Hostname: page-component-6bb9c88b65-vpjdr Total loading time: 0 Render date: 2025-07-23T14:03:27.049Z Has data issue: false hasContentIssue false

Militarization, industrial progress, and their carbon footprints in North Atlantic Treaty Organization countries: a panel econometric analysis

Published online by Cambridge University Press:  14 July 2025

Mahmud Hasan Riaz*
Affiliation:
Department of Economics & Banking, International Islamic University Chittagong, Chattogram 4318, Bangladesh
Musa Khan
Affiliation:
Department of Statistics, Pirojpur Science and Technology University, Pirojpur 8500, Bangladesh
Zobayer Ahmed
Affiliation:
Bangladesh Institute of Governance and Management, Dhaka 1207, Bangladesh
Shah Asadullah Mohd Zobair
Affiliation:
Department of Economics & Banking, International Islamic University Chittagong, Chattogram 4318, Bangladesh
*
Corresponding author: Mahmud Hasan Riaz; Email: mahmudhriaz@gmail.com
Rights & Permissions [Opens in a new window]

Abstract

This study uniquely explores the impact of militarization on carbon emissions in North Atlantic Treaty Organization (NATO) countries from 1985 to 2019 using panel econometric techniques. NATO countries, characterized by substantial defense budgets, advanced technologies, high industrialization, and significant energy consumption, offer a unique context for examining these factors. Employing the Pooled Mean Group Autoregressive Distributed Lag (PMG-ARDL) and FMOLS models, the research analyzes the long-term and short-term dynamics across three groups: traditional NATO members (Group 1), new NATO members (Group 2), and a combined group (Group 3). Relevant variables used in the estimation are industrialization, technological innovation, energy consumption, and economic growth. Findings reveal that in Group 1, military expenditure and energy consumption significantly increase carbon emissions, while industrialization and technological innovation reduce them. In Group 2, increased military spending and industrialization reduce emissions, but energy consumption and technological innovation increase them. For Group 3, economic growth significantly drives emissions, whereas industrial advancements and selective technological innovations mitigate them. The study underscores the need for tailored environmental policies and technological advancements to reduce carbon emissions, contributing to sustainable development within military alliances. These insights are crucial for policymakers aiming to balance defense needs with environmental sustainability in NATO countries.

Information

Type
Research Article
Creative Commons
Creative Common License - CCCreative Common License - BY
This is an Open Access article, distributed under the terms of the Creative Commons Attribution licence (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted re-use, distribution and reproduction, provided the original article is properly cited.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of Northeastern Agricultural and Resource Economics Association

Introduction

The Earth’s surface temperature has been rising steadily over recent decades, causing significant environmental changes such as rising sea levels, reduced polar ice cover, species extinction, and numerous other ecological problems (D. Li et al., Reference Li, Cao, Zhou, Zhao, Du and Chen2023; Olivier et al., Reference Olivier, Schure and Peters2017). The primary driver of these changes is the increased concentration of carbon dioxide (CO2) and other greenhouse gases in the atmosphere. Among the various contributors to CO2 emissions are militarization, industrialization, technological innovation, energy consumption, and economic growth. While each of these factors has been studied independently, their collective impact on carbon emissions, particularly within a specific geopolitical context like North Atlantic Treaty Organization (NATO) countries, remains underexplored. This study seeks to address this gap by analyzing the interconnected effects of these factors on carbon emissions, focusing on NATO countries from 1985 to 2019.

NATO, the North Atlantic Treaty Organization, is a military alliance comprising 31 member states, each characterized by substantial defense budgets, advanced technological capabilities, high levels of industrialization, and considerable energy consumption. These factors contribute to making NATO countries some of the largest global CO2 emitters. NATO’s significant military expenditures, industrialized arms production, and energy use have raised concerns about its environmental footprint. For instance, research has shown that military organizations’ technologies, armaments, and operations can lead to substantial environmental degradation (Çolak et al., Reference Çolak, Özuyar and Bölükbaşı2022; Jorgenson & Clark, Reference Jorgenson and Clark2011).

Since the end of World War II, there has been a dramatic rise in global CO2 emissions, driven by military expansion, economic growth, and technological advancements. By 2015, global CO2 emissions had reached 32.1 billion tons, with minimal change from 2013 (IEA, 2016). In 2014 alone, CO2 emissions associated with energy consumption increased by 5,406 million metric tons (EIA, 2015). Carbon emissions are a significant contributor to greenhouse gases, accounting for approximately 60% of the global increase in temperature (Franco et al., Reference Franco, Mandla and Rao2017). Recent studies suggest that higher energy use negatively impacts environmental sustainability (Chiu & Zhang, Reference Chiu and Zhang2022). At the NATO Summit in 2021, the alliance’s leaders committed to achieving carbon neutrality by 2050 (NATO, 2022). However, the effectiveness of current NATO policies, which mainly emphasize raising awareness, sharing information, providing training, and assisting member countries in aligning with environmental regulations, remains in question.

NATO members are particularly relevant for this research due to their defense budgets, advanced technological capabilities, high levels of industrialization, and massive energy consumption. These characteristics make them significant CO2 emitters globally. For example, six of the world’s top fifteen military spenders – the USA, UK, Germany, France, Italy, and Canada – are NATO members. Furthermore, the United States, Germany, Canada, and Turkey are also among the top fifteen most significant CO2-emitting countries worldwide (IEA, 2022). NATO’s influence is underscored by its accounting for 55% of global defense spending in 2020 (SIPRI, 2021). The substantial energy consumption associated with military activities in NATO countries potentially causes enormous environmental harm.

The selection of “traditional” and “new” NATO members is based on their distinct historical, economic, political, and security contexts, providing a comprehensive framework for analyzing disparities in military expenditure determinants across the alliance. The “traditional” members, including Belgium, Canada, France, Germany, and the United States, represent economically advanced nations with established defense policies and significant historical contributions to NATO’s strategic objectives. These countries typically exhibit more stable political systems, mature economies, and long-standing defense practices, offering a benchmark for evaluating alliance-wide commitments. In contrast, the “new” NATO members, such as Poland, Latvia, Estonia, and Turkey, bring diverse economic capacities and unique security concerns, often shaped by their more recent political transformations and geographical proximity to regions of strategic importance. Their inclusion highlights the alliance’s expansion since the 1990s and the evolving security landscape, particularly in Eastern Europe and the Balkans. These members often face greater economic and political challenges, such as higher inflation risks, fiscal deficits, or external pressures, which can influence their ability to meet NATO’s 2% GDP defense expenditure target (Odehnal, Reference Odehnal2015; Odehnal & Neubauer, Reference Odehnal and Neubauer2020).

Despite the extensive literature on the impact of factors such as military spending, industrialization, technological innovation, energy use, and economic growth on carbon emissions, a significant gap exists in understanding their collective impact within NATO countries. Previous studies have largely concentrated on individual factors or specific regions, such as the effects of military spending on emissions in Myanmar, Turkey, Pakistan, and India (Ahmed et al., Reference Ahmed, Zafar and Mansoor2020; Gokmenoglu et al., Reference Gokmenoglu, Taspinar and Rahman2021a; Ullah et al., Reference Ullah, Andlib, Majeed, Sohail and Chishti2021) or similar dynamics in the Mediterranean and G20 countries (Erdogan et al., Reference Erdogan, Gedikli, Çevik and Öncü2022; Uddin et al., Reference Uddin, Rashid, Ahamad and Ehigiamusoe2023). However, there is a notable lack of integrated analysis addressing how these factors collectively influence carbon emissions within NATO nations.

Despite extensive studies on NATO’s economic, security, and political determinants of military spending, the existing literature inadequately distinguishes between “traditional” and “new” NATO members, a critical distinction for understanding the varied geopolitical, economic, and environmental contexts of these groups (Odehnal, Reference Odehnal2015; Odehnal & Neubauer, Reference Odehnal and Neubauer2020). Traditional members, characterized by mature economies, established defense policies, and advanced industrialization, exhibit different patterns of technological development and environmental impact compared to newer members, whose recent political transformations and economic constraints create distinct challenges. While previous research has focused on free-riding behavior and disparities in military expenditure commitments, the influence of militarization, industrialization, technological innovation, and energy consumption on NATO’s environmental sustainability remains underexplored.

This gap highlights the need for a nuanced analysis that integrates environmental sustainability into discussions of military expenditure and technological innovation, particularly in the context of NATO’s diverse member states. By examining how these factors intersect with economic, security, and political determinants, this paper aims to provide a comprehensive framework for assessing the environmental implications of militarization and technological advancements across NATO, addressing a critical yet overlooked aspect of alliance-wide sustainability efforts. Additionally, research has not sufficiently explored the impacts on a combined group encompassing both traditional and new NATO members, leaving a critical gap in understanding the unique and collective challenges faced by these groups.

Lastly, while some studies have identified specific links between militarization and environmental harm within NATO (Çolak et al., Reference Çolak, Özuyar and Bölükbaşı2022), they have not proposed strategies for achieving environmental sustainability in this context. The combined influence of militarization, industrialization, technological innovation, and energy consumption on NATO’s environmental sustainability remains largely unexplored, highlighting the need for a more comprehensive analysis of how these factors interact and impact NATO members’ efforts toward sustainability (Asumadu-Sarkodie & Owusu, Reference Asumadu-Sarkodie and Owusu2017; Erdogan, Reference Erdogan2021; K. Li & Lin, Reference Li and Lin2015; Pata, Reference Pata2018; Xinmin et al., Reference Xinmin, Hui, Hafeez, Aziz, Akbar and Mirza2020; Yii & Geetha, Reference Yii and Geetha2017).

This study explores the intersection of militarization, industrialization, technological innovation, energy consumption, and environmental sustainability within NATO countries – a critical yet underexamined area. NATO nations, representing over half of global defense spending and including some of the world’s largest CO2 emitters, have a significant environmental footprint due to their military and industrial activities. By examining these dynamics, this study provides crucial insights into how military alliances can align their operations with global sustainability goals, such as achieving carbon neutrality by 2050. Furthermore, the distinction between ‘traditional’ and ‘new’ NATO members offers a deeper understanding of the disparities in environmental impacts stemming from diverse economic, political, and technological contexts. These findings are essential for policymakers, defense strategists, and environmental advocates working to balance security needs with sustainability imperatives.

