proceeds with the fixed effect with D-KSE, FGLS, and PCSE in columns (1) to (3). The long-run coefficient of environmental-related technologies variable is negative at 1% significance level. Specifically, 1% increase in environmental-related technologies in terms of patents (per capita) significantly yields a 0.025%, 0.036%, or 0.045% reduction in ecological footprint (per capita). With regard to the CS-ARDL approach, the findings are summarized in Table 7 columns (4) and (5). The magnitude of long-run coefficient is 0.015% and statistically significant at 5% level. This negative long-run relationship is consistent with the finding of Hussain and Dogan (2021) and Ahmad et al. (2020a, b) on the beneficial impact of environmental-related technologies on ecological footprint. It is also supported by the empirical works on the nexus between green technologies and carbon emissions (Ahmed 2020; Alvarez-Harranz et al. 2017; Danish and Ulucak 2020; Wang, et al. 2020). However, it is found that the short-run connection between ecological footprint and environmental-related technologies is not statistically established although the sign of the coefficient is negative. This outcome is similar to the conclusion of Shao et al. (2021) who find that technologies only affect carbon emissions in the long-run. Briefly, an accumulation in the number of patents related to the environment will help OECD countries to achieve sustainable environmental goals over the long-term. Therefore, not only do enterprises and research institutes invest in research and development activities related to green technologies but the governments should implement policies that encourage such activities as well as the adoption of such innovations in practice.

Table 7 Main results
Full size table

The coefficient for the error correction term is both negative (-0.872) and significant at 99% confidence level. It implies that there is 87.2% adjustment from a short-run divergence to the long-run equilibrium between interested variables. This speed of correction is much higher than those in the case of 20 OECD countries (Ahmed 2020) and of BRICS countries (Hussain and Dogan 2021) but similar to those in the case of G7 countries (Wang et al. 2020) and 22 emerging countries (Ahmad et al. 2020a, b). Such differences rely on the different measures of environmental quality (carbon emissions in Admed (2020) and different samples (BRICS countries in Hussain and Dogan (2021)).

With regard to other control variables, higher trade openness and renewable energy lessen the environmental externalities while higher energy intensity and income per capita degrade the quality of environment. A 1% rise in trade openness leads to 0.04 to 0.05% corresponding reduction in ecological footprint in the long-run. Similarly, a 1% higher proportion of renewable energy consumption mitigates ecological footprint by 0.02 to 0.05%. These two findings are similar to those of Danish and Ulucak (2020)), Hussain et al (2020), and Shao et al. (2021). However, in the short-run, higher economic integration causes harmful environmental impact. To mitigate the environmental degradation, all stakeholders in OECD countries such as the governments, enterprises, households are advised to shift their reliance from non-renewable to renewable energy sources. Moreover, the impact of trade on environmental quality must be scrutinized to avoid the case that OECD countries export pollution to their trading partners.

Energy efficiency exerts a statistical impact on environmental pollution. A 1% increase energy intensity causes 0.6 to 0.8% surges in ecological footprint. The literature is consistent with this outcome (Chu 2021; Danish et al. 2020; Pham et al. 2020). It is worth mentioning that higher income per capita is identified as a key element of the footprint both in the short- and long-run. Ahmed (2020) and Hussain et al. (2021) find a similar pattern.

The study checks the sensitivity of the main findings by substituting environmental related technologies by its two sub-indicators, including patents related to environmental management and climate change mitigation. The same methods are applied for two substituting variables and are posted in Table 8. While the signs and magnitude of independent variables’ coefficients are similar to those of the main findings, there are two remarkable points. First, environmental-related technologies, measured by both sub-indicators, significantly increase ecological footprint in the short-run. Specifically, in a standalone basis, a 1% rise in each sub-indicator leads to a growth by 0.014 to 0.016% in ecological footprint. It signifies the rebound effect of technological advance as discovered by Wang and Wei (2020). Only by combining two types of technology, the beneficial effects of environmental-related technologies are achieved. Second, the speed of adjustment to permit convergence among the variables in the long-run is significantly lower in the case of two sub-indicators of patents (about 56%) than in the case of total patents. This finding is intuitive because the synergy of two types of technological advances will bring more benefits for the ecosystem.

Table 8 Environmental management and climate change mitigation patents estimation
Full size table

To be assured of our main results in terms of estimation method, we employ a system-generalized method of moments (hereafter system-GMM). This estimation approach is chosen because it deals rigorously with heterogeneity and autocorrelation. The results reported in Table 9 reveal that the environmental-related technologies lead to a reduction in ecological footprint.

Table 9 System GMM estimation
Full size table

With regard to model specification, this study challenges Eq. (2) by adding the square of gross domestic products per capita to take into consideration the EKC. The EKC hypothesis suggests the existence of an inverted U-shaped connection between per capita income and per capita pollution. The output (not shown here to save space) indicate that the nexus between environmental technologies and ecological footprint is robust even when we control for the EKC. In sum, the outcomes of three robustness tests validate the statistically negative impact of green technologies on the ecological footprint in the long-run.

