Geopolitical Risk and National Green Economic Efficiency: Evidence from G20 Member Countries

Abstract

This study investigates how geopolitical risk shaped the green economic efficiency (GEE) of 19 countries in the G20 group from 2000 to 2022. Using the Super-SBM model, we construct a cross-country measure of GEE and empirically examine both its determinants and underlying mechanisms. The results show that rising geopolitical risk significantly undermines GEE, indicating that external uncertainty disrupts countries’ ability to balance economic growth with environmental performance. Mechanism analysis reveals that geopolitical tensions heighten energy security concerns, leading to increased fossil fuel consumption, and trigger exchange rate depreciation to decrease green economic efficiency. Moreover, foreign direct investment mitigates the adverse effects of geopolitical risk by facilitating technology spillovers and capital inflows. Moreover, geopolitical risks have different impacts on the efficiency of a country’s green economy, varying across levels such as the country’s economic development level, resource endowment, and trade openness. The findings highlight geopolitical risk as a constraint on global green transition. Policymakers should strengthen energy source diversity, stabilize exchange rate environments, and promote FDI to enhance national resilience. Building institutional capacity is essential in sustaining green economic efficiency under rising geopolitical uncertainty.

1. Introduction

As global economies expand, the negative impact of environmental pollution on various countries has become increasingly prominent, and governments around the world have gradually realized that addressing climate change is crucial for the development of their nations and even regions [1]. Under the framework of global climate governance, countries have begun to promote energy transition and reduce greenhouse gas emissions, but in the complex international environment, achieving green transition and sustainable development still faces many difficulties. In recent years, the global political and economic landscape has become increasingly turbulent, with geopolitical risks continuing to rise due to armed conflicts, trade frictions, supply chain crises, and anti-globalization trends [2,3,4,5]. These external uncertainties are reshaping the global governance landscape and have already made the efforts of various countries to achieve sustainable development goals more complex [6]. Most existing studies focus on the factors influencing green transition and green economic development within individual economies, while research on broader cross-country levels is scarce [7]. Therefore, in order to continuously promote the sustainable development of the global economy and in the context of increasing geopolitical risks, it is particularly important to explore the factors influencing the efficiency of green economies in different countries.

Green economic efficiency (GEE) is a comprehensive efficiency indicator based on production frontier theory. It is used to measure a country’s ability to achieve the goal of maximizing economic and ecological benefits while minimizing resource inputs and environmental pollution. It is widely used to assess a country’s progress in green transition and sustainable economic development. Unlike traditional economic efficiency indicators that only focus on production inputs and economic outputs, the GEE systematically incorporates factor inputs, desirable outputs, and undesirable outputs. This allows it to comprehensively reflect the coordinated development of the economy and the environment, and it serves as a core standard for assessing a country’s sustainable development performance. It reflects the use of energy, labour, and capital, as well as the actions taken to mitigate climate change. In recent years, the increase in geopolitical risks has had a significant negative impact on climate and energy policies [8,9,10,11]. Moreover, the rise in geopolitical risks has reduced the effectiveness of multinational cooperation and increased the costs of international coordination, thereby undermining climate cooperation [12,13]. Although we have already understood the impact of geopolitical risks on climate, such specific studies (especially those targeting the efficiency of the green economy) are rarely mentioned. After reviewing a large amount of the literature, we found that existing research on green economy efficiency remains highly fragmented [14,15]. In EU countries, in order to achieve green economic growth, the government will increase the degree of economic freedom to promote green economic development [16]. In APEC countries, researchers have found that economic policy uncertainty has adverse effects on economic sustainability and energy efficiency [17]. However, existing research has not yet extended to a broader cross-country context, nor has it explored the mechanisms affecting global green economic efficiency. Given the insufficient academic research on national green economy efficiency, our study aims to fill this gap by exploring its potential impact mechanisms from multiple perspectives.

To address these gaps, this study uses data from 19 G20 countries (excluding the European Union) from 2000 to 2022 to examine the impact of geopolitical risk on green economic efficiency, answering the following questions:

(1)

Does geopolitical risk significantly affect green economic efficiency across countries?

(2)

Through which mechanisms does this influence occur?

(3)

Why might developed and emerging economies respond differently to geopolitical shocks?

The G20 economies are particularly relevant because they represent major global emitters, key geopolitical actors, and diverse institutional and developmental contexts. We employ the Super-SBM model to measure GEE, which provides more accurate efficiency evaluations than traditional DEA methods. We then analyze the direct impact of geopolitical risk, explore its transmission channels, fossil-fuel consumption and exchange rate fluctuations, and examine the moderating role of foreign direct investment.

This study makes three main contributions. First, it is the first to apply the Super-SBM model to measure the green economic efficiency of G20 countries, offering a more precise cross-country comparison and provides empirical evidence that rising geopolitical risk significantly reduces GEE among G20 members. Second, it identifies heterogeneous effects across developed and emerging economies, suggesting that institutional soundness and development levels shape countries’ responses to geopolitical shocks. Third, it reveals two key mechanisms—energy security-driven increases in fossil fuel consumption and exchange rate depreciation—through which geopolitical risk undermines GEE, while also showing that foreign direct investment can mitigate these negative effects.

Overall, geopolitical risks have become an important external constraint on global green transformation. Understanding their impact on green economic efficiency is crucial for policymakers seeking to balance energy security, economic stability, and environmental sustainability in an increasingly uncertain world.

The remainder of this paper is organized as follows: Section 2 reviews the literature and proposes hypotheses; Section 3 describes the research design, including the empirical model, sample selection, and variable definitions; Section 4 presents a series of empirical results; and Section 5 discusses the conclusions and policy implications of this study.

2. Literature Review and Theoretical Hypothesis

2.1. Geopolitical Risks and Green Economic Efficiency

Green economic efficiency reflects a country’s capacity to maximize economic output while prioritizing environmental protection and climate action, serving as a vital indicator of sustainable development [18,19]. In recent years, the continuous intensification of anti-globalization and geopolitical frictions has put economic development in a difficult position, making the external environment for green transition increasingly complex and subjecting green economic efficiency to multiple severe impacts. Against this backdrop, the fulfilment of environmental protection commitments made by various countries is facing numerous obstacles.

The resource dependency theory suggests that if a country is highly dependent on external resources for energy or key technologies, an increase in geopolitical risks will heighten uncertainty, thereby hindering the green transition process. Zeraibi et al. (2025a) suggest a negative correlation between geopolitical risk and renewable energy [20], indicating that geopolitical tensions may inhibit investment in clean energy. For specific industries, manufacturing remains the main driver of economic activity in G20 countries. Previous research reports have indicated a U-shaped relationship between manufacturing agglomeration and green economic efficiency: moderate agglomeration may help reduce environmental pollution, with industrial structure playing a crucial role as an important intermediary channel [21,22]. Therefore, promoting renewable energy to support the green transition of industrial structures is crucial. However, the intensification of geopolitical risks may cause external shocks to the industrial structure, thereby affecting its transformation process. In addition, geopolitical risks can also affect the efficiency of the green economy by influencing financial instruments. A study by Mertzanis et al. (2025) on the cross-border financial market indicates that the rise in geopolitical risk has, to some extent, driven the increase in green bond issuance [23]. Further research has found that geopolitical risk has a negative impact on the prices of green bonds [24]. Research on BRICS countries also indicates a connection between green investment and green economic growth [25]. Because green bonds serve as an important tool for attracting green investment, changes in this market will affect the scale of green investment, which is a key driver of green economic growth. Geopolitical risks may influence the efficiency of the green economy by shaping the deployment of clean energy, industrial restructuring, and the role of green financial markets.

From an institutional perspective, the rise in geopolitical risks is often accompanied by greater policy uncertainty. Institutional theory suggests that geopolitical instability can undermine the continuity and effectiveness of policy implementation. Evidence from the United States indicates that the uncertainty of climate policies is associated with a decrease in the level of green innovation [26], but the broader empirical situation is more complex. Evidence from China shows that economic policy uncertainty is positively correlated with green economic efficiency, with market mechanisms playing an important regulatory role [27]. These studies illustrate that the relationship between geopolitical risks and green economic development varies significantly across countries with distinct institutional backgrounds. This stems from the role the government plays in the market, which is determined by the institutional framework. Wang et al. (2020) used data from G20 economies to show that governments often mobilize additional resources to address external uncertainties and support sustainable energy development in the name of energy security [28].