This research makes several key contributions to the existing body of knowledge. First, it integrates multiple factors – militarization, industrialization, technological innovation, and energy consumption – to provide a holistic view of their combined impact on carbon emissions in NATO countries. Second, it highlights the environmental disparities between “traditional” and “new” NATO members, offering a nuanced understanding of how historical, economic, and geopolitical differences influence sustainability efforts. Third, the study employs advanced econometric techniques, such as the PMG-ARDL and FMOLS models, to analyze long-term and short-term dynamics, ensuring robust and actionable findings. Finally, it offers practical policy recommendations tailored to the unique challenges faced by NATO members, contributing to global efforts toward sustainable development and environmental preservation within military contexts.

The objective of this study is to examine the environmental impact of militarization, industrialization, technological innovation, and energy consumption in NATO countries over the period 1985 to 2019. By focusing on both “traditional” and “new” NATO members, the study aims to explore the distinct contributions of these groups to carbon emissions, considering their unique economic, political, and security contexts. Additionally, the research seeks to analyze the short-term and long-term relationships between key variables, including military expenditure, industrial activity, technological advancements, energy consumption, and economic growth, using advanced econometric techniques such as the PMG-ARDL and Fully Modified Ordinary Least Squares (FMOLS) models. The study also aims to identify disparities in environmental impacts among NATO member groups, providing insights into how these differences shape alliance-wide sustainability efforts. Ultimately, this research strives to offer actionable recommendations for policymakers, enabling them to develop tailored strategies that mitigate carbon emissions while maintaining NATO’s defense capabilities, thereby contributing to the broader goal of achieving environmental sustainability within military alliances.

In addition, the analysis includes an examination of the Pairwise Granger causality test. In addition to the introductory section, the following sections are organized in the following manner: In Part 2, we will delve into the literature review, examining the existing research and theories. Moving on to Section 3, we will discuss the methodology and conceptual framework, explaining the approach used for data collection, analysis, and model application. Section 4 focuses on the empirical findings and analysis, providing a thorough examination of the study’s results. Moving forward, Section 5 delves into a comprehensive discussion of the policy implications that arise from these outcomes.

Literature review

This research reflects prior discussion examining the connection between militarization, industrialization, advancement in technology, use of energy, economic growth, and discharge of carbon towards achieving environmental sustainability in NATO countries. Understanding the intricate relationship between militarization and environmental degradation has become increasingly critical in addressing global climate change. Various studies have examined how military activities, economic systems, and technological advancements contribute to carbon emissions and ecological harm. The following sections synthesize key research that applies theoretical frameworks such as the Treadmill of Production (ToP), Treadmill of Destruction (ToD), Ecological Modernization Theory (EMT), and World-Systems Theory to investigate these dynamics. These studies highlight the multifaceted impacts of militarization, modernization, and global inequality on environmental outcomes, offering valuable insights into the structural factors driving ecological crises.

Previous research has shown a correlation between G7 nations’ militarization and the discharge of CO2 using annual panel statistics from 1985 to 2015. The panel ARDL model investigated the long-term link among CO2 discharge, militarization, GDP per capita, and energy usage, revealing an extended interaction between these factors (Bildirici, Reference Bildirici2017). Further, another study demonstrated a noteworthy improvement in the correlation between militarization and CO2 emissions in G20 countries by employing panel cointegration and causality methods (Bildirici, Reference Bildirici2019). This study verified a one-way causal relationship between militarism and CO2 emissions.

Jorgenson et al. (Reference Jorgenson, Clark and Kentor2010) examined the validity of the treadmill of destruction framework in 72 nations from 1970 to 2000. Using standard panel data analysis (fixed effects estimator), their findings suggest that militarization directly contributes to ecological deterioration. Additionally, Gokmenoglu et al. (Reference Gokmenoglu, Taspinar and Rahman2021b) found that military expenditure in Turkey from 1960 to 2014, analyzed using FMOLS, led to decreased environmental quality. Conversely, (Chang et al., Reference Chang, Chen and Song2023) showed that military spending significantly impacted CO2 emissions in 15 countries using the CS-ARDL approach, highlighting how increased military expenditure in the Asia-Pacific region has exacerbated environmental damage.

Erdogan, (Reference Erdogan2021) utilized the global vector autoregressive (GVAR) model to analyze the relationship between military spending and environmental changes. This study demonstrated that boosting military expenditure leads to elevated carbon emissions at both local and global levels, emphasizing the destructive influence of military activities on nature. Similarly, Ullah et al. (Reference Ullah, Andlib, Majeed, Sohail and Chishti2021), using the NARDL technique to analyze data from Pakistan and India between 1985 and 2018, revealed that reducing military spending could lower CO2 emissions, while also demonstrating the unequal connections between these factors. In contrast, Konuk et al. (Reference Konuk, Kaya, Akpınar and Yıldız2023) suggested that military spending decreases environmental harm in G7 nations.

Smith & Lengefeld (Reference Smith and Lengefeld2020) investigate the environmental impacts of militarization, focusing on asymmetric warfare and risk-transfer militarism from 2000 to 2010 using panel data from 126 countries. Their study, grounded in theoretical frameworks such as the Treadmill of Production (ToP), Treadmill of Destruction (ToD), and World-Systems Theory, highlights the differentiated effects of militarization on carbon emissions in developed and less-developed nations. In developed countries, reliance on advanced military technologies – reflected in higher military expenditures per soldier (MEPS) – drives significant carbon emissions. Conversely, less-developed nations contribute less to emissions but bear the physical and environmental consequences of traditional militarization. The study underscores the environmental inequality perpetuated by global militarization, wherein developed nations externalize the costs of warfare, exacerbating global environmental degradation, particularly in the Global South. These findings illuminate the hidden ecological costs of modern warfare and emphasize the urgency of addressing the environmental consequences of military activities.

Lengefeld & Smith (Reference Lengefeld and Smith2013) examine the environmental consequences of militarism, capitalism, and modernization during the period from 2001 to 2007, analyzing data from 136 countries. Employing random effects generalized least squares regression, the study evaluates theories such as Ecological Modernization Theory (EMT), the Treadmills of Production (ToP) and Destruction (ToD), and World-Systems Theory. While EMT posits that technological advancements, such as civilian nuclear energy, can mitigate carbon emissions, the findings reveal a positive correlation between nuclear energy use and increased emissions. Moreover, ToP and ToD theories underscore the environmental toll of capitalist and militaristic expansions, with nuclear weapons production significantly contributing to carbon emissions. The study highlights the disparity among core, semi-periphery, and periphery nations, demonstrating that militarization in core nations amplifies environmental degradation while shifting risks to less-developed regions. These findings challenge the notion that modernization and nuclear technology inherently lead to environmental reform, instead revealing their detrimental impact on global carbon emissions.

Jorgenson & Clark (Reference Jorgenson and Clark2009) analyze the ecological impacts of militarization, economic development, and international trade using panel data from 1975 to 2000 across 53 developed and less-developed countries. Through generalized least squares random effects regression, the study integrates theories of ecological modernization, the treadmill of production, the treadmill of destruction, and ecologically unequal exchange. The results reveal that both economic development and militarization independently exacerbate environmental degradation, with MEPS positively correlated with higher per capita ecological footprints. Furthermore, less-developed countries bear a disproportionate ecological burden through unequal trade relationships with wealthier, militarily dominant nations that externalize their environmental costs. By extending the treadmill of destruction framework to the international level, this study underscores the structural inequalities of the global economy, where developed nations expand their ecological footprints while less-developed nations endure resource suppression and environmental degradation.

Jorgenson et al., (Reference Jorgenson, Clark, Thombs, Kentor, Givens, Huang, El Tinay, Auerbach and Mahutga2023) investigate how militarization moderates the relationship between economic growth and carbon emissions, focusing on 106 nations from 1990 to 2016. Utilizing longitudinal models and moderation analysis, the study identifies two key characteristics of militarization – capital intensiveness (measured by MEPS) and size (measured by military participation rate) – as amplifiers of the environmental impact of economic growth. The findings indicate that nations with larger, more capital-intensive militaries experience heightened carbon pollution as economic growth accelerates. This research extends the treadmill of destruction framework by demonstrating the intertwined nature of militarization and economic activities in exacerbating global carbon emissions.

Smith et al., (Reference Smith, Hooks and Lengefeld2024) examine the disproportionate contributions of nation-states to carbon emissions through the lenses of the treadmill of production and treadmill of destruction frameworks. Employing fuzzy set Qualitative Comparative Analysis (fsQCA) across 179 countries, the study identifies the interplay of population, economic growth (GDP per capita), and militarism (measured by military spending as a percentage of GDP) as key drivers of carbon emissions. The findings reveal that population growth and the treadmill of production are necessary conditions for elevated emissions, with militarism further amplifying per capita carbon pollution. The study emphasizes the outsized responsibility of a few nations – such as China, the United States, India, and Russia – for global carbon emissions. It advocates for targeted climate policies that address the combined effects of economic growth and militarism, highlighting the overlooked role of the military in climate change discussions.

York (Reference York2008) explores the effects of de-modernization on CO2 emissions in former Soviet republics from 1992 to 2000 using a STIRPAT model. The study finds that declines in GDP per capita and urbanization reduced emissions, while industrial decline had no significant effect due to “infrastructural momentum,” wherein existing infrastructure continues to drive emissions. Population reductions had a disproportionately large impact on lowering emissions, while militarization independently increased emissions, supporting the treadmill of destruction hypothesis. Additionally, foreign direct investment was linked to higher emissions, offsetting some of the reductions achieved through de-modernization. York underscores the limited ability of de-modernization to reverse historical ecological damage and stresses the importance of policies addressing militarization and infrastructure-driven emissions.

Most empirical research highlights that industrialization significantly influences carbon emissions, exhibiting both positive and negative effects. K. Li & Lin (Reference Li and Lin2015) employed Johansen’s cointegration technique and VECM to assess the impact of industrialization on CO2 emissions in Nigeria from 1980 to 2011, finding that industrial value added had a significant negative effect on emissions. In contrast, Asumadu-Sarkodie & Owusu (2016a) examined the relationship between industrialization and carbon dioxide emissions in Benin from 1980 to 2012 using ARDL, discovering a positive correlation between industrialization and emissions over time. Similarly, Pata (Reference Pata2018) reported that industrial development positively impacts carbon emissions in Turkey.

Studies analyzing broader regions also reveal consistent patterns. Zafar et al. (Reference Zafar, Ullah, Majeed and Yasmeen2020), using FMOLS, found that industrialization positively influences CO2 emissions across 46 countries from 1991 to 2017. Ekwueme & Zoaka (Reference Ekwueme and Zoaka2020) concluded that industrialization is a significant driver of carbon emissions in the MENA region. Additionally, Dong et al., (Reference Dong, He, Li, Mou and Dong2020) observed that updating industrial structures in 41 countries contributes to reduced emissions, suggesting that modernization can mitigate the adverse environmental effects of industrialization.