Conclusion and policy implications

This study relates two important issues, one is the wake of Industrial revolution 4.0 and the second is the increasing high pressures of human activities on environmental quality. The main question to be answered is whether cutting-edge environmental technologies are effective in protecting and sustaining the environment. Based on such premises, this on-hand study employs recently developed data to empirically examine both long- and short-run relationships between environmental related technologies and ecological footprint for 20 OECD countries. Specifically, it provides new understandings about the dynamic impacts of environmental-related technologies, energy intensity, renewable energy, and trade openness on ecological footprint from 1990 to 2015. The Peasaran (2015)’s cross-sectional dependence, Pesaran and Yamagata (2008)’s slope heterogeneity, Pesaran (2021)’s stationary, and Westerlund (2005)’s co-integration tests are employed to check the long-run connection between studied variables. Finally, a series of estimation techniques such as fixed effects with D-KSE, FGLS, PCSE, CS-ARDL, and system-GMM are conducted to provide robust empirical results.

The primary tests first confirm that all cross-sectional variables are co-integrated. Second, all estimation techniques reveal that ecological footprint is significantly related to environmental-related technologies over the long-run. However, the result of CS-ARDL indicates that the environmental influence of green technologies is not significant in the short-run. In addition, it takes more than one year for a short-term deviation to converge to the long-run equilibrium. This finding is intuitive because the adoption of green technologies is a long-term process that demands huge investment and associates with high uncertainties related to many legal and socio-economic factors such as environmental regulations and consumption habitats (Shabalov et al. 2021; Wada et al., 2021). Thus, the environmental evaluation of green technologies should be conducted and evaluated on a long-term perspective, which considers all the possible externalities. Renewable energy and international trade help control the ecosystem degradation in OECD countries (Anwar et al. 2021; Chen and Lei 2018; Cheng et al. 2021; Chu and Hoang 2021; Qi et al. 2019). Similarly, the efficient use of energy should mitigate environmental degradation (Xie et al. 2021; Salman et al. 2019).

Based on empirical results, this study proposes several relevant policies related to technological innovation and environmental quality. First, there is an urgent need for research focusing on ecological footprint as a benchmark of environmental quality besides a huge existing literature on the determining factors such as carbon and GHG emissions (Danish and Wang 2019; Destek and Sarkodie 2019; Sharif et al. 2020). This new strand of literature will contribute to the achievement of several sustainable development goals related to good health and well-being (SDG 3), water and sanitation (SDG 6), life below water (SDG 14), and life on land (SDG 15).

Second, both governments and enterprises in OCED countries should be able to take advantage of innovations from the Industrial revolution 4.0 on the development of environmental-related technologies. To facilitate such a process, the policymakers are advised to introduce standards for green technologies together with the stricter environmental regulations (Aichele and Felbermayr 2013; de Angelis et al. 2019; Danish et al. 2020; Korhonen et al. 2015; Wang et al. 2019; Zhang 2016). Environmental regulation can affect environmental sustainability through the idea that stringent policies make activities that negatively affect the environment more costly to change the behaviors of economic agents. The government can also design policies to inspire the invention and adoption of environmentally friendly technologies, which effectively limit environmental deprivation. Moreover, they must consider sponsoring the research and development of green technologies through both public and public–private programs or introduce tax cuts on these activities (Bekun et al. 2021). The establishment of a market for such patents to be traded is also essential for the diffusion of green technologies.

Third, at the end of the day, the success of green technologies in sustaining environmental quality depends on the enterprises that adopt them in producing goods and services. The investment cost of adopting clean technologies should be shared between all stakeholders, the governments, enterprises, and consumers through an appropriate mechanism. It is also the objective of SDG 17 which requires the global partnership for accelerating sustainable solutions to all the world’s biggest challenges. The governments and international donors should actively engage with private sectors to help catalyze their business investment and to ensure their responsibility, sustainability, and inclusivity of business activities.

Last, given the fact that both environmental-related technologies and renewable energy are powerful in dealing with environmental degradation, environmental-related technologies should be treated as a valuable instrument to boost the share of renewable energy production and consumption (Adedoyin et al. 2021; Sharif et al. 2020). Raising the carbon price on traditional energy sources and using the revenue to sponsor renewable energy projects should in made in parallel to optimize the coordination of the two strategies (Khan et al. 2021).

There are several potential ways to extend this study in the future. Future work directed toward a larger sample of high-income countries or smaller samples such as G7 or E7 countries would contribute to a more consolidated understanding of the technology-environmental quality nexus. It would be also excited to explore the connection between green technologies and ecological footprint while taking into account the role of institutional quality (Li 2019; Ouyang et al. 2019), economic policy uncertainty (Akadiri et al. 2020), or green finance (Shahbaz et al. 2016), for example Table 10, Figs. 1 and 2.