Overall, the resource dependence theory and institutional theory indicate that the intensification of geopolitical risks hinders a country’s access to external resources and exacerbates institutional pressures caused by policy uncertainty. Consequently, this disrupts the development of national industries and impedes the green transformation, ultimately proving detrimental to green economic efficiency.

Based on this, we propose the following hypothesis:

H1. 

Geopolitical risk has a negative impact on the green economic efficiency of countries.

2.2. Geopolitical Risk, Foreign Direct Investment and Green Economic Efficiency

Since the era of globalization, attracting foreign direct investment has become an important means for many countries to promote economic development. Through this funding, host countries can develop infrastructure and promote international cooperation [29]. As sustainable development gains global prominence, governments increasingly emphasize achieving a win–win situation for both the environment and the economy when attracting investment. Therefore, many scholars have begun to explore the mechanisms by which foreign direct investment contributes to achieving sustainable development goals. Existing research indicates that foreign direct investment affects a country’s energy consumption through various channels, including influencing corporate expansion, technological development, and market distribution, thereby impacting energy demand [30,31]. However, in recent years, the frequent occurrence of geopolitical conflicts has brought new uncertainties and challenges to the cross-border flow of foreign direct investment. Therefore, it is necessary to explore the impact mechanism of foreign direct investment on the green economic efficiency of various countries in this context.

First, the global value chain theory indicates that the involvement of foreign direct investment can enable host countries to attract capital, create job opportunities, and achieve technology sharing, thereby allowing regions with uneven development to also benefit from these advancements [32,33,34]. Shinwari et al. (2024) found that foreign direct investment inflows to countries along the “Belt and Road” initiative have a positive impact on energy consumption and promote the development of a green economy [35]. Some scholars believe that the capital and technological advantages brought about by foreign investment help developing countries achieve sustainable development goals and fulfil their environmental governance commitments. For example, some studies suggest that foreign direct investment can reduce the environmental costs behind economic development [36]. Empirical analysis of East Asian economies has also found that foreign direct investment has a positive impact on the green growth process [37]. From the perspective of the pollution halo hypothesis, the technological advancements brought about by the increase in foreign direct investment are beneficial for promoting the development of green technologies in the host country [38]. These studies provide positive evidence for understanding the impact of foreign direct investment on the development of a country’s green economy.

However, there are also some studies that present the opposite view. Some scholars believe that some developed countries may transfer high-pollution and high-energy-consuming production processes to host countries with lower environmental standards through foreign investment, thereby evading their own environmental regulatory responsibilities. In this context, foreign direct investment may have adverse effects on the sustainable development of the host country. The literature reveals a correlation between the increase in foreign direct investment and the rise in carbon emissions. For example, Abbas et al. (2023) [39] found in their study of South Africa that for every 1% increase in foreign direct investment, local per capita carbon emissions would increase by an average of 0.03%. This supports the “pollution haven” hypothesis and indicates that its impact on carbon emissions can also affect surrounding areas [39]. Clearly, the increase in carbon emissions will harm a country’s green economic development.

By reviewing this literature, we found that the researchers’ opposing views mainly stem from differences in institutional quality, environmental regulation intensity, industrial structure, and technology absorption capacity among different countries. Institutional quality is considered one of the key factors influencing the shape of the Environmental Kuznets Curve (EKC). In emerging economies, due to the uncertainty of environmental policies and regulations, the impact of foreign direct investment on green economic efficiency is more complex. Countries with sound institutions, strict regulations, and strong technological absorption capabilities are more likely to promote green economic growth through FDI [40,41]. Therefore, introducing FDI facilitates the development of green capital and technology, effectively mitigating the adverse impacts of geopolitical shocks.

Based on this, we propose the following hypothesis:

H2. 

Foreign direct investment weakens the impact of geopolitical risk on the efficiency of the green economy.

2.3. Geopolitical Risks, Fossil Energy Consumption and Green Economic Efficiency

The consumption of fossil energy has long been regarded as an important indicator of a country’s energy development level and economic development stage, as well as a significant factor influencing the efficiency of the green economy [42]. As the global climate change issue becomes increasingly severe, the international community has reached a broad consensus on reducing dependence on fossil fuels, promoting energy structure transformation, and achieving sustainable development goals [43]. However, the fundamental role of fossil energy in economic growth and social stability often creates path dependence and institutional constraints, preventing countries from rapidly advancing in their green transitions. Numerous studies indicate that factors such as the degree of trade openness, the level of technological innovation, tax policies, and government subsidies can significantly impact a country’s energy consumption structure, and these factors are closely related to the complex and ever-changing geopolitical environment [44,45,46,47]. With deepening globalization, cross-border industrial and energy cooperation has strengthened, leaving many countries highly dependent on external energy supplies. Consequently, geopolitical risks have become a critical external factor affecting energy structures and green economic efficiency [48,49].

Existing research generally suggests that geopolitical risks affect the performance of green development by influencing energy security, thereby forcing countries to adjust their energy structures. First, the rise in geopolitical risk will increase the uncertainty of energy supply. For countries that heavily rely on imported oil and natural gas, external conflicts, policy changes, and transportation disruptions directly threaten energy security [50,51]. Energy security theory suggests that when the external environment is unstable, countries often adopt a “conservative energy strategy”, which involves increasing their domestic fossil energy reserves and enhancing the production capacity of traditional fossil energy to ensure the stability of economic operations. This reliance on traditional fuels stems from three factors: renewable energy requires long construction periods and high initial investments, making it difficult to deploy quickly during crises; geopolitical risks raise capital costs and exchange rate volatility, complicating green financing; and fossil fuels provide the immediate, controllable reserves needed when external uncertainties rise. While this “conservative energy strategy” alleviates short-term energy crises, the resulting surge in fossil fuel consumption ultimately hinders the improvement of green economic efficiency.

Moreover, geopolitical tensions may also indirectly affect the green energy transition by hindering international energy cooperation. Trade theory indicates that international conflicts and political tensions increase the costs of cross-border transactions, weaken energy cooperation and green technology exchange between countries, and slow down the global dissemination of green technology. Shen and Yang (2024) believe that the intensification of geopolitical conflicts often triggers protectionist responses in the energy sector [52]. These responses include implementing stricter controls on the export of key renewable energy technologies and taking measures in cross-border cooperation to prevent strategic resources from falling under external control. Overall, geopolitical risks have increased trade costs by causing policy uncertainty, limiting countries’ access to green technologies and green resources, thereby forcing them to rely on traditional fossil fuels to ensure national energy security. This reliance on traditional fossil fuels ensures short-term national energy security but severely negatively impacts green economic efficiency.

Based on this, we make the following hypothesis:

H3. 

Geopolitical risk increases fossil energy consumption and reduces the green economic efficiency of countries.

2.4. Geopolitical Risks, Exchange Rates and Green Economic Efficiency

As an important macroeconomic indicator, the exchange rate has a significant impact on a country’s green economic growth. Exchange rate fluctuations affect almost all industries, influencing economic operations and the green transformation process through various channels. On the one hand, exchange rate fluctuations are closely related to the relative prices of import and export goods: a depreciation of the domestic currency will increase import costs, while an appreciation will have the opposite effect. These price changes can affect the import costs of clean energy technologies and equipment, as well as the expectations of businesses and consumers regarding energy prices, thereby impacting the development of the green economy [53]. On the other hand, exchange rate fluctuations can also affect cross-border investment and capital flows, thereby impacting the financing environment for domestic green projects. In today’s increasingly complex global political and economic landscape, geopolitical risks have introduced significant uncertainty to exchange rates. Yilmazkuday (2025) [54] conducted a study on 35 countries and found that geopolitical tensions often lead to greater currency depreciation in economies that are deeply integrated into global value chains. Consequently, the exchange rates of highly open economies are more susceptible to geopolitical events, subjecting their green economic development to greater uncertainty [54].

Exchange rate fluctuations caused by geopolitical risks primarily affect the efficiency of the green economy through capital flows and price effects. When geopolitical tensions escalate, conflicts intensify, or the risk of sanctions increases, international capital often swiftly shifts to safe-haven assets, leading to greater depreciation pressure on the currencies of some emerging market countries or those with higher risk exposure [55,56]. This currency depreciation significantly inflates import costs, making it excessively expensive to introduce renewable energy equipment and other green production factors, thereby hindering clean technology diffusion and industrial upgrading. In addition, significant fluctuations in exchange rates can indirectly affect the efficiency of the green economy by impacting the government’s fiscal capacity [57]. On one hand, continuous currency depreciation will increase the burden of government debt denominated in foreign currency, and the government may need to consume more financial resources to stabilize the exchange rate and repay the debt. On the other hand, the increased fiscal pressure will weaken the government’s subsidies and support for the green industry, affecting investments in clean energy projects and environmental protection infrastructure [58].