Empirical study focuses on the bond between technological innovation and ${\rm{C}}{{\rm{O}}_2}\;$ emissions from energy-efficient practices, ultimately reducing ecological deterioration (Balsalobre-Lorente et al., Reference Balsalobre-Lorente, Shahbaz, Roubaud and Farhani2018; Samargandi, Reference Samargandi2017). Thus, Bugden (Reference Bugden2022) analyzed the impact of technological innovation on environmental outcomes within the framework of ecological modernization theory. Using global patent data on environmental technologies across 35 countries from 1982 to 2016, the study employed panel regression analyses to test whether advancements in environmental technologies mitigate the ecological impacts of economic activity. The findings provide limited support for ecological modernization theory. While the development of environmental technologies marginally reduces the ecological footprint associated with economic growth, the direct effect of environmental patents was found to increase, rather than decrease, a nation’s ecological footprint. On the other hand, York et al. (Reference York, Rosa and Dietz2003) critically evaluated modernization theory’s assumptions about environmental sustainability within a cross-national framework. Using the ecological footprint as a comprehensive measure of environmental impact, the study examined whether factors associated with neo-liberal modernization – such as political freedom, civil liberties, and state-led environmentalism – significantly reduce environmental harm. The findings directly contradict the claims of modernization theory. Factors derived from modernization, including institutional and policy reforms, showed no measurable effect on reducing environmental impacts. Instead, the study highlighted the dominant role of material conditions, such as population size, economic production, and urbanization, in driving environmental degradation.

Furthermore, Fisher-Vanden et al. (Reference Fisher-Vanden, Jefferson, Liu and Tao2004) assert that technical innovation is vital in decreasing energy intensity. Moreover, Rafique et al. (Reference Rafique, Li, Larik and Monaheng2020) used the augmented mean group method to study the correlation among financial development, technical innovation, foreign direct investment, and ${\rm{C}}{{\rm{O}}_2}\;$ emissions in the BRICS nations between 1990 and 2017. He figured out that technical advancements helped reduce carbon discharge to a substantial extent. Furthermore, Yii & Geetha (Reference Yii and Geetha2017) figured out a direct connection between technological developments, growth, power use, energy pricing, and ${\rm{C}}{{\rm{O}}_2}\;$ levels in Malaysia between 1971 and 2013. The study showed that scientific advancements led to a short-term decrease in ${\rm{C}}{{\rm{O}}_2}\;$ levels, but no lasting impact was observed. Then, Wang et al. (Reference Wang, Chang, Rizvi and Sari2020) also examine the correlation between carbon dioxide discharge and financial stability in G7 economies, together with technological innovation. Technology and renewable energy capabilities were found to considerably impact reducing carbon dioxide emissions, as shown by the results. This will significantly reduce the energy industry’s coal consumption and ${\rm{C}}{{\rm{O}}_2}\;$ emissions. Another work, from 1971 to 2013 in Malaysia, Yii & Geetha (Reference Yii and Geetha2017) identified a correlation between technological innovation, economic growth, power use, energy costs, and ${\rm{C}}{{\rm{O}}_2}\;$ emissions. Technological advancement resulted in a short-term decrease in ${\rm{C}}{{\rm{O}}_2}\;$ emissions, even without a long-term connection. Another researcher assessed the influence of the high-tech environmental protection factory on reducing the release of ${\rm{C}}{{\rm{O}}_2}\;$ in Asian nations. The study found that low-carbon automobiles, green technology, and energy-saving innovations are significantly essential for reducing ${\rm{C}}{{\rm{O}}_2}\;$ emissions (C. T. Lee et al., Reference Lee, Hashim, Ho, Fan and Klemeš2017). Contrastingly, Omri & Hadj (Reference Omri and Hadj2020) found that technological innovation had a notable adverse effect on the four indices of ${\rm{C}}{{\rm{O}}_2}\;$ emission in developing nations from 1996 to 2014, applying the generalized method of moments (GMM) technique. Further, Shahbaz et al. (Reference Shahbaz, Raghutla, Song, Zameer and Jiao2020) discovered that new technologies negatively impacted the release of carbon dioxide in China from 1984 to 2018 using the bootstrapping autoregressive distributed lag modeling financial mechanics. In Addition, Destek & Manga (Reference Destek and Manga2021) discovered that technological innovation harmed ${\rm{C}}{{\rm{O}}_2}\;$ emissions in ten countries from 1995 to 2016. However, there is no notable connection between technological innovation and environmental imprint. Then, other studies have investigated the correlation between eco research and development (R&D) and carbon emissions. They found a positive relationship between the two in Japanese production companies between 2001 and 2010 (K.-H. Lee & Min, Reference Lee and Min2015).

Prior studies conducted in Indonesia and Turkey revealed that economic growth increased the use of energy and raised ${\rm{C}}{{\rm{O}}_2}\;$ emissions. Thus, Ang (Reference Ang2007) investigates the causal link between ecological quality, use of energy, and economic production in France. She posits a connection between these factors, with evidence suggesting that growth in the economy results in higher energy usage and pollution over time. Additionally, Jorgenson & Clark (Reference Jorgenson and Clark2012) investigate whether economic development and environmental harm have decoupled over time, conducting a comparative international analysis from 1960 to 2005 across 86 developed and less-developed countries. Using panel regression models with Prais-Winsten estimations, they assess three measures of carbon dioxide emissions: total emissions, per capita emissions, and emissions per GDP. The study evaluates competing propositions of ecological modernization and treadmill of production theories, which respectively argue for decoupling versus persistent or increasing environmental degradation with development. The findings reveal evidence of relative decoupling in developed countries, where economic development’s impact on total emissions decreased modestly over time. However, this pattern was not observed in less-developed countries, where the relationship remained stable. The study underscores the role of structural global inequalities, as developed nations achieve eco-efficiency while less-developed countries bear disproportionate environmental burdens due to their integration into the global economy. Jorgenson and Clark emphasize the need to consider transnational production structures and trade patterns in understanding the dynamics of environmental degradation and economic development.

Moreover, Kongkuah et al. (Reference Kongkuah, Yao and Yilanci2022) discovered that economic growth and energy consumption directly correlate with the increase in ${\rm{C}}{{\rm{O}}_2}\;$ emissions in China. Contrary to expectations, Saud et al. (Reference Saud, Chen and Haseeb2020) employed the pooled mean group (PMG-ARDL) method from 1990–2014 in selected countries partaking in the one-belt-one-road project to illustrate a positive connection among economic improvement, energy use, financial growth, and carbon dioxide emissions. Similarly, Riaz et al., (Reference Riaz, Alam, Ali, Ahmed and Raihan2025) employed FMOLS, DOLS, CCR, and GMM techniques from 1991–2019 in BRICS+ countries to reveal that economic growth increases the environmental pressure. Then, Ehigiamusoe & Lean (Reference Ehigiamusoe and Lean2019) analyzed how power utilization, economic development, and financial development affect carbon dioxide venting in 122 nations by employing several econometric models such as FMOLS, DOLS, GMM, CCEMG, and dynamic CCEMG. The study showed that energy utilization, economic growth, and financial improvement undoubtedly impact ${\rm{C}}{{\rm{O}}_2}\;$ levels. Furthermore, Lean & Smyth (Reference Lean and Smyth2010) use commission information from Asian nations to demonstrate long-term causation among energy utilization and ${\rm{C}}{{\rm{O}}_2}\;$ emissions and economic expansion, as well as short-term Granger adversity between ${\rm{C}}{{\rm{O}}_2}\;$ discharge and power utilization. Furthermore, Abbasi et al. (Reference Abbasi, Parveen, Khan and Kamal2020) Abbasi (2020) found a favorable relationship between GDP per capita, energy consumption, and financial growth with carbon dioxide emissions in eight Asian countries from 1982 to 2017, implying fully modified ordinary least squares (FMOLS). In addition, Tahir et al. (Reference Tahir, Luni, Majeed and Zafar2021) employed fully modified ordinary least squares (FMOLS), dynamic least squares (DOLS), and pooled mean group (PMG) to determine that release of carbon dioxide had a favorable long-term impact on GDP per capita, power consumption, as well as financial development in south Asian countries between 1990 and 2014. Then, (Dogan & Turkekul (Reference Dogan and Turkekul2016) found that consumption of power firmly influences ${\rm{C}}{{\rm{O}}_2}\;$ emissions, and expansion of the economy unfavorably influences ${\rm{C}}{{\rm{O}}_2}\;$ emissions, besides improvement of the finance has a noteworthy consequence on ${\rm{C}}{{\rm{O}}_2}\;$ release in the USA from 1960–2010 applying the autoregressive distributed lag (ARDL) model.

Methodology of the study

Conceptual framework

This research framework explores the interconnections between militarization, industrialization, technological innovation, energy usage, and economic growth, focusing on their combined impact on carbon emissions and environmental sustainability, presented in Fig. 1. The framework integrates two theoretically divergent perspectives: the Treadmill of Destruction (ToD) and Ecological Modernization Theory (EMT), each offering distinct interpretations of the relationship between socio-economic development and environmental outcomes.

Figure 1. Visualization of the association between chosen variables.

The upper pathway of the framework is informed by the Treadmill of Destruction, which emphasizes the environmentally deleterious consequences of entrenched institutional structures – particularly those associated with military power, industrial expansion, and the logic of capitalist accumulation (Hooks & Smith, Reference Hooks and Smith2004). Militarization is identified as a significant contributor to carbon emissions through its reliance on energy-intensive defense infrastructures and prolonged ecological exploitation (Jorgenson et al., Reference Jorgenson, Clark and Kentor2010). Similarly, industrialization, characterized by large-scale production and fossil fuel dependency, drives emissions upward by increasing the scale and intensity of resource extraction and consumption (York et al., Reference York, Rosa and Dietz2003). Within this framework, economic growth is not inherently benign; rather, it is conceptualized as a systemic force that amplifies carbon emissions through elevated energy demand, expanded industrial activity, and heightened material throughput (Jorgenson, Reference Jorgenson2007).

Conversely, the lower pathway draws upon Ecological Modernization Theory, which posits that technological innovation and structural modernization can serve as mechanisms for mitigating environmental degradation (Mol & Spaargaren, Reference Mol and Spaargaren2000). From this perspective, technological advancements – particularly in clean energy, energy efficiency, and sustainable production – are instrumental in reducing carbon emissions (Dong et al., Reference Dong, He, Li, Mou and Dong2020). Furthermore, economic growth, when strategically aligned with environmental innovation and supported by effective institutional arrangements, may facilitate the decoupling of economic expansion from environmental harm.