As mentioned earlier, the resource dependency theory indicates that countries reliant on energy imports are more susceptible to exchange rate fluctuations under geopolitical shocks. The institutional theory emphasizes that the weaker the quality of institutions, the more susceptible exchange rates are to external shocks. The trade theory points out that exchange rate fluctuations increase capital costs, thereby hindering the acquisition of green technologies. Within this framework, geopolitical risks lead to exchange rate fluctuations and currency depreciation, which in turn result in rising trade costs. Additionally, these risks cause fluctuations in energy costs, preventing countries from acquiring the necessary green technologies and resources, which ultimately impairs green economic efficiency.

Based on this, we make the following hypothesis:

H4. 

Geopolitical risks cause currency depreciation, which in turn reduces the green economic efficiency of the country.

Figure 1 shows the theoretical framework of this paper.

3. Research Design

3.1. Data and Variables

Following Appiah-Otoo et al. (2023) and Luo et al. (2024) [59,60], we selected G20 member countries as they span a vast global range and are significant global trading nations. They account for a significant proportion of global greenhouse gas emissions, play an important role in geopolitical and global governance, and exhibit notable differences in institutional aspects. Therefore, this study selected 19 G20 countries as a sample (excluding the European Union) to examine the impact of geopolitical risks on the efficiency of the green economy from 2000 to 2022 [59,60]. The data are primarily sourced from the World Bank’s World Development Indicators (WDI) database, which is widely recognized for its high credibility and extensive application in academic research. During the data preprocessing stage, we used the 3-year mean imputation method to fill in a small number of missing values in the raw data to ensure data continuity. Subsequently, to reduce the potential bias that extreme outliers might bring, all continuous variables were winsorized at the 1st and 99th percentiles. Additionally, to ensure numerical stability and facilitate the interpretation of regression coefficients, certain covariates were scaled accordingly. Finally, any observations that still had missing values and could not be imputed were excluded from the final sample. The final sample includes 437 country observations.

3.1.1. Explanatory Variable: Geopolitical Risk (GPRH)

The core independent variable is geopolitical risk, measured using the Geopolitical Risk Historical Index (GPRH) constructed by Caldara and Iacoviello (2022) [61]. The index quantifies geopolitical risk by counting the frequency of news reports involving events such as war threats, terrorist attacks, and diplomatic crises. The higher the value, the higher the level of geopolitical tension and uncertainty. The original data is published on a monthly frequency. In order to match the annual panel data, we take the arithmetic average of the monthly index for each country to obtain the annual Geopolitical Risk Index from 2000 to 2022. The index ensures cross-country comparability and captures economic and environmental uncertainties stemming from global geopolitical events, including conflicts, terrorist attacks, and diplomatic frictions. Figure 2 shows the trend in the GPRH of the sample countries from 2000 to 2022.

3.1.2. Explained Variable: Green Economic Efficiency (GEE)

In this paper, green economic efficiency (GEE) is used as the dependent variable. In view of the limitations of radial and angular assumptions in the traditional data envelopment analysis (DEA) model, it is difficult to account for slack variables and undesirable outputs. Therefore, following Tone (2002) [62], we use the super-slack-based measure (super-SBM) model including undesirable output to evaluate the green economic efficiency of G20 countries.

Suppose there are n decision-making units (DMUs). In this study, each DMU is defined as a country-year observation. Each DMU is characterized by q input variables $x\in R_{+}^q$, ${\mathrm{u}}_1$ desirable output variables $y^g\in R_{+}^{u1}$, and $u_2$ undesirable output variables $y^b\in R_{+}^{u2}$. The input matrix is defined as $\left.X=\left[x_1,\cdots ,x_n\right.\right]\in R^{q\times n}>0$, the desirable output matrix as $\left.Y^g=\left[\begin{array}{c}{y}_1^g,\cdots ,y_n^g\end{array}\right.\right]\in R^{u_1\times n}>0$, and the undesirable output matrix as $\left.Y^b=\left[\begin{array}{c}{y}_1^b,\cdots ,y_n^b\end{array}\right.\right]\in R^{u_2\times n}>0$. Based on the production possibility set $P=\left\{\right(x, y^g, y^b\left)\right|x\ge X\lambda , y^g\le Y^g\lambda ,{ y}^b\ge Y^b\lambda , \lambda \ge 0\}$, the efficiency score in the SBM framework is formulated as follows:

$$\rho =min{\displaystyle \frac{1-\frac{1}{q}{\sum }_{i=1}^q\frac{s_i^{-}}{x_{i0}}}{1+\frac{1}{u_1+u_2}({\sum }_{r=1}^{u_1}\frac{s_r^g}{y_{r0}^g}+{\sum }_{l=1}^{u_2}\frac{s_l^b}{y_{l0}^b})}}$$

$$x_0=X\lambda +s^{-}$$

$$y_0^g=Y^g\lambda -s^g$$

$$y_0^b=Y^b\lambda +s^b$$

$$s^{-}\ge 0,s^g\ge 0,s^b\ge 0,\lambda \ge 0$$

In the above formula, $x_{i0}$, $y_{r0}^g$ and $y_{l0}^b$ represent the input items, desirable output and undesirable output of the decision-making unit; $s^{-}$, $s^g$ and $s^b$ denote the slack variables for input, desirable output and undesirable output, respectively; $\lambda$ is the weight vector; and the subscript “0” denotes the unit under evaluation. When $\rho$ = 1, $s^{-}$, $s^g$ and $s^b$ are all 0, the decision-making unit is efficient. When $\rho$ < 1, it indicates that the decision-making unit has efficiency loss, and the input and output need to be further optimized at this time. However, under the SBM model with undesirable outputs, multiple decision-making units may be evaluated as efficient simultaneously, with all efficiency scores equal to 1, making it difficult to distinguish and rank them. If multiple decision-making units are evaluated as efficient in the measurement results, the Super-SBM model with undesirable output needs to be used to solve the problem. Super-SBM can clearly identify various factors affecting the efficiency value, so as to improve the accuracy and rationality of efficiency measurement. The Super-SBM model is formulated as follows:

$${\rho }^{*}=min{\displaystyle \frac{\frac{1}{q}{\sum }_{i=1}^q\frac{{\overline{x}}_i}{x_{i0}}}{\frac{1}{u_1+u_2}({\sum }_{r=1}^{u_1}\frac{{\overline{y}}_r^g}{y_{r0}^g}+{\sum }_{l=1}^{u_2}\frac{{\overline{y}}_l^b}{y_{l0}^b})}}$$

$$\overline{x}\ge \sum _{j=1,j\ne 0}^n{\lambda }_j{x}_j ,{\overline{y}}^g\le \sum _{j=1,j\ne 0}^n{\lambda }_j{y}_j^g$$

$${\overline{y}}^b\ge \sum _{j=1,j\ne 0}^n{\lambda }_j{y}_j^b ,\overline{x}\ge x_0, {\overline{y}}^g\le y_0^g, {\overline{y}}^b\ge y_0^b$$

$${\overline{y}}^g\ge 0, \lambda \ge 0$$

In this formula, the variables in the Super-SBM model are further defined, with the predicted targets $\overline{x}$, ${\overline{y}}^g$, and ${\overline{y}}^b$ representing the optimal targets (decision variables) determined on the adjusted frontier; while the variables $x_{i0}$, $y_{r0}^g$, and $y_{l0}^b$ represent the actual observed values of the unit currently being evaluated. At this stage, the slack variables capture the deviations between these target values and the actual observed values. The symbols $i$, $r$, and $l$ represent the indices for inputs, desirable outputs, and undesirable outputs, respectively, while the symbol j represents the decision-making unit. Specifically, the condition $j$ ≠ 0 in the summation indicates a super-efficiency adjustment, which strictly excludes the evaluated unit (i.e., ${\mathrm{D}\mathrm{M}\mathrm{U}}_0$) from the reference set. This adjustment allows the calculated efficiency score ${\rho }^{\mathrm{*}}$ to exceed 1.0, thereby enabling a comprehensive ranking of all efficient units. ${\rho }^{\mathrm{*}}$ represents green economic efficiency. The input and output variables used in this study are described below.

Referring to the studies of Zhao et al. (2020a) and Kong et al. (2023), we selected the following input and output variables in the efficiency measurement [63,64]. The input factors were selected from three perspectives: energy input, labour input and capital input. Energy input is measured by energy consumption (kg of oil equivalent), labour input by the total labour force, and capital input by gross fixed capital formation. In terms of desirable output, GDP and forest coverage rate are regarded as desirable output. In terms of undesirable outputs, we focus on major greenhouse gas emissions, including carbon dioxide (${\mathrm{C}\mathrm{O}}_2$), methane (${\mathrm{C}\mathrm{H}}_4$), and nitrous oxide (${\mathrm{N}}_2\mathrm{O}$). All the above data are obtained from the World Development Indicators (WDI).

Table 1. Indicators for Measuring Green Economic Efficiency.

Type	Index	Description
Input	Energy input	Total energy consumption (kilograms of oil equivalent)
	Labour input	Total labour force
	Capital investment	Gross fixed capital formation (constant 2015 US dollars)
Desirable output	Economic output	GDP (constant 2015 US dollars)
	Environmental optimization	Forest coverage rate (as a percentage of land area)
Undesirable output	Carbon dioxide	Total carbon dioxide (${\mathrm{C}\mathrm{O}}_2$) emissions (excluding LULUCF, million tons of ${\mathrm{C}\mathrm{O}}_2$ equivalent)
	Methane	Total methane (${\mathrm{C}\mathrm{H}}_4$) emissions (excluding LULUCF, million tons of ${\mathrm{C}\mathrm{O}}_2$ equivalent)
	Nitrous oxide	Total nitrous oxide (${\mathrm{N}}_2\mathrm{O}$) emissions (excluding LULUCF, million tons of ${\mathrm{C}\mathrm{O}}_2$ equivalent)

lists the specific indicators used in the construction of the green economic efficiency index, Figure 3 shows the GEE gap of countries in 2022, Figure 4 shows the change trend in GEE of countries from 2000 to 2022, and Figure 5 presents the trends in green economic efficiency for developed countries, emerging countries, and the 19 G20 countries.

3.2. Control Variables

In order to control other factors that may affect the green economic efficiency, this paper includes a series of control variables in the model. The selection of control variables is based on Su et al. (2022) and Jóźwik et al. (2025) [65,66]. Controlling for these variables helps reduce omitted-variable bias and improves the robustness of the results [65,66].

The level of economic development (AGDP) is measured by GDP per capita. A higher level of economic development usually provides more capital and technological resources for clean production and energy-efficiency improvement, which may lead to better environmental performance. Population density (PD) is measured by the number of people per square kilometre. On the one hand, higher population density may increase pressure on resources and the environment; on the other hand, it may improve infrastructure utilization efficiency. Its effect on green economic efficiency is uncertain. Macroeconomic stability is measured by the inflation rate (INF, expressed as the annual increase in the GDP deflator). Large inflation fluctuations can discourage investment in long-term green projects, whereas a more stable inflation environment tends to support green economic development. Trade openness (TRA) is measured by the ratio of total trade (the sum of imports and exports) to GDP. Greater trade openness may improve green economic efficiency through technology diffusion and international competition, although its effects may vary across countries. The urbanization rate (UP) is measured by the proportion of urban population to the total population. Urbanization is usually accompanied by improved infrastructure and more efficient resource allocation, and it may also strengthen public environmental awareness.

3.3. Mechanism Variables

In addition to the direct effect, this paper further examines the impact of geopolitical risk on green economic efficiency through the mediating effect of energy structure and the moderating effect of foreign direct investment.

3.3.1. Mediating Variables: Fossil Fuel Energy Consumption Ratio (FFEC) and Exchange Rate (ER)

The fossil fuel energy consumption ratio (FFEC) is measured as the percentage of fossil fuel energy consumption in primary energy consumption, reflecting the degree of cleanliness of a country’s energy structure. Geopolitical tensions may prompt countries to adjust their energy mix by disrupting energy supply chains and causing energy price fluctuations. For example, high geopolitical risk may lead a country to rely more on domestic fossil energy to ensure energy security, thereby increasing the share of fossil fuels; in some cases, it may also promote greater energy autonomy. However, over-reliance on fossil fuels is generally not conducive to improving green economic efficiency, as its high carbon emissions will reduce ecological efficiency.

The exchange rate (ER) is measured as the annual average exchange rate of the local currency against the US dollar (US dollars per unit of local currency). ER reflects the stability of a country’s currency value and the cost of international trade. Rising geopolitical risks often trigger capital flight to safe-haven assets, causing large exchange rate fluctuations that directly affect the import costs of clean technologies and the financing capacity of green projects denominated in foreign currencies. For example, depreciation of the local currency raises the import prices of green and energy-saving equipment and may discourage firms from investing in green projects; drastic exchange rate fluctuations also increase the repayment risk of cross-border green bonds and reduce resource allocation efficiency. Unstable exchange rates are generally not conducive to improving the green economic efficiency, because green technology innovation and promotion depend on a stable funding environment.

3.3.2. Moderating Variables

Foreign direct investment (FDI) is measured by net inflows (current US$). Foreign capital is not only a source of funds, but also an important channel for the diffusion of technology and the introduction of management experience. This paper argues that FDI may affect the relationship between geopolitical risk and green economic efficiency. On the one hand, the technology spillover effects of FDI help improve energy utilization efficiency and environmental governance, thereby alleviating the negative impact of geopolitical uncertainty on green development to some extent. On the other hand, geopolitical conflicts often undermine investor confidence and inhibit foreign capital inflows, thereby amplifying the adverse effects of geopolitical risk.

Table 2. Variable Definitions.

Variable Type	Variable Name	Variable Symbol	Variable Definition
Independent variable	Geopolitical Risk Historical Index	GPRH	The Geopolitical Risk Index (historical index) constructed by Caldara and Iacoviello (2022) [61] is adopted to measure the geopolitical uncertainty of a country. The larger the index, the higher the risk.
Dependent variable	Green economic efficiency	GEE	The green economic efficiency index calculated based on the Super-SBM model, which integrates input factors, desirable outputs and undesirable outputs, indicates a higher green economic efficiency with a larger value.
Control variable	Per capita GDP	AGDP	GDP per capita (in ${10}^5$ 2015 US dollars) is used to indicate the economic development level of a country.
	Population density	PD	Population density (${10}^3$ people per ${\mathrm{k}\mathrm{m}}^2$) reflects the degree of population spatial concentration.
	GDP deflator inflation rate	INF	The GDP deflator inflation rate (expressed as a decimal between 0 and 1) reflects the macro price level and economic stability.
	Trade openness	TRA	Trade openness, measured by the ratio of total imports and exports to GDP (scaled by ${10}^3$).
	The proportion of urban population to the total population	UP	The level of urbanization is measured by the proportion of urban population to the total population (0–1).
	Total labour force participation rate	TLFP	Labour force participation rate (0–1), measured by the labour force participation rate of the population aged 15 and above (simulated ILO estimates).
Mediating variable	The percentage of fossil fuel energy consumption in total energy consumption	FFEC	The percentage of fossil fuel energy consumption in total energy consumption is used to indicate the degree of dependence on fossil energy in the energy structure.
	Exchange rate	ER	The annual average exchange rate (US dollars per unit of local currency, scaled by ${10}^4$).
Moderating variable	Net inflow of foreign direct investment	FDI	Net inflow of foreign direct investment (BOP basis, in trillions of current US dollars) is used to measure the scale of foreign capital inflows.

presents the definitions of the research variables.

3.4. Model

3.4.1. Main Regression Model

To investigate the impact of geopolitical risks on green economic efficiency (GEE) in G20 countries, we employ the following benchmark regression model:

$${GEE}_{i,t}={\alpha }_0+{\alpha }_1{GPRH}_{i,t}+{\sum }_{j=1}{\beta }_j{Controls}_{i,t}+{\epsilon }_{i,t}$$

(1)

In the above model, i and t denote location and year, respectively. GEE represents green economic efficiency in a given year, while GPRH captures geopolitical risk. ${\alpha }_0$ denotes the constant term, and ${\epsilon }_{i,t}$ is the error term, ${Controls}_{i,t}$ refers to a set of control variables. Equation (1) incorporates both time and region fixed effects.