Data and sources

This research aims to analyze the impact of militarization, industrialization, technological innovation, energy consumption, and economic growth on carbon emissions in NATO nations, with the objective of advancing environmental sustainability. To achieve this, the study classifies the 26 NATO countries into two primary groups based on their membership status, as outlined in Appendix Table A1.

The first group, referred to as Group 1, comprises 14 traditional NATO countries, which joined the alliance before or during its early years. The second group, Group 2, consists of 12 new NATO member countries, which joined the alliance in more recent decades. This classification follows Jakub Odehnal (Reference Odehnal2015) framework, which separates NATO members based on their historical accession and associated security and economic factors. Additionally, a combined group, Group 3, is created by aggregating all 26 NATO countries for a comprehensive analysis. Table 1 displays the chosen variables, abbreviations, descriptions, units, and sources. The data extracted from World Development Indicators (WDI) from 1985 to 2019.

Table 1. Variables’ description and data source

Econometric analysis

Model development

The objective of this research is to examine the impact of military expenditure (lnME), industrialization (lnIND), technical innovation (lnTECH), energy consumption (lnEC), and economic growth (lnGDP) on carbon emissions ( $lnC{O_2}$ ) in NATO member nations. Building on this foundation, the study adopts the mechanism of the Cobb–Douglas production function (Cobb & Douglas, Reference Cobb and Douglas1928) to develop the following economic model. This approach allows for the systematic analysis of how key factors interact to influence carbon emissions while maintaining flexibility in representing non-linear relationships. The functional form is specified as follows:

(1) $$C{O_2} = f\;\left( {{\rm{MEX}},{\rm{IND}},{\rm{TECH}},{\rm{\;EC}},{\rm{\;GDP}}} \right)$$

where, $C{O_2}$ indicates carbon emissions, MEX denotes military expenditure, IND stands for industrialization, GDP represents economic growth, TECH shows technical innovation, EC energy consumption represents energy consumption, while GDP represents economic growth.

The equation that establishes the associations between the variables under examination is established afterward (Riaz et al., Reference Riaz, Alam, Ali, Ahmed and Raihan2025; Uddin et al., Reference Uddin, Rashid, Ahamad and Ehigiamusoe2023). Equation 1 serves as a foundational model, chosen for its suitability in capturing the nuances of the relationships. To enhance clarity and statistical robustness, the model is reformulated using natural logarithms:

(2) $$lnCO{2_{it}} = {\beta _0} + {\beta _1}lnME{X_{it}} + {\beta _2}lnIN{D_{it}} + {\beta _3}lnTEC{H_{it}} + {\beta _4}lnE{C_{it}} + {\beta _5}lnGD{P_{it}} + {\varepsilon _{it}}$$

Where i represents the number of countries, where i can be any integer starting from 1, where N indicates the cross country, and t is a positive integer starting from 1. The variable T denotes the time in panel data. Equation (1) denotes the coefficients β0, β1, β2, β3, β4, and β5 for each explanatory variable. Furthermore, the model includes the εit error component, and β0 represents a constant coefficient that differs among countries but remains constant over time. The equation specifies that the dependent variable is represented by $lnC{O_2}$ . Eq. (2) includes the independent variables lnMEX, lnIND, lnTECH, lnEC, and lnGDP, as defined in Table A1.

In the initial phase of analysis, a comprehensive tabulation of descriptive statistics was conducted for each variable to assess central tendencies and dispersions. This step provided insights into the basic characteristics of the data set, laying a foundation for subsequent analytical procedures. Following this, a correlation analysis examined linear relationships among the variables and detected potential multicollinearity. The severity of multicollinearity was further investigated using the Variance Inflation Factor (VIF) analysis, quantifying its impact on the econometric model’s robustness and reliability. Additionally, tests for heterogeneity within the panel data were performed to ensure the validity of subsequent analyses, accounting for potential variations across cross-sectional units and supporting the study’s findings’ accuracy and generalizability. Pesaran’s CD test (2004), Friedman’s test (1937), and Frees’ test (1995) were employed to examine cross-sectional dependency within the panel data, providing insights into the strength of dependencies across panel units.

Detection of cross-sectional dependencies necessitates advanced unit root tests due to potential biases in conventional tests. Therefore, second-generation unit root tests, specifically the Cross-sectional Im, Pesaran, and Shin (CIPS) test and the Cross-sectional Augmented Dickey-Fuller (CADF) test (Pesaran, Reference Pesaran2007) were utilized. These tests adjust for cross-sectional dependencies by incorporating cross-sectional averages, offering more reliable insights into the data’s stationarity properties. The CIPS test extends the traditional Im, Pesaran, and Shin (IPS) test by accommodating cross-sectional dependence, while the CADF test enhances the robustness of the Augmented Dickey-Fuller (ADF) check when cross-sectional dependencies are present.

The study used the (Pedroni, Reference Pedroni1999, ) and Kao Residual Cointegration tests Kao (Reference Kao1999) to investigate long-term cointegration among the selected variables, building on the results of these unit root tests. The Pedroni test allows for heterogeneity across cross-sectional units in short-term dynamics and long-term cointegrating vectors, whereas the Kao Residual Cointegration test assumes homogeneity in long-term cointegrating relationships. These tests validated the theoretical framework by confirming how variables move together over time and verifying the presence of long-term cointegration, crucial for understanding sustainable relationships among economic variables.

The panel ARDL model

In this study, we utilize an econometric approach to analyze panel data related to carbon emissions, focusing on variables such as military expenditure, industrialization, technical innovation, energy consumption, and economic growth. Traditional econometric techniques like Vector Auto Regression, Engle and Granger causality, Ordinary Least Squares (OLS), and Johansen cointegration have known limitations that can affect the reliability of results. The ARDL model, developed by Pesaran & Smith (Reference Pesaran and Smith1995) Pesaran et al., (Reference Pesaran, Shin and Smith2001), has been widely employed in agricultural and climate studies to analyze both long- and short-term relationships (Bildirici, Reference Bildirici2017; Uddin et al., Reference Uddin, Rashid, Ahamad and Ehigiamusoe2023). In this study, it is utilized to address these methodological constraints effectively. This model is particularly well-suited for our research due to its numerous significant benefits. Firstly, the ARDL method is effective with limited sample sizes. Statistically, it is comparable to the standard error correction model, though it calculates standard errors differently, leading to unbiased estimates. Secondly, the ARDL approach, through linear transformation, allows us to evaluate both long- and short-term relationships between the variables of interest. Thirdly, the ARDL model assesses the stationarity of variables using unit root tests to identify level I(0) and the first difference I(1) values. This method ensures no variable is stationary at the second difference I (2), as supported by Brown et al., (Reference Brown, Durbin and Evans1975). Secondly, the linear transformation of the ARDL technique serves to evaluate the long-term and short-term associations between the relevant variables. Additionally, the ARDL model examines the stationarity of variables by conducting unit root tests to determine if they are at level I (0) or first difference I (1) values. This method guarantees that there is no stationary variable at the second difference, I (2), as evidenced by the research conducted by Brown et al. (Reference Brown, Durbin and Evans1975).

Moreover, three estimators are used in ARDL estimation: the mean group (MG), the pooled mean group (PMG), and the dynamic fixed effect (DFE). We conducted a Hausman test to assess the appropriateness of the PMG estimator in comparison to the Mean Group (MG) estimator. The MG estimator can accommodate fluctuations in both short-term and long-term estimators and is renowned for its constancy. Suppose the null hypothesis, which suggests that there is no significant difference between the PMG and MG, is confirmed. In that case, it indicates that the PMG estimator is superior in terms of effectiveness and reliability compared to the MG. The Pooled Mean Group (PMG) approach was employed to ascertain the enduring and immediate connections between the chosen variables and explore the dynamic diversity features among the countries. The Pooled Mean Group (PMG) model, proposed by Pesaran (Reference Pesaran, Shin and Smith1999), is a suitable ARDL model for assessing the dynamic panel with error correction. It pertains to variables having a combination of integrated orders. The model is defined in Equation 3.

(3) $${Y_{it}} = \mathop \int \nolimits_{j = 1}^{p - 1} \gamma _y^i{\left( {{y_i}} \right)_{t - j}} + \mathop \int \nolimits_{j = 1}^{q - 1} \delta _y^i({\left( {{X_i}} \right)_{t - j}} + {\varphi ^i}{\left( {{y_i}} \right)_{t - 1}} + {\mu _i} + {\varepsilon _{it}}w$$

Where ${\left( {{X_i}} \right)_{t - j}}$ denotes the vector (k x l) represents the explanatory variables for group i, and ${\mu _i}$ represents the random effect. Within an unbalanced panel, the values of p and q have the potential to differ across different countries. Therefore, it is possible to reparametrize this as a Vector Error Correction Model (VECM) system in Equation (4):

(4) $$\Delta {Y_{it}} = {\theta _i}\left( {{Y_{i,t - 1}} - {\beta _i}{X_{i,t - 1}}} \right) + \mathop \int \nolimits_{j = 1}^{q - 1} \gamma _y^i\Delta {Y_{i,t - j}} + \mathop \int \nolimits_{j = 0}^{q - 1} \delta _y^i\Delta {\left( {{X_i}} \right)_{,t - j}} + {\mu _i} + {\varepsilon _{it}}{\rm{\;}}$$

In this context, βi represents the parameters that are applicable in the long term, while θi refers to the parameters that account for error correction. Within the framework of the PMG approach, the long-run parameters exhibit a high degree of similarity across different countries. Eq. (5) displays the subsequent model:

(5) $$\Delta {y_{it}} = {\theta _i}\left( {{Y_{i,t - 1}} - {\beta _i}{X_{i,t - 1}}} \right) + \mathop \int \nolimits_{j = 1}^{q - 1} \gamma _y^i{\left( {{Y_i}} \right)_{t - j}} + \mathop \int \nolimits_{j = 0}^{q - 1} \delta _y^i\Delta {\left( {{X_i}} \right)_{,t - j}} + {\mu _i} + {\varepsilon _{it}}{\rm{\;}}$$

The $lnC{O_2}$ is represented by the dependent variable ‘y,’ while the explanatory variables are represented by ‘X.’ Furthermore, the short-run coefficients of the dependent and explanatory variables are denoted by “γ” and “δ.” “β” stands for the long-run coefficients, whereas “θ” is the coefficient that modifies the long-run coefficient and represents the adjustment rate. The countries and times are denoted by the variables “i” and “t,” respectively. The regression for long-term improvement is shown in square brackets.