3.4.2. Mechanism Testing

As previously discussed, foreign direct investment and fossil fuel energy consumption may serve as key mechanisms through which geopolitical risks influence green economic efficiency. To examine these pathways, this study employs both mediating and moderating effect analyses. The corresponding econometric models are specified as follows:

$${GEE}_{i,t}={\beta }_0+{\beta }_1{FFEC}_{i,t}+{\beta }_2{GPRH}_{i,t}+{\beta }_3{Controls}_{i,t}+{\epsilon }_{i,t}$$

(2)

$${GEE}_{i,t}={\beta }_0+{\beta }_1{ER}_{i,t}+{\beta }_2{GPRH}_{i,t}+{\beta }_3{Controls}_{i,t}+{\epsilon }_{i,t}$$

(3)

$${GEE}_{i,t}={\beta }_0+{\beta }_{1}FDI\times {GPRH}_{i,t}+{\beta }_2{GPRH}_{i,t}+{\beta }_3{Controls}_{i,t}+{\epsilon }_{i,t}$$

(4)

4. Empirical Results

4.1. Descriptive Statistical Analysis

Table 3. Descriptive Statistical Analysis.

Variable	N	Mean	SD	p25	p75	Min	Max
GEE	437	0.580	0.306	0.326	0.860	0.0444	1.012
GPRH	437	0.397	0.651	0.0787	0.420	0.0142	3.776
AGDP	437	0.236	0.176	0.0833	0.380	0.00890	0.598
PD	437	0.141	0.149	0.0162	0.235	0.00260	0.529
INF	437	0.0613	0.0852	0.0156	0.0742	−0.0232	0.494
TRA	437	0.0517	0.0170	0.0402	0.0622	0.0219	0.0961
UP	437	0.736	0.151	0.684	0.823	0.289	0.919
TLFP	437	0.603	0.0582	0.575	0.644	0.469	0.729
FFEC	437	82.21	12.26	78.28	89.94	47.41	100
FDI	437	0.0524	0.0767	0.00900	0.0569	−0.0171	0.381

reports the descriptive statistical results of the variables involved in the regression analysis. According to the information in Table 3, the average value of green economic efficiency (GEE) is about 0.58, this means that overall, there is still significant room for improvement in the green production processes of various countries, with the average efficiency level reaching only 58% of the potential optimal level. This figure reflects that most countries have not yet achieved effective coordination of resource utilization, energy structure, and environmental governance, and the green transition is still at a moderate level. In addition, there is a significant gap between the maximum and minimum values of the GEE, indicating that the green development capabilities of different countries are highly unbalanced, showing clear transnational disparities. Geopolitical risk (GPRH) averages about 0.40. Its relatively large standard deviation and wide range indicate substantial cross-country variation over the sample period, in line with the rise in regional conflicts and terrorist incidents worldwide in recent years. Higher geopolitical risk often leads to declines in investment and employment and increases the risk of economic downturn. This suggests that geopolitical turmoil may weaken investment confidence and related channels, thereby hindering green investment and efficiency improvement.

As for the control variables, the minimum value of GDP per capita (AGDP) is 0.0089, and the maximum value is 0.598, indicating substantial differences in the level of economic development across the sample. AGDP reflects a country’s level of economic development, which in turn may affect its technological capacity, energy structure, and industrial structure. The mean and interquartile range of PD (population density) indicate substantial cross-country differences in population density. As a key demographic factor, population density is closely related to resource consumption. Population size and density can affect a country’s energy demand and pollution emissions. The range of inflation (INF) varies from mild deflation (minimum = −0.023) to high inflation (maximum = 0.494), indicating significant differences in macroeconomic stability across different countries and periods, which may affect the conditions for green investment. The standard deviation of trade openness (TRA) is 0.017, indicating relatively limited variation across countries. The minimum value of the urbanization rate (UP) is 0.289, and the maximum value is 0.919, indicating significant differences in the level of urbanization between different countries, which may impact the energy structure of these countries. The labour force participation rate (TLFP) is relatively concentrated (mean = 0.603, standard deviation = 0.058), indicating that the dispersion of labour supply is smaller than that of other macro factors. Overall, the descriptive statistics show sufficient variation across variables, providing a reliable basis for the subsequent regression analysis.

4.2. Main Regression

To test the impact of geopolitical risk on green economic efficiency, this paper first conducts a benchmark regression analysis.

Table 4. Benchmark Regression Results.

	(1)	(2)
	GEE	GEE
GPRH	−0.0885 ***	−0.1001 ***
	(−4.0933)	(−5.9697)
AGDP		0.6541 ***
		(7.1214)
PD		0.6020 ***
		(6.6664)
INF		−0.3337 ***
		(−2.9764)
TRA		0.1543
		(0.2499)
UP		0.3782 ***
		(3.0634)
TLFP		−0.7172 ***
		(−3.7258)
_cons	0.6149 ***	0.5467 ***
	(44.4099)	(3.6101)
Time fixed effect	YES	YES
Location fixed effect	YES	YES
N	437	437
R2	0.0120	0.7634

t statistics in parentheses. *** p < 0.01.

reports the main regression results of the impact of geopolitical risk on green economic efficiency. Column (1) is the regression result containing only the core explanatory variable GPRH and the double fixed effect, and column (2) adds all the control variables to form a complete model. According to the regression results, the coefficient of GPRH is negative and significant at the 1% level, regardless of whether control variables are included. The coefficients are −0.0885 and −0.1001, respectively, based on the descriptive statistics, the standard deviation of GPRH is 0.651, and the mean of GEE is 0.580. Holding other factors constant, a one-standard-deviation increase in geopolitical risk is associated with an average decrease of approximately 0.065 units in green economic efficiency. This corresponds to an 11.2% decline relative to the sample mean. Therefore, hypothesis H1 is confirmed. These results indicate that rising geopolitical uncertainty constitutes an obstacle to the green transformation of the world’s major economies.

Further examination of the estimated coefficients of the control variables in Table 4 reveals that the impact direction of most variables is consistent with expectations. In column (2), the AGDP coefficient is positive and significant, meaning that the higher the level of economic development, the higher the green economic efficiency of the country. The PD coefficient is positive and significant, indicating that the green economic efficiency of countries with higher population density is relatively higher. The INF coefficient is negative and significant, indicating that the decline in macroeconomic stability (i.e., the rise in inflation) is not conducive to green economic efficiency, which may be because high inflation weakens the ability to invest in environmental protection. The coefficient of TRA is not significant, indicating that the impact of trade on green efficiency varies from country to country. The coefficient of UP is positive and significant, indicating that the increase in urbanization can bring about economies of scale and more complete environmental protection infrastructure, thereby helping to improve green economic efficiency. The TLFP (total labour force participation rate) coefficient is negative and significant, which may mean that economies with high labour force participation are more dependent on traditional labour-intensive industries, which to some extent weakens the potential for improving green economic efficiency.

4.3. Analysis of Mechanism Effect

4.3.1. Moderating Effect

We first examined the moderating effect of the interaction between foreign direct investment (FDI) and geopolitical risk (GPRH) on green economic efficiency (GEE). The regression results in column (1) show that the coefficient of the FDI*GPRH interaction term is positive and significant, indicating that FDI has a significant moderating effect. In countries with higher levels of FDI inflows, the adverse impact of geopolitical risk on green economic efficiency is significantly reduced, thus supporting hypothesis H2. Specifically, high levels of foreign direct investment can offset the negative impact of geopolitical risk to a certain extent, help the host country with the capital and technology it needs, and enhance economic resilience, thereby alleviating the inhibitory effect of geopolitical uncertainty on green economic efficiency.

4.3.2. Test of the Mediating Effect

Columns (2) to (5) of

Table 5. Mediation Analysis Results.

	(1)	(2)	(3)	(4)	(5)
	GEE	FFEC	GEE	ER	GEE
FFEC			−0.0078 ***
			(−12.3030)
ER					0.1651 ***
					(4.3399)
FDI	−1.1293 ***
	(−5.9505)
FDI × GPRH	0.4872 ***
	(3.9063)
GPRH	−0.1391 ***	3.7992 ***	−0.0704 ***	−0.0524 **	−0.0914 ***
	(−5.2547)	(3.3767)	(−4.8569)	(−2.4356)	(−5.5324)
AGDP	0.7152 ***	3.3390	0.6802 ***	−0.1695	0.6821 ***
	(7.6125)	(0.5416)	(8.6801)	(−1.4366)	(7.5700)
PD	0.6463 ***	−13.7164 **	0.4949 ***	0.1221	0.5819 ***
	(7.3721)	(−2.2628)	(6.3862)	(1.0521)	(6.5760)
INF	−0.4095 ***	32.6990 ***	−0.0784	0.1934	−0.3656 ***
	(−3.7598)	(4.3452)	(−0.8018)	(1.3430)	(−3.3255)
TRA	−0.1457	81.4457 *	0.7900	−1.0652	0.3302
	(−0.2385)	(1.9644)	(1.4927)	(−1.3424)	(0.5451)
UP	0.2408 *	5.3449	0.4199 ***	−1.0892 ***	0.5580 ***
	(1.8735)	(0.6450)	(3.9864)	(−6.8684)	(4.3695)
TLFP	−0.2332	−39.4463 ***	−1.0251 ***	0.0237	−0.7212 ***
	(−1.1373)	(−3.0526)	(−6.1729)	(0.0958)	(−3.8284)
_cons	0.4044 ***	95.4721 ***	1.2919 ***	0.9398 ***	0.3916 **
	(2.6035)	(9.3915)	(9.0566)	(4.8308)	(2.5688)
Time fixed effect	YES	YES	YES	YES	YES
Location fixed effect	YES	YES	YES	YES	YES
N	437	437	437	437	437
R2	0.7827	0.3346	0.8283	0.4207	0.7740

t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

show the regression results of the mediating effect. First, column (2) shows the impact of geopolitical risk (GPRH) on the fossil fuel energy consumption rate (FFEC). The results show that the effect of GPRH on FFEC is positive and significant at the 1% level, with a coefficient of 3.7992, indicating that rising geopolitical risk significantly increases the share of fossil fuel consumption in the energy structure. Column (3) includes both FFEC and GPRH in the regression of GEE. The results show that the impact of the two on GEE is significant at the 1% statistical level. The coefficient of FFEC is negative, indicating that the increase in fossil energy consumption will reduce the green economic efficiency. Geopolitical risks reduce the green economic efficiency by increasing the country’s fossil energy consumption. These results support hypothesis H3.

In addition to the energy structure channel, exchange rate changes are also an important channel through which geopolitical risk affects green economic efficiency. Column (4) shows the impact of geopolitical risk (GPRH) on the exchange rate (ER). The results show that the impact of GPRH on ER is negative and significant at the 5% statistical level, indicating that when geopolitical risk rises, the exchange rate of the local currency shows a significant downward trend. Column (5) includes both ER and GPRH in the regression of GEE, and the results show that both are significantly correlated at the 1% statistical level. These results suggest that rising geopolitical risk weakens national green economic efficiency by disrupting exchange rate stability. The decline of the exchange rate increases the uncertainty of the introduction and investment of green technology, which to some extent amplifies the negative impact of GPRH on GEE. This finding supports hypothesis H4, which states that geopolitical risk causes currency depreciation and thus reduces a country’s green economic efficiency.

4.4. Heterogeneity Analysis

Although the benchmark regression analysis indicates that the rise in geopolitical risk significantly reduces the green economic efficiency of G20 countries, this relationship may vary due to core characteristics such as the level of economic development, energy endowment characteristics, and degree of trade openness in different countries. Therefore, we referenced relevant data from the International Monetary Fund (IMF) and the World Bank (WDI) to conduct grouped regressions based on three dimensions: economic development level, energy endowment, and trade openness.

Table 6. Heterogeneity Analysis Results.

	(1)	(2)	(3)	(4)	(5)	(6)
	GEE	GEE	GEE	GEE	GEE	GEE
GPRH	−0.0307 ***	−0.0706	−0.1203 **	−0.0330	−0.1240 *	−0.1919 ***
	(−3.1351)	(−1.3817)	(−2.3760)	(−1.1132)	(−1.7612)	(−4.4600)
AGDP	0.0053	−6.1721 ***	0.1468	1.4129 ***	1.4550 ***	0.7260 ***
	(0.0327)	(−8.0155)	(0.8135)	(8.1812)	(7.8418)	(4.2274)
PD	0.1455 *	0.7349	1.2770 ***	−1.1968 ***	−0.2172	0.4063 **
	(1.9733)	(1.0717)	(5.8676)	(−4.8976)	(−0.9603)	(2.3217)
INF	0.5877 *	−0.0130	−0.7748 ***	0.0091	0.0625	−0.5993 ***
	(1.6546)	(−0.1067)	(−5.9276)	(0.0536)	(0.3227)	(−4.0145)
TRA	2.2829 ***	−2.9455 ***	−1.6329 *	2.4699 **	0.7920	−6.9547 ***
	(4.2390)	(−3.0826)	(−1.7912)	(2.1563)	(0.8329)	(−3.0907)
UP	0.8061 ***	0.2260	0.7368 ***	−1.5677 ***	0.2035	0.4900 **
	(3.9205)	(0.5496)	(3.8357)	(−5.1814)	(0.9168)	(2.2411)
TLFP	−1.7735 ***	−0.1075	−0.4720 *	0.1964	−1.7379 ***	0.0987
	(−5.8634)	(−0.2844)	(−1.6666)	(0.5220)	(−7.2173)	(0.2499)
_cons	1.1051 ***	0.8982 *	0.2614	1.1634 ***	1.1323 ***	0.3263
	(10.5305)	(1.8825)	(1.1767)	(3.0355)	(5.4498)	(0.9702)
Time fixed effect	YES	YES	YES	YES	YES	YES
Location fixed effect	YES	YES	YES	YES	YES	YES
N	207.0000	230.0000	259.0000	178.0000	218.0000	219.0000
r2	0.8189	0.6884	0.7879	0.9062	0.8936	0.7446

t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

presents the research results.

The results in column (1), for the sample of developed countries, show that geopolitical risk has a significant negative impact on green economic efficiency, with a coefficient of −0.0307 that is significant at the 1% level. This result is consistent with the baseline regression. The intensification of geopolitical risks increases regional uncertainty, which hinders policy advancement to some extent. The commitments made by developed countries to environmental protection and climate action may be weakened under heightened uncertainty.

In contrast, in the sample of emerging economies shown in column (2), the coefficient of GPRH is negative but not significant. This indicates that in emerging market countries, the direct impact of geopolitical risk on green economic efficiency is not significant, reflecting a notable heterogeneity between developed and emerging economies. This difference may stem from the varying stages of development, industrial structures, institutional completeness, and environmental governance capabilities between the two types of economies. Therefore, the transmission channels and intensity through which geopolitical risks affect green economic efficiency differ across countries with different national characteristics. Therefore, policy-making should take this heterogeneity into account: developed economies need to focus on maintaining geopolitical stability to ensure the fulfilment of their environmental commitments, while emerging economies should strive to improve institutional development and enhance their resilience to external shocks.

Resource endowment is the core factor determining a country’s energy structure, transformation dynamics, and sensitivity to geopolitical risks. Based on the resource dependence theory, the economy and finances of energy-dependent economies are highly tied to their domestic fossil energy industries, while the energy supply of energy-independent economies is highly reliant on international markets. The impact of geopolitical risks on the green economic efficiency of these two types of economies differs. Columns (3) and (4) show the grouping of samples from energy-exporting and energy-importing countries. We dynamically grouped the samples over time based on the net energy imports data from the World Bank Database (WDI). The results indicate that, in the resource-rich group, the coefficient of GPRH on GEE is −0.1203 and significant at the 5% level, whereas in the resource-poor group, the effect of GPRH on GEE is not significant.

For energy-importing countries, their energy supply is highly dependent on the international market. Geopolitical risks leading to global energy price fluctuations and disruptions in cross-border energy transportation can directly impact the stability of their energy supply. To ensure energy security, these economies may have to increase fossil energy consumption in the short term. At the same time, higher levels of geopolitical risk can raise the import costs of green technologies and clean energy equipment, significantly negatively impacting the efficiency of the green economy.

For energy-exporting economies, their domestic fossil energy supply is sufficient, and the direct impact of geopolitical risks on their energy supply is relatively weak. At the same time, during periods of rising geopolitical risks, the increase in global energy prices boosts their fossil energy export revenues, which in turn weakens the short-term motivation for green transition, ultimately resulting in the negative impact of geopolitical risks not being statistically significant.

Finally, considering that trade openness may alter the impact of geopolitical shocks on technological development and green economic efficiency, we referred to the study by Suleman et al. (2025) and divided the sample into high-trade-openness and low-trade-openness groups [67]. Columns (5) and (6) report the estimation results: in the high-trade-openness group, the GPRH coefficient is −0.1240 and significant at the 10% level; in the low-openness group, the coefficient is −0.1919, significant at the 1% level (t = −4.4600), and the absolute value is larger.