Lastly, diagnostic techniques were employed to validate the results after determining the long-term effects of the variables. These included the Wooldridge test for autocorrelation, the Ramsey RESET test for model stability, and the Breusch-Pagan/Cook-Weisberg test for heteroskedasticity. Additionally, the robustness of the model was verified using the FMOLS methodology, which provides supplementary confirmation of the stability and reliability of the results. Figure 2 illustrates the comprehensive methodological framework adopted in this investigation, incorporating a variety of diagnostic tests and methodologies to ensure meticulous analysis and robust interpretation of the findings.

Figure 2. Flow chart of the analytical techniques employed in the research.

H1: Militarization significantly contributes to an increase in carbon dioxide emissions in NATO countries.

H2: Industrialization has a positive effect on carbon dioxide emissions in NATO countries.

H3: Technological innovation contributes to a reduction in carbon dioxide emissions in NATO countries.

H4: Energy consumption significantly increases carbon dioxide emissions in NATO countries.

H5: Economic growth contributes to a reduction in carbon dioxide emissions in NATO countries.

Results and discussion

Descriptive statistics

The summary statistics for the variables are presented in Table A2, which includes the results of normality tests such as skewness and kurtosis. The variables display positive skewness. Positively skewed variables indicate that all subsequent data points for that variable will exceed one. The kurtosis data indicate that all variables exhibit a leptokurtic distribution with values below 3. Furthermore, a higher VIF value, presented in Table A3, indicates a greater likelihood of multicollinearity, which requires further investigation. A VIF value exceeding 10 suggests significant multicollinearity that needs to be addressed. The findings suggest that multicollinearity doesn’t constitute a substantial concern when the VIF value is less than 10.

Heterogeneity of slope test

Neglecting the heterogeneity in slope coefficients may lead to inaccurate approximations and biased results. Table A4 demonstrates that the Pesaran and Yamagata (2008) test supports this study’s alternative hypothesis of slope coefficient heterogeneity. Robust estimators built upon this principle were utilized for diverse scenarios and interconnection across different sections.

Cross-sectional dependence test

The results of the CSD tests are represented in Table A5. When cross-sectional variables are tested using the CSD test, statistically significant P-values lead to the rejection of the null hypothesis of cross-sectional independence. This indicates that CSD exists for the studied variables at a significance level of 1% and 5%. The econometric calculations accounted for CSD, showing a substantial cross-sectional relationship in panel group error terms.

Panel unit root test

Table A6 displays the parallel integration order, as the panel root test results show. The second-generation panel unit root tests, such as CIPS and CADF, tackle the CSD issue by demonstrating that the unit root null hypothesis remains valid irrespective of temporal trends and levels. After implementing the initial modification, the six variables remained consistent regardless of the significance levels tested, which included 1%, 5%, and 10%. In econometrics, variables were classified as order 0, denoted as I (0), and subsequently as order 1, denoted as I (1). This opportunity allows for a more comprehensive examination of the extended-term balance between the variables.

Pedroni and Kao panel cointegration test

Each of the variables were determined to be stationary at the first difference. It is imperative to establish the stationarity of variables in order to proceed to the subsequent phase of panel cointegration testing. Table A7 shows four data points for Group 1, five for Group 2, and five data for Group 3 from a total of seven statistics indicating the failure to accept the null hypothesis according to the Pedroni test, both intra-dimensional and inter-dimensional. This conclusion confirms a long-term cointegration among $C{O_2}\;$ discharges, militarization, industrialization, technical innovation, and energy usage. The outcome of the Kao ADF cointegration test provides evidence in favor of the existence of a long-term cointegration connection.

Long-run and short-run result discussion

To accurately analyze both short-term and long-term relationships among carbon emissions, militarization, industrialization, technological innovation, energy usage, and economic growth, it was essential to first confirm that the variables were cointegrated, which established the presence of a long-term relationship. The Hausman test was then conducted to determine the most suitable estimator among the MG, PMG, and DFE estimators. The test results confirmed that the PMG estimator exhibited greater efficiency for assessing long-term relationships. Consequently, the PMG estimation technique was employed to explore both long-term and short-term dynamics within ‘traditional NATO member countries’ (Group 1), ‘new NATO member countries’ (Group 2), and a combined group (Group 3). The detailed results of these analyses are presented in Table 2, highlighting the relationships and impacts of various factors on carbon emissions and environmental sustainability.

Table 2. Parameter estimation of panel PMG models for all groups

Note: ***, ** and * represent statistical significance at 1%, 5%, 10% level, respectively.

Militarization and carbon emissions

The study identifies a significant long-term relationship between military spending and carbon emissions in traditional NATO member states (Group 1). Specifically, a 1% increase in military expenditure is associated with a 17.68% rise in emissions, highlighting the considerable environmental burden of military activities. This result aligns with previous research by Jorgenson and Clark (2016), Baldric (2019), Bildirici (2017b), and A.K. Jorgenson (Reference Jorgenson, Clark and Kentor2010), which emphasize the high energy consumption of military operations as a major contributor to greenhouse gas emissions. However, in the short term, military spending in Group 1 shows no significant impact on carbon emissions, suggesting that the environmental consequences of military activities in these nations are more evident in the long run.

In contrast, new NATO member countries (Group 2) display a different dynamic, with military spending leading to a 2.59% reduction in carbon emissions over the long term. This finding supports recent studies by Konuk et al. (Reference Konuk, Kaya, Akpınar and Yıldız2023) and Ullah et al. (Reference Ullah, Andlib, Majeed, Sohail and Chishti2021), indicating that new NATO members have likely adopted more energy-efficient technologies and sustainable military practices. While this positive trend is evident in the long run, a short-term positive relationship between military spending and emissions persists in Group 2, suggesting that increased military activities still contribute to short-term emissions before the benefits of improved practices and technologies materialize.

For the combined NATO group (Group 3), the overall effect of military spending on carbon emissions is negative, with a 1.11% reduction in emissions for every 1% increase in expenditure. This reflects the mitigating influence of sustainable practices in new NATO members, which offset the emissions increase observed in traditional members. In the short term, the impact of military spending on emissions in Group 3 is insignificant, indicating a balance between the immediate emissions from military activities and the mitigating long-term effects. The long-term reduction in emissions in Group 3 highlights the importance of technological advancements and energy efficiency, particularly in new NATO member states, even as traditional members continue to rely on energy-intensive military operations.

Industrialization and carbon emissions

The study reveals that industrialization has a substantial negative impact on long-term carbon emissions in traditional NATO member countries (Group 1). Specifically, a 1% increase in industrialization is associated with a 19.45% reduction in emissions, indicating that advancements in industrial processes, particularly through technological innovation and energy efficiency measures, have led to significant reductions in emissions over time. This suggests that traditional NATO members have successfully modernized their industrial sectors, shifting towards cleaner production methods. However, the short-term impact of industrialization on carbon emissions in Group 1 is negligible, implying that the immediate benefits of industrial advancements may not be apparent in the short term.

In new NATO member countries (Group 2), industrialization also demonstrates a significant negative effect on long-term carbon emissions, with a 3.42% reduction in emissions for every 1% increase in industrialization. These findings are consistent with studies by Dong et al. (Reference Dong, He, Li, Mou and Dong2020) and Wang et al. (2011), which highlight the role of technological advancements, energy efficiency, and the adoption of cleaner energy sources in reducing emissions. The results suggest that both traditional and new NATO members can reduce emissions by fostering innovation, implementing stricter government regulations and incentives, and promoting energy-efficient practices and environmental awareness. This long-term reduction underscores the importance of sustained efforts in industrial modernization for achieving environmental sustainability in these regions.

For the combined NATO member group (Group 3), industrialization continues to exert a significant negative impact on carbon emissions, with a 0.84% reduction in emissions for every 1% increase in industrial activity. This outcome reflects the collective progress of both traditional and new NATO members, where industrial sectors have embraced cleaner technologies and energy-efficient processes. The findings align with previous studies by Wang (2011) and Dong et al. (Reference Dong, He, Li, Mou and Dong2020), which emphasize the critical role of technological innovation and stricter environmental regulations in mitigating emissions, even in highly industrialized nations. Similar to Group 1, the short-term effect of industrialization on emissions in Group 3 is insignificant, suggesting that the environmental benefits of industrial advancements are realized over time, rather than immediately. The overall negative impact of industrialization on emissions across Group 3 demonstrates a clear shift towards cleaner production methods in both older and newer NATO members, contributing to reduced carbon footprints in the long run.

Technological innovation and carbon emissions

Technological innovation has a significant negative impact on carbon emissions in traditional NATO member countries (Group 1) over the long term. Specifically, a 1% increase in technological innovation leads to a 2.99% reduction in carbon emissions, highlighting the role of technological advancements in fostering environmental sustainability. This finding aligns with the results of Adebayo (2022) and Yii and Geetha (Reference Yii and Geetha2017), who emphasize that innovations enhance energy efficiency and cleaner production methods. Moreover, in the short term, technological innovation continues to demonstrate a substantial reduction in emissions for Group 1, reinforcing its environmental benefits across different time horizons.

For new NATO member countries (Group 2), the impact of technological innovation on carbon emissions is more nuanced. While long-term effects show a marginally positive relationship, in the short term, a 1% rise in technological innovation leads to a 1.47% reduction in emissions. This suggests that technological advancements in these countries may contribute to short-term environmental sustainability, but the long-term outcomes are less pronounced than in traditional NATO members. This disparity could be attributed to the varying stages of technological development and adoption between the two groups, as newer members may not yet fully integrate advanced, energy-efficient technologies.

In Group 3, technological innovation unexpectedly exhibits a significant positive impact on emissions in the long run, with a 1% increase in innovation leading to a 12.20% rise in emissions. This counterintuitive result could be driven by the reliance of newer NATO members on industries with high energy consumption and emissions during the early stages of technological development. While Group 1 countries benefit more consistently from green technologies, Group 3’s outcome suggests that these nations may still be in the process of adopting energy-efficient innovations. However, in the short term, technological innovation in all groups, including Group 3, reduces emissions, showing a 3.63% decrease per 1% rise in innovation, which aligns with findings by Yii and Geetha (Reference Yii and Geetha2017). This reflects the immediate benefits of technological advancements, though long-term impacts remain dependent on the nature and maturity of the technologies adopted.

Energy consumption and carbon emissions

Energy consumption is found to be strongly and positively correlated with carbon emissions in traditional NATO member countries (Group 1) over the long term. Specifically, the study reveals that a 1% increase in energy consumption leads to a 109.44% rise in carbon emissions. This finding underscores the significant environmental impact of energy use in these countries, where high dependence on fossil fuels in sectors such as energy production, transportation, and industry exacerbates carbon emissions. Similar results are observed in the short term, where energy consumption continues to contribute to rising emissions, consistent with the conclusions of previous studies by Gokmenoglu et al. (2021), Shahbaz et al. (2013), and Kongkuah (Reference Kongkuah, Yao and Yilanci2022), which highlight the pervasive role of energy use in environmental degradation.

The positive impact of energy consumption on carbon emissions is equally prominent in new NATO member countries (Group 2), where a 1% increase in energy consumption results in a 123.73% rise in emissions over the long term. These findings suggest that the newer NATO members, like their traditional counterparts, remain heavily reliant on fossil fuels, with little progress toward cleaner energy alternatives. In both the short and long term, energy consumption remains a major contributor to emissions, reinforcing NATO’s role in global carbon output, particularly due to the energy-intensive nature of military and industrial activities. This is consistent with previous research, which confirms that fossil fuel use is a primary driver of emissions across various sectors in NATO countries.

In Group 3, which comprises a diverse range of NATO members, energy consumption also exhibits a strong positive correlation with carbon emissions, with a 1% increase in energy consumption leading to an 84.54% rise in emissions. This finding aligns with those from Groups 1 and 2, further illustrating the extensive reliance on fossil fuels across NATO countries. The results indicate that energy consumption remains a central factor driving emissions, with renewable energy investments not yet sufficient to offset the growing energy demands in military and industrial sectors. The consistency of these findings across all groups highlights a critical need for NATO members to transition toward cleaner energy sources to mitigate their contribution to climate change.

Economic growth and carbon emissions

In traditional NATO member countries (Group 1), the long-term impact of economic growth on carbon emissions is relatively insignificant, with a modest 10% increase in emissions for every 1% rise in GDP. This suggests that the influence of economic expansion on emissions diminishes over time, echoing the findings of Adebayo (2022) and Yii & Geetha (Reference Yii and Geetha2017), who argue that the relationship between economic growth and carbon emissions weakens in the long run. However, in the short term, economic growth continues to exert a significant upward pressure on emissions, primarily due to heightened energy consumption and industrial activity that accompany immediate economic expansion.

In contrast, for the new NATO member countries (Group 2), the long-term effect of economic growth on emissions is more pronounced, with a notable 18.44% reduction in carbon emissions for each 1% increase in GDP. This underscores a stronger decoupling of economic growth from carbon emissions compared to Group 1, aligning with the insights of Dong et al. (Reference Dong, He, Li, Mou and Dong2020) and Wang et al. (2011), who emphasize that emerging economies often exhibit significant emissions growth in the early stages of economic expansion. Nevertheless, in the short term, these nations continue to experience considerable increases in emissions, reflecting their reliance on fossil fuels and energy-intensive processes, despite the potential for adopting cleaner technologies in the future.

For Group 3, economic growth exerts the most significant long-term impact on carbon emissions, with a 65.17% increase in emissions for every 1% rise in GDP. This sharp rise illustrates the close link between economic expansion and increased energy use, as seen in both Groups 1 and 2, where growth continues to drive environmental degradation. The analysis suggests that, across all NATO member countries, economic growth remains tightly coupled with fossil fuel consumption and energy-intensive industries, particularly in the short term. The findings highlight the urgent need for decoupling economic growth from fossil fuel dependency to mitigate the adverse environmental impacts associated with development.

Robustness estimates of FMOLS

The robustness test serves to validate the reliability of the long-run outcomes derived from the PMG-ARDL model by employing the FMOLS methodology, as detailed in Table 3. This additional analytical approach offers supplementary confirmation of the significance and stability of the variables influencing carbon emissions within NATO member countries. Consistent findings across these methods underscore the robustness of the initial PMG-ARDL results. This rigorous validation process ensures the credibility and dependability of the conclusions, providing a robust foundation for policy recommendations and avenues for future research.

Table 3. Robustness estimates of FMOLS

Note: ***, ** and * represent statistical significance at 1%, 5%, 10% level respectively.

Diagnostic checks

Once the short-term and long-term impacts of the independent variables on carbon discharges were investigated using the PMG estimator, the accuracy of the model was extensively confirmed. The model’s accuracy was verified through a variety of diagnostic tests, such as the Wooldridge measure for serial correlation, the Breusch-Pagan/Cook-Weisberg test for heteroskedasticity, and the Ramsey RESET assess for the specification of the model. The model’s robustness was confirmed by the absence of heteroskedasticity and serial correlation, as evidenced by the results presented in Table A8.

Conclusion and policy implications

This study explores the collective impact of militarization, industrialization, technological innovation, energy consumption, and economic growth on carbon emissions in NATO countries from 1985 to 2019, using an advanced econometric approach with PMG estimation methods. The findings reveal distinct patterns across traditional NATO members (Group 1), new NATO members (Group 2), and the combined group (Group 3). For Group 1, military expenditure and energy consumption significantly increase carbon emissions, while industrialization and technological innovation help reduce them. In Group 2, military spending and industrialization are associated with a reduction in emissions, but energy consumption and technological innovation lead to an increase. For Group 3, economic growth is a significant driver of emissions, while industrial advancements and selective technological innovations help mitigate them. These results provide policymakers, authorities, and academicians with valuable insights into the specific factors influencing carbon emissions within NATO countries.

To reduce carbon emissions and promote environmental sustainability within NATO countries, legislative bodies in traditional NATO member states (Group 1) should prioritize cutting military spending and regulating activities that contribute to emissions. This includes limiting missile tests, constraining nuclear power, and reducing the deployment of energy-intensive military equipment. By mitigating the environmental impact of military actions, these nations can align their defense strategies with sustainability goals. Additionally, reallocating savings from military expenditures to fund research and development in green technologies will further enhance environmental efforts.

New NATO member states (Group 2) should focus on adopting and developing eco-friendly technologies to lower carbon emissions. Policymakers need to create incentives and regulatory frameworks that discourage the use of harmful technologies and encourage innovation in sustainable practices. Investing in clean energy projects, building renewable energy infrastructure, and improving energy efficiency will support long-term economic growth while enhancing environmental protection.

Across all NATO countries, including Group 1, Group 2, and the combined group (Group 3), integrating renewable energy sources, such as solar, wind, and hydroelectric power, into national energy grids is essential. Policymakers should set ambitious targets for renewable energy use, enforce regulations to reduce reliance on fossil fuels, and create collaborative platforms for sharing expertise and resources. Advancing sustainable industrial practices, such as energy-efficient manufacturing and green supply chains, along with regional cooperation through a NATO Environmental Sustainability Task Force, will enable coordinated efforts to achieve both security and sustainability objectives across all member states.

This study has several limitations. While it distinguishes between traditional and new NATO members, it does not account for intra-group variations, such as differences in energy reliance between countries like the U.S. and Norway. The use of total military expenditure as a proxy for militarization overlooks variations in activities like training, combat, and research, which may have distinct environmental impacts. Technological innovation is measured through patent applications, which may not reflect the actual deployment or effectiveness of green technologies. The exclusion of some NATO countries due to data unavailability limits the comprehensiveness of the findings. Additionally, the study does not evaluate the effectiveness of NATO’s environmental policies, such as the 2021 carbon neutrality pledge. As the focus is primarily on developed, industrialized, and militarized nations, the findings may not be generalizable to underdeveloped or non-militarized countries.

This study highlights several areas for future research to deepen the understanding of the environmental impacts of militarization, industrialization, technological innovation, energy consumption, and economic growth in NATO countries. First, the scope and duration of the analysis should be extended to include all 31 NATO member states. This broader coverage would allow for a more comprehensive examination of regional variations in environmental impacts and provide a nuanced understanding of how these dynamics unfold in different economic and geopolitical contexts. Second, future research should focus on investigating the environmental effects of specific military activities, such as weapons testing, military exercises, and logistics operations. This approach would help identify high-emission practices and develop targeted strategies to mitigate their environmental impact. Additionally, exploring the distinct components of military spending, such as investments in weapons development, personnel wages, and operational expenditures, could provide valuable insights into their varied environmental implications. Third, further studies should assess the potential for NATO countries to transition from fossil fuels to renewable energy in military operations. This includes exploring the feasibility, challenges, and long-term benefits of integrating renewable energy sources like solar, wind, and biofuels into defense strategies, thereby aligning military activities with global sustainability goals. Fourth, researchers should examine the asymmetric effects of technological innovations and industrial activities on carbon emissions in both traditional and new NATO members. By identifying the differing impacts of these advancements, policy recommendations can be better tailored to the unique needs and capacities of each group. Fifth, advanced econometric models, such as the Nonlinear Autoregressive Distributed Lag (NARDL), should be employed to capture the nonlinear and lagged relationships between military spending, industrialization, and environmental impacts. These models can offer deeper insights into the complex dynamics and asymmetric effects that may not be evident in linear analyses. Lastly, sector-specific studies should be conducted to identify high-emission industries within NATO economies. This approach would enable the development of targeted mitigation strategies, such as promoting energy-efficient technologies and adopting cleaner production methods in critical sectors like manufacturing, transportation, and energy production.

Data availability statement

Data are available upon reasonable request from the corresponding author.

Acknowledgements

The authors are sincerely grateful to Md. Harun Ur Rashid, Assistant Professor of Accounting, Department of Economics and Banking, International Islamic University Chittagong, for his valuable guidance throughout the research process. The authors also extend heartfelt thanks to Dr. Anna Klis, Editor of the Agricultural and Resource Economics Review, for her continuous support and assistance during the publication process.

Competing interests

The authors declare no competing interests.

Funding statement

This research received no specific grant from any funding agency, commercial, or not-for-profit sectors.

Appendix

Table A1. List of selected North Atlantic Treaty Organization (NATO) countries

Note: Albania is excluded because the data is unavailable.

Table A2. Descriptive statistics

Table A3. Variance inflation factor

Table A4. Heterogeneity of slope test

Note: ***, ** and * represent statistical significance at 1%, 5%, 10% level respectively.

Table A5. Cross-sectional dependency

Note: ***, ** and * represent statistical significance at 1%, 5%, 10% level respectively.

Table A6. Panel unit root test

Note: ***, ** and * represent statistical significance at 1%, 5%, 10% level respectively.

Table A7. Pedroni and Kao panel cointegration test

Note: (1) ***, ** and * represent statistical significance at 1%, 5%, and 10% levels, respectively. (2) Schwarz information criteria were used to select the optimal lag lengths.