This indicates that geopolitical risks have a more pronounced impact on the green economic efficiency of low-openness economies. Low-openness economies typically face higher barriers to technology acquisition and weaker external financing channels. Geopolitical risks can lead to capital flight, increased import costs, and restricted access to critical green equipment, which can directly impact insufficient green investment and declining efficiency. In contrast, although high-openness economies are more exposed to external shocks, their trade partners and supply chain networks are more diversified, allowing them to mitigate risks and buffer the negative effects of geopolitical risks to some extent. As a result, they show a suppressed but weaker impact.

4.5. Robustness Test

To ensure the reliability of the above conclusions, we designed a number of robustness tests.

4.5.1. Changing the Calculation Method of Explanatory Variables

We first changed the measurement method of the core explanatory variable, recalculated the Geopolitical Risk Index (GPRH) using the geometric mean, and conducted regression tests using this new index. The geometric mean, by reducing the impact of short-term sudden fluctuations, can provide a more stable measure of potential risk levels, whereas the arithmetic mean is very sensitive to extreme values. Geopolitical risk data may be skewed and prone to extreme values. The geometric mean can help us, to some extent, to avoid errors caused by skewed distributions or short-term extreme values, making our experimental results more robust. However, since the calculation of the geometric mean is very sensitive to zero values (i.e., a single zero value can cause the entire annual index to be zero), we calculated the geometric mean under the condition of strictly positive monthly observations. Specifically, months with a GPRH value of zero were excluded from the annual aggregation of this robustness check. The formula is as follows:

$$\mathrm{G}\mathrm{P}\mathrm{R}\mathrm{H}1=\mathrm{e}\mathrm{x}\mathrm{p}\left({\displaystyle \frac{1}{{\mathrm{N}}_{\mathrm{t}}}}{\sum }_{\mathrm{m}=1,{\mathrm{G}\mathrm{P}\mathrm{R}\mathrm{H}}_{\mathrm{m}}>0}^{12}\mathrm{l}\mathrm{n}\left({\mathrm{G}\mathrm{P}\mathrm{R}\mathrm{H}}_{\mathrm{m}}\right)\right)$$

Here, ${\mathrm{N}}_{\mathrm{t}}$ represents the number of months (m) in year t where GPRH is strictly greater than 0. This method ensures that the alternative index can capture the average composite level of geopolitical risk during the periods when the risk is actually observed. The results are shown in column (1) of

Table 7. Robustness Tests.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
	GEE	GEE	GEE	GEE	GEE	GEE	GEE	CI
L.GPRH					−0.1115 ***
					(−6.4910)
GPRH1	−0.1065 ***
	(−5.9181)
GPRH		−0.0942 ***	−0.0692 ***	−0.0714 ***		−0.1001 ***	−0.1064 ***	0.1744 ***	−0.0980 ***
		(−6.1438)	(−6.7790)	(−4.8276)		(−7.9649)	(−3.5949)	(9.6890)	(−5.8656)
GEPU_shock									0.0007 **
									(2.2263)
AGDP	0.6568 ***	0.7665 ***	0.7658 ***	0.7526 ***	0.6739 ***	0.6541 ***	0.6571 ***	−2.0262 ***	0.6373 ***
	(7.1402)	(9.3953)	(8.4644)	(9.8690)	(7.4912)	(6.4979)	(4.1647)	(−20.5463)	(6.9487)
PD	0.6024 ***	0.6694 ***	0.5808 ***	0.6059 ***	0.6294 ***	0.6020 ***	0.5916 ***	−1.1002 ***	0.5968 ***
	(6.6611)	(8.0376)	(9.3814)	(7.3660)	(7.0532)	(11.1336)	(3.8323)	(−11.3475)	(6.6394)
INF	−0.3307 ***	−0.8410 ***	−0.3097 ***	−0.2536	−0.4358 ***	−0.3337	−0.4294 **	0.1763	−0.3323 ***
	(−2.9467)	(−8.2989)	(−4.3098)	(−1.1955)	(−3.8065)	(−1.1760)	(−2.1252)	(1.4648)	(−2.9788)
TRA	0.1408	−1.3193 **	1.6168 ***	0.5330	−0.3129	0.1543	0.0059	5.0937 ***	0.2412
	(0.2270)	(−2.3895)	(3.5000)	(0.8709)	(−0.5114)	(0.2013)	(0.0055)	(7.6809)	(0.3916)
UP	0.3729 ***	0.5319 ***	0.4965 ***	0.1018	0.3857 ***	0.3782 ***	0.3843 *	1.0609 ***	0.4057 ***
	(3.0147)	(4.3060)	(5.0572)	(0.7924)	(3.1280)	(4.5451)	(1.8107)	(8.0049)	(3.2864)
TLFP	−0.7154 ***	−0.8840 ***	−1.2924 ***	−0.7607 ***	−0.7372 ***	−0.7172 ***	−0.7220 **	2.2935 ***	−0.7021 ***
	(−3.7123)	(−4.8373)	(−8.4574)	(−3.1388)	(−3.9180)	(−8.3726)	(−2.1901)	(11.0964)	(−3.6629)
_cons	0.5490 ***	0.5936 ***	0.6911 ***	0.9425 ***	0.5759 ***	0.6325 ***	0.5642 **	−1.3109 ***	0.4800 ***
	(3.6212)	(4.0447)	(6.1545)	(4.3490)	(3.8539)	(11.5158)	(2.1765)	(−8.0621)	(3.1237)
Time fixed effect	YES	YES	YES	YES	YES	YES	YES	YES	YES
Location fixed effect	YES	YES	YES	YES	YES	YES	YES	YES	YES
N	437	285	437	437	418	437	152	437	437
R2	0.7630	0.8856			0.7838	0.7634	0.7790	0.8567	0.7663
Pseudo R2			0.6804	0.5494

t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

. After changing the measurement method, the coefficient of GPRH is still negative and significant at the 1% level, and the conclusion is consistent with the benchmark regression. This shows that the main conclusion is not affected by changes in the measurement of the explanatory variable and effectively rules out potential measurement error.

4.5.2. Regression by Time Period

Furthermore, considering that changes in the macroeconomic environment across periods may affect the results, we shortened the sample period. Given that the global financial crisis occurred in 2008, which had a certain degree of impact on energy prices, capital flows, and national economic development, we chose 2008–2022 as the new time window for regression analysis. Column (2) of Table 7 presents the regression results from 2008 to 2022. GPRH remains significantly negatively correlated with green economic efficiency at the 1% statistical level, with a coefficient of −0.0942. This is basically consistent with our benchmark regression results. These results tests indicate that the conclusions drawn from this study remain reliable under different international environments and time spans, further supporting the research hypothesis.

4.5.3. Replacing the Regression Model

In order to test the robustness of the model results, this paper uses the method of replacing the regression model to further analyze the regression relationship. We selected the 25% and 75% quantiles of the explained variable GEE for regression estimation to examine whether the impact of the explanatory variables was consistent at different conditional distribution levels. Columns (4) and (5) of Table 7 report the regression results of the 25% quantile and the 75% quantile, respectively. The estimation results under each quantile are basically consistent with the benchmark regression, indicating that the core explanatory variable of geopolitical risk maintains a significant negative impact in countries with low and high green economic efficiency. Specifically, the coefficient of GPRH is negative and significant at the 1% level in the regression of both the 25% quantile (low-efficiency group) and the 75% quantile (high-efficiency group). There is no difference in the direction of the impact of geopolitical risk shocks on countries with high green economic efficiency and countries with low green economic efficiency, both of which show a decline in green economic efficiency, further confirming the robustness of the benchmark conclusion.

4.5.4. Lagged One-Period Explanatory Variable

Furthermore, to avoid potential reverse causality in the study, we use lagged explanatory variables for testing. Column (5) of Table 7 reports the regression results with the explanatory variable lagged by one period. The results indicate that after applying a one-period lag, the coefficient of GPRH remains negative and significant at the 1% level. This suggests that the adverse impact of geopolitical risk on green economic efficiency is persistent and that reverse causality is unlikely, confirming the robustness of the results.