Table A8. Diagnostic test

References

Abbasi, M. A., Parveen, S., Khan, S., and Kamal, M. A. 2020. “Urbanization and energy consumption effects on carbon dioxide emissions: Evidence from Asian-8 countries using panel data analysis.” Environmental Science and Pollution Research 27(15): 1802918043. https://doi.org/10.1007/s11356-020-08262-w Google ScholarPubMed
Ahmed, Z., Zafar, M. W., and Mansoor, S. 2020. “Analyzing the linkage between military spending, economic growth, and ecological footprint in Pakistan: Evidence from cointegration and bootstrap causality.” Environmental Science and Pollution Research 27(33): 4155141567. https://doi.org/10.1007/s11356-020-10076-9 CrossRefGoogle ScholarPubMed
Ang, J. B. 2007. “CO2 emissions, energy consumption, and output in France.” Energy Policy 35(10): 47724778. https://doi.org/10.1016/j.enpol.2007.03.032 CrossRefGoogle Scholar
Asumadu-Sarkodie, S., and Owusu, P. A. 2017. “The causal effect of carbon dioxide emissions, electricity consumption, economic growth, and industrialization in Sierra Leone.” Energy Sources, Part B: Economics, Planning, and Policy 12(1): 3239. https://doi.org/10.1080/15567249.2016.1225135 Google Scholar
Balsalobre-Lorente, D., Shahbaz, M., Roubaud, D., and Farhani, S. (2018. “How economic growth, renewable electricity and natural resources contribute to CO2 emissions?Energy Policy 113, 356367. https://doi.org/10.1016/j.enpol.2017.10.050 Google Scholar
Bildirici, M. 2017. “CO 2 emissions and militarization in G7 countries: Panel cointegration and trivariate causality approaches.” Environment and Development Economics 22(6): 771791. https://doi.org/10.1017/S1355770X1700016X Google Scholar
Bildirici, M. 2019. “CO2 emissions and militarisation in G20 countries: Panel cointegratrion and causality approaches.” International Journal of Oil, Gas and Coal Technology 22(1): 104121.Google Scholar
Brown, R. L., Durbin, J., and Evans, J. M. 1975. “Techniques for Testing the Constancy of Regression Relationships Over Time.” Journal of the Royal Statistical Society Series B: Statistical Methodology 37(2): 149163. https://doi.org/10.1111/j.2517-6161.1975.tb01532.x Google Scholar
Bugden, D. 2022. “Technology, decoupling, and ecological crisis: Examining ecological modernization theory through patent data.” Environmental Sociology 8(2): 228241. https://doi.org/10.1080/23251042.2021.2021604 Google Scholar
Chang, S., Chen, B., and Song, Y. 2023. “Militarization, renewable energy utilization, and ecological footprints: Evidence from RCEP economies.” Journal of Cleaner Production 391, 136298. https://doi.org/10.1016/j.jclepro.2023.136298 Google Scholar
Chiu, Y.-B., and Zhang, W. 2022. “Effects of energy and economic growth on CO2 emissions: What does globalization matter?Environment, Development and Sustainability 24(7). https://doi.org/10.1007/s10668-022-02522-0 Google Scholar
Cobb, C. W., and Douglas, P. H. 1928. “A Theory of Production.” The American Economic Review 18(1): 139165. JSTOR.Google Scholar
Çolak, O., Özuyar, S. E. G., and Bölükbaşı, Ö. F. 2022. “Asymmetric Effects of the Defense Burden on Environmental Degradation: Evidence from NATO Countries.” Sustainability 15(1): 573. https://doi.org/10.3390/su15010573 CrossRefGoogle Scholar
Destek, M. A., and Manga, M. 2021. “Technological innovation, financialization, and ecological footprint: Evidence from BEM economies.” Environmental Science and Pollution Research 28(17): 2199122001. https://doi.org/10.1007/s11356-020-11845-2 CrossRefGoogle ScholarPubMed
Dogan, E., and Turkekul, B. 2016. “CO2 emissions, real output, energy consumption, trade, urbanization and financial development: Testing the EKC hypothesis for the USA.” Environmental Science and Pollution Research 23(2): 12031213. https://doi.org/10.1007/s11356-015-5323-8 Google ScholarPubMed
Dong, J., He, J., Li, X., Mou, X., and Dong, Z. 2020. “The Effect of Industrial Structure Change on Carbon Dioxide Emissions: A Cross-Country Panel Analysis.” Journal of Systems Science and Information 8(1): 116. https://doi.org/10.21078/JSSI-2020-001-16 Google Scholar
Ehigiamusoe, K. U., and Lean, H. H. 2019. “Effects of energy consumption, economic growth, and financial development on carbon emissions: Evidence from heterogeneous income groups.” Environmental Science and Pollution Research 26(22): 2261122624. https://doi.org/10.1007/s11356-019-05309-5 Google ScholarPubMed
EIA. 2015. U.S. Energy-related Carbon Dioxide Emissions, 2014, Washington, DC: US Department of Energy, [Available at] https://www.eia.gov/environment/emissions/carbon/pdf/2014co2analysis.pdf.Google Scholar
Ekwueme, D. C., and Zoaka, J. D. 2020. “Effusions of carbon dioxide in MENA countries: Inference of financial development, trade receptivity, and energy utilization.” Environmental Science and Pollution Research 27(11): 1244912460. https://doi.org/10.1007/s11356-020-07821-5 CrossRefGoogle ScholarPubMed
Erdogan, S. 2021. “Dynamic Nexus between Technological Innovation and Building Sector Carbon Emissions in the BRICS Countries.” Journal of Environmental Management 293, 112780. https://doi.org/10.1016/j.jenvman.2021.112780 Google ScholarPubMed
Erdogan, S., Gedikli, A., Çevik, E. İ., and Öncü, M. A. 2022. “Does military expenditure impact environmental sustainability in developed Mediterranean countries?Environmental Science and Pollution Research 29(21): 3161231630. https://doi.org/10.1007/s11356-021-18226-3 CrossRefGoogle ScholarPubMed
Fisher-Vanden, K., Jefferson, G. H., Liu, H., and Tao, Q. 2004. “What is driving China’s decline in energy intensity?Resource and Energy Economics 26(1): 7797. https://doi.org/10.1016/j.reseneeco.2003.07.002 Google Scholar
Franco, S., Mandla, V. R., and Rao, K. R. M. 2017. “Urbanization, energy consumption and emissions in the Indian context A review.” Renewable and Sustainable Energy Reviews 71, 898907.Google Scholar
Gokmenoglu, K. K., Taspinar, N., and Rahman, M. M. 2021a. “Military expenditure, financial development and environmental degradation in Turkey: A comparison of CO2 emissions and ecological footprint.” International Journal of Finance & Economics 26(1): 986997. https://doi.org/10.1002/ijfe.1831CrossRefGoogle Scholar
Gokmenoglu, K. K., Taspinar, N., and Rahman, M. M. (2021b. “Military expenditure, financial development and environmental degradation in Turkey: A comparison of CO2 emissions and ecological footprint.” International Journal of Finance & Economics 26(1): 986997.Google Scholar
Hooks, G., and Smith, C. L. 2004. “The Treadmill of Destruction: National Sacrifice Areas and Native Americans.” American Sociological Review 69(4): 558575. https://doi.org/10.1177/000312240406900405 Google Scholar
IEA. 2016. ‘Decoupling of global emissions and economic growth confirmed’, International Energy Agency, Paris, [Available at] https://www.iea.org/newsroom/news/2016/march/decoupling-of-global-emissions-and-economic-growthconfirmed.Html.Google Scholar
IEA. 2022. International Energy Agency (IEA). Global Energy Review 2021: CO Emissions 2022. Available online: https://www.iea.org/reports/global-energy-review-2021/CO-emissions (accessed on 14 April 2022).Google Scholar
Jorgenson, A. K. 2007. “Does foreign investment harm the air we breathe and the water we drink? A cross-national study of carbon dioxide emissions and organic water pollution in less-developed countries, 1975 to 2000.” Organization & Environment 20(2): 137156.Google Scholar
Jorgenson, A. K., and Clark, B. 2009. “The Economy, Military, and Ecologically Unequal Exchange Relationships in Comparative Perspective: A Panel Study of the Ecological Footprints of Nations, 1975–2000.” Social Problems 56(4): 621646. https://doi.org/10.1525/sp.2009.56.4.621 CrossRefGoogle Scholar
Jorgenson, A. K., and Clark, B. 2011. “Societies consuming nature: A panel study of the ecological footprints of nations, 1960–2003.” Social Science Research 40(1): 226244.10.1016/j.ssresearch.2010.09.004CrossRefGoogle Scholar
Jorgenson, A. K., and Clark, B. 2012. “Are the Economy and the Environment Decoupling? A Comparative International Study, 1960–2005.” American Journal of Sociology 118(1): 144. https://doi.org/10.1086/665990 Google Scholar
Jorgenson, A. K., Clark, B., and Kentor, J. 2010. “Militarization and the Environment: A Panel Study of Carbon Dioxide Emissions and the Ecological Footprints of Nations, 1970–2000.” Global Environmental Politics 10(1): 729. https://doi.org/10.1162/glep.2010.10.1.7 Google Scholar
Jorgenson, A. K., Clark, B., Thombs, R. P., Kentor, J., Givens, J. E., Huang, X., El Tinay, H., Auerbach, D., and Mahutga, M. C. 2023. “Guns versus Climate: How Militarization Amplifies the Effect of Economic Growth on Carbon Emissions.” American Sociological Review 88(3): 418453. https://doi.org/10.1177/00031224231169790 CrossRefGoogle Scholar
Kao, C. 1999. “Spurious regression and residual-based tests for cointegration in panel data. Journal of Econometrics 90(1): 144.Google Scholar
Kongkuah, M., Yao, H., and Yilanci, V. 2022. “The relationship between energy consumption, economic growth, and CO2 emissions in China: The role of urbanisation and international trade.” Environment, Development and Sustainability 24(4): 46844708. https://doi.org/10.1007/s10668-021-01628-1 CrossRefGoogle Scholar
Konuk, F., Kaya, E., Akpınar, S., and Yıldız, Ş. (2023. The Relationship Between Military Expenditures, Financial Development and Environmental Pollution in G7 Countries. Journal of the Knowledge Economy 15, 10871102. https://doi.org/10.1007/s13132-023-01122-1 CrossRefGoogle Scholar
Lean, H. H., and Smyth, R. 2010. “CO2 emissions, electricity consumption and output in ASEAN.” Applied Energy 87(6): 18581864. https://doi.org/10.1016/j.apenergy.2010.02.003 CrossRefGoogle Scholar
Lee, C. T., Hashim, H., Ho, C. S., Fan, Y. V., and Klemeš, J. J. 2017. “Sustaining the low-carbon emission development in Asia and beyond: Sustainable energy, water, transportation and low-carbon emission technology.” Journal of Cleaner Production 146, 113. https://doi.org/10.1016/j.jclepro.2016.11.144 CrossRefGoogle Scholar
Lee, K.-H., and Min, B. 2015. “Green R&D for eco-innovation and its impact on carbon emissions and firm performance.” Journal of Cleaner Production 108, 534542. https://doi.org/10.1016/j.jclepro.2015.05.114 Google Scholar
Lengefeld, M. R., and Smith, C. L. 2013. “Nuclear shadows: Weighing the environmental effects of militarism, capitalism, and modernization in a global context, 2001-2007.” Human Ecology Review 20(1): 1125. JSTOR.Google Scholar
Li, D., Cao, L., Zhou, Z., Zhao, K., Du, Z., and Chen, Y. 2023. “Correlating CO2 emissions of cities with the inter-city carbon compensation mechanism: A regional perspective in the middle reaches of the Yangtze River (MRYR), China.” Environment, Development and Sustainability 26, 91859216. https://doi.org/10.1007/s10668-023-03089-0 Google Scholar
Li, K., and Lin, B. 2015. “Impacts of urbanization and industrialization on energy consumption/CO2 emissions: Does the level of development matter?Renewable and Sustainable Energy Reviews 52, 11071122. https://doi.org/10.1016/j.rser.2015.07.185 CrossRefGoogle Scholar
Mol, A. P. J., and Spaargaren, G. 2000. “Ecological modernisation theory in debate: A review.” Environmental Politics 9(1): 1749. https://doi.org/10.1080/09644010008414511 Google Scholar
NATO. 2022. North Atlantic Treaty Organization (NATO). Brussels Submit Communiqué 2021. Available online: https://www.nato.int/cps/en/natohq/news_185000.htm (accessed on 17 December 2022).Google Scholar
Odehnal, J. 2015. “Military Expenditures and Free-Riding in NATO.” Peace Economics, Peace Science and Public Policy 21(4): 479487. https://doi.org/10.1515/peps-2015-0015 CrossRefGoogle Scholar
Odehnal, J., and Neubauer, J. 2020. “Economic, Security, and Political Determinants of Military Spending in NATO Countries.” Defence and Peace Economics 31(5): 517531. https://doi.org/10.1080/10242694.2018.1544440 CrossRefGoogle Scholar
Olivier, J. G., Schure, K. M., and Peters, J. 2017. “Trends in global CO2 and total greenhouse gas emissions.” PBL Netherlands Environmental Assessment Agency 5, 111.Google Scholar
Omri, A., and Hadj, T. B. 2020. “Foreign investment and air pollution: Do good governance and technological innovation matter?Environmental Research 185, 109469.CrossRefGoogle ScholarPubMed
Pata, U. K. 2018. “Renewable energy consumption, urbanization, financial development, income and CO2 emissions in Turkey: Testing EKC hypothesis with structural breaks.” Journal of Cleaner Production 187, 770779. https://doi.org/10.1016/j.jclepro.2018.03.236 Google Scholar
Pedroni, P. 1999. “Critical Values for Cointegration Tests in Heterogeneous Panels with Multiple Regressors.” Oxford Bulletin of Economics and Statistics 61(s1): 653670. https://doi.org/10.1111/1468-0084.61.s1.14 Google Scholar
Pesaran, M. H. 2004. General Diagnostic Tests for Cross Section Dependence in Panels (SSRN Scholarly Paper 572504). https://doi.org/10.2139/ssrn.572504 Google Scholar
Pesaran, M. H. 2007. “A simple panel unit root test in the presence of cross-section dependence.” Journal of Applied Econometrics 22(2): 265312. https://doi.org/10.1002/jae.951 CrossRefGoogle Scholar
Pesaran, M. H., Shin, Y., and Smith, R. J. 2001. “Bounds testing approaches to the analysis of level relationships.” Journal of Applied Econometrics 16(3): 289326. https://doi.org/10.1002/jae.616 CrossRefGoogle Scholar
Pesaran, M. H., Shin, Y., and Smith, R. P. 1999. “Pooled Mean Group Estimation of Dynamic Heterogeneous Panels.” Journal of the American Statistical Association 94(446): 621634. https://doi.org/10.1080/01621459.1999.10474156 CrossRefGoogle Scholar
Pesaran, M. H., and Smith, R. 1995. “Estimating long-run relationships from dynamic heterogeneous panels.” Journal of Econometrics 68(1): 79113. https://doi.org/10.1016/0304-4076(94)01644-F Google Scholar
Rafique, M. Z., Li, Y., Larik, A. R., and Monaheng, M. P. 2020. “The effects of FDI, technological innovation, and financial development on CO2 emissions: Evidence from the BRICS countries.” Environmental Science and Pollution Research 27(19): 2389923913. https://doi.org/10.1007/s11356-020-08715-2 Google ScholarPubMed
Riaz, M. H., Alam, M., Ali, A., Ahmed, Z., and Raihan, A. 2025. “The role of growth, energy, and resource utilization in ecological footprints: Navigating BRICS + environmental challenges.” Letters in Spatial and Resource Sciences 18(1): 8. https://doi.org/10.1007/s12076-025-00403-4 Google Scholar
Samargandi, N. 2017. “Sector value addition, technology and CO2 emissions in Saudi Arabia.” Renewable and Sustainable Energy Reviews 78, 868877. https://doi.org/10.1016/j.rser.2017.04.056 CrossRefGoogle Scholar
Saud, S., Chen, S., Haseeb, A., and Sumayya. 2020. “The role of financial development and globalization in the environment: Accounting ecological footprint indicators for selected one-belt-one-road initiative countries.” Journal of Cleaner Production 250, 119518. https://doi.org/10.1016/j.jclepro.2019.119518 CrossRefGoogle Scholar
Shahbaz, M., Raghutla, C., Song, M., Zameer, H., and Jiao, Z. 2020. “Public-private partnerships investment in energy as new determinant of CO2 emissions: The role of technological innovations in China.” Energy Economics 86, 104664. https://doi.org/10.1016/j.eneco.2020.104664 CrossRefGoogle Scholar
SIPRI. 2021. Stockholm International Peace and Research Institute. SIPRI Yearbook 2021: Armaments, Disarmament and International Security 2021. Available online: https://www.sipri.org/yearbook/2021 (accessed on 14 April 2022).Google Scholar
Smith, C. L., Hooks, G., and Lengefeld, M. 2024. “Linking the treadmills of production and destruction to disproportionate carbon emissions.” Npj Climate Action 3(1): 72. https://doi.org/10.1038/s44168-024-00156-8 CrossRefGoogle Scholar
Smith, C. L., and Lengefeld, M. R. 2020. “The Environmental Consequences of Asymmetric War: A Panel Study of Militarism and Carbon Emissions, 2000–2010.” Armed Forces & Society 46(2): 214237. https://doi.org/10.1177/0095327X19832615 CrossRefGoogle Scholar
Tahir, T., Luni, T., Majeed, M. T., and Zafar, A. 2021. “The impact of financial development and globalization on environmental quality: Evidence from South Asian economies.” Environmental Science and Pollution Research 28(7): 80888101. https://doi.org/10.1007/s11356-020-11198-w CrossRefGoogle ScholarPubMed
Uddin, M., Rashid, Md. H. U., Ahamad, S., and Ehigiamusoe, K. U. 2023. “Impact of militarization, energy consumption, and ICT on CO2 emissions in G20 countries. Environment, Development and Sustainability 26, 1177111793. https://doi.org/10.1007/s10668-023-03483-8 CrossRefGoogle Scholar
Ullah, S., Andlib, Z., Majeed, M. T., Sohail, S., and Chishti, M. Z. 2021. “Asymmetric effects of militarization on economic growth and environmental degradation: Fresh evidence from Pakistan and India.” Environmental Science and Pollution Research 28(8): 94849497. https://doi.org/10.1007/s11356-020-11142-y Google ScholarPubMed
Wang, L., Chang, H.-L., Rizvi, S. K. A., and Sari, A. 2020. “Are eco-innovation and export diversification mutually exclusive to control carbon emissions in G-7 countries?Journal of Environmental Management 270, 110829. https://doi.org/10.1016/j.jenvman.2020.110829 CrossRefGoogle ScholarPubMed
Xinmin, W., Hui, P., Hafeez, M., Aziz, B., Akbar, M. W., and Mirza, M. A. 2020. “The nexus of environmental degradation and technology innovation and adoption: An experience from dragon.” Air Quality, Atmosphere & Health 13(9): 11191126. https://doi.org/10.1007/s11869-020-00868-w Google Scholar
Yii, K.-J., and Geetha, C. 2017. “The Nexus between Technology Innovation and CO 2 Emissions in Malaysia: Evidence from Granger Causality Test.” Energy Procedia 105, 31183124. https://doi.org/10.1016/j.egypro.2017.03.654 CrossRefGoogle Scholar
York, R. 2008. “De-Carbonization in Former Soviet Republics, 1992–2000: The Ecological Consequences of De-Modernization.” Social Problems 55(3): 370390. https://doi.org/10.1525/sp.2008.55.3.370 Google Scholar
York, R., Rosa, E. A., and Dietz, T. 2003. “Footprints on the Earth: The Environmental Consequences of Modernity.” American Sociological Review 68(2): 279300. https://doi.org/10.1177/000312240306800205 Google Scholar
Zafar, A., Ullah, S., Majeed, M. T., and Yasmeen, R. 2020. “Environmental pollution in Asian economies: Does the industrialisation matter?OPEC Energy Review 44(3): 227248. https://doi.org/10.1111/opec.12181 CrossRefGoogle Scholar
Figure 0

Figure 1. Visualization of the association between chosen variables.

Figure 1

Table 1. Variables’ description and data source

Figure 2

Figure 2. Flow chart of the analytical techniques employed in the research.

Figure 3

Table 2. Parameter estimation of panel PMG models for all groups

Figure 4

Table 3. Robustness estimates of FMOLS

Figure 5

Table A1. List of selected North Atlantic Treaty Organization (NATO) countries

Figure 6

Table A2. Descriptive statistics

Figure 7

Table A3. Variance inflation factor

Figure 8

Table A4. Heterogeneity of slope test

Figure 9

Table A5. Cross-sectional dependency

Figure 10

Table A6. Panel unit root test

Figure 11

Table A7. Pedroni and Kao panel cointegration test

Figure 12

Table A8. Diagnostic test