4.5.5. Driscoll–Kraay Robustness Test

Given the potential for common shocks and economic spillovers among G20 countries, further robustness checks are warranted; therefore, we re-estimated the baseline regression model using Driscoll–Kraay robust standard errors. In the context of globalization and regional economic integration, the macroeconomic performance of the sample countries often exhibits spatial spillover effects, implying the potential presence of cross-sectional dependence in the panel data. To formally diagnose this, we first employed the Pesaran CD test. The results show that the CD statistics for the core variables, green economic efficiency (GEE) and geopolitical risk (GPRH), are 5.67 and 24.31, respectively (both with p < 0.01). This indicates significant cross-sectional dependence among G20 countries, meaning that the error terms are correlated across individuals. This correlation stems from the fact that G20 economies are jointly exposed to global shocks such as geopolitical tensions, energy price fluctuations, and international climate policies. Such dependence could bias the standard errors and inflate statistical significance. Consequently, we applied Driscoll–Kraay standard errors to obtain robust inferences. The corresponding results are presented in column 6 of Table 7. After accounting for potential cross-sectional and serial correlation using these robust standard errors, GPRH remains significantly and negatively correlated with green economic efficiency at the 1% statistical level, with a coefficient of −0.1001. This is consistent with the baseline regression results, demonstrating that our findings remain highly robust even after addressing these potential estimation issues.

4.5.6. Multi-Year Average Estimation

In order to further verify the long-term stability of the relationship between geopolitical risk and green economic efficiency, we re-estimated the model using multi-year averages. Annual macroeconomic data may be affected by short-term business cycle fluctuations or random shocks, making it difficult to accurately reflect long-term structural trends. Therefore, we convert the original annual panel data into non-overlapping three-year averages (with the last period consisting two years to maintain sample integrity). By eliminating interannual fluctuations, this method can isolate the long-term impact of geopolitical risk.

The estimated results obtained using the average values over these three years are listed in column (7) of Table 7. The coefficient of GPRH remains −0.0784 and is significant at the 1% statistical level. These results strongly demonstrate that the inhibitory effect of geopolitical risk on green economic efficiency is not a temporary phenomenon driven by short-term fluctuations but a persistent effect that remains stable over a longer time frame. This confirms the robustness of our core conclusion, unaffected by short-term macroeconomic noise.

4.5.7. Changing the Core Dependent Variable

To further test robustness, we used national carbon intensity (CI) as a substitute for green economic efficiency. National carbon intensity is measured as total carbon dioxide emissions divided by GDP. The results are shown in Table 7, column (8). After replacing the core dependent variable, GPRH is significantly positively associated with carbon intensity, indicating that higher geopolitical risk is associated with higher carbon intensity and, therefore, lower green economic efficiency.

4.5.8. Other Exogenous Shocks

To rule out the possibility that the baseline results might be confounded by omitted global macroeconomic shocks, we conducted an additional robustness check. Following the methodology of Caldara and Iacoviello (2022) [61], we selected the Global Economic Policy Uncertainty (GEPU) index to proxy for global macroeconomic shocks. We then interacted it with the historical average geopolitical risk level of each region to construct a measure of region-specific exposure to global shocks (GEPU_shock). As shown in column 9 of Table 7, even after introducing this global control variable, the coefficient of the core explanatory variable (GPRH) remains significantly negative at the 1% level, with an estimate of −0.0980. This finding is highly consistent with the baseline regression, demonstrating that our main results are not driven by broader international economic disturbances and further verifying the robustness of the study.

4.6. Endogeneity Test

If there is an endogeneity problem between geopolitical risk (GPRH) and green economic efficiency (GEE), the empirical results of this study may be biassed. To address potential endogeneity in the model, this paper employs the instrumental variable method.

This paper selects two instrumental variables. The first is military expenditure (as a percentage of GDP, MS), which is significantly related to geopolitical risk. The greater the geopolitical security pressure and the higher the geopolitical risk a country faces, the more likely it is to increase defence investment and military expenditure. Military expenditure is a direct reflection of geopolitical risk at the national fiscal level. At the same time, the scale of military expenditure is determined by a country’s long-term defence policy and the surrounding geopolitical situation and is not directly related to the country’s green economic efficiency. Second, the number of refugees accepted by each country in a given year (REF) directly reflects the level of regional geopolitical conflict and political instability. The higher the geopolitical risk in a region, the larger the refugee flows and, consequently, the more refugees host countries tend to accept. This variable is highly correlated with the core explanatory variable, geopolitical risk. The scale of refugee acceptance is mainly determined by the geopolitical conflict situation in the source country, which is exogenous to the host country’s green economic efficiency and has no direct correlation with the country’s energy transition process.

The first-stage regression results in columns (3) and (5) of

Table 8. Endogeneity Tests.

	(1)	(2)	(3)	(4)
	GPRH	GEE	GPRH	GEE
MS	0.3607 ***	0.0376 ***
	(16.1880)	(3.0716)
AGDP	1.4842 ***	0.7637 ***	0.2973	0.6232 ***
	(6.8627)	(7.8207)	(1.1357)	(6.6804)
PD	1.4059 ***	0.6570 ***	1.2580 ***	0.5994 ***
	(6.9770)	(7.2082)	(5.1291)	(6.6541)
INF	0.2496	−0.3302 ***	0.2226	−0.3423 ***
	(0.9620)	(−2.9759)	(0.7079)	(−3.0582)
TRA	−11.6811 ***	−0.0146	−14.9417 ***	0.1346
	(−8.8956)	(−0.0238)	(−9.5606)	(0.2185)
UP	−2.3496 ***	0.2099	−0.7330 **	0.4097 ***
	(−8.0841)	(1.5678)	(−2.1074)	(3.2927)
TLFP	0.0730	−0.8076 ***	1.8610 ***	−0.6727 ***
	(0.1619)	(−4.1893)	(3.4694)	(−3.4728)
GPRH		−0.1412 ***		−0.1106 ***
		(−6.6220)		(−6.2249)
REF			0.0116 ***	0.0011 *
			(7.2049)	(1.7523)
_cons	1.3345 ***	0.6341 ***	0.2510	0.5031 ***
	(3.8033)	(4.1568)	(0.5835)	(3.2859)
Kleibergen–Paap rk LM		78.789		13.258
Cragg–Donald Wald F		262.052		51.910
Kleibergen–Paap rk Wald F		115.435		27.950
Time fixed effect	YES	YES	YES	YES
Location fixed effect	YES	YES	YES	YES
N	437.0000	437.0000	437.0000	437.0000
R2	0.7189	0.7688	0.5885	0.7652

t statistics in parentheses. * p < 0.1, ** p < 0.05, *** p < 0.01.

show that both instrumental variables are significantly related to geopolitical risk (GPRH), meeting the relevance requirement for instrumental variables. The second-stage regression results in columns (4) and (5) show that, when the share of military expenditure and the number of accepted refugees are used as instrumental variables, the coefficients of geopolitical risk (GPRH) on green economic efficiency (GEE) are both significantly negative, at −0.1412 and −0.1106, respectively, and both are significant at the 1% level. This is consistent with the core conclusions of the baseline regression.

The results of the instrumental variable tests indicate that both instrumental variables are appropriate. The Kleibergen–Paap rk LM statistics for the underidentification test are 78.789 and 13.258, respectively, rejecting the null hypothesis of “instrumental variable underidentification,” indicating no underidentification problem. The Cragg–Donald Wald F statistics are 262.052 and 51.910, respectively, passing the weak instrument test, and the Kleibergen–Paap rk Wald F statistics are 115.435 and 27.950, indicating no weak instrument problem. Furthermore, a Hansen J overidentification test was conducted, with a test statistic of 0.263 and a p-value of 0.6078, confirming that both instrumental variables satisfy the exclusion restriction, indicating that the choice of instrumental variables is appropriate.

In summary, after using the instrumental variable method to address potential endogeneity issues, the core conclusion of this paper remains valid, namely, that the rise in geopolitical risk has a significant inhibitory effect on a country’s green economic efficiency, and the research results are robust.

5. Conclusions

5.1. Results

This paper empirically examines the impact of geopolitical risk on green economic efficiency and its underlying mechanisms using panel data from G20 countries. The main conclusions are as follows: first, geopolitical risk significantly reduces green economic efficiency, and this effect differs across developed and emerging economies, resource endowments, and levels of trade openness. Specifically, in developed economies, the negative impact of geopolitical risk on green economic efficiency is more significant, while in emerging economies, the effect is not statistically significant. This suggests that different national systems and economic development levels will affect the decision-making and response capabilities of countries in terms of geopolitical risks. In other words, countries with higher levels of economic development have greater resilience when facing geopolitical risks and are less likely to be adversely affected. Furthermore, the heterogeneity analysis indicates that the impact of geopolitical risks is more pronounced in countries that rely on external energy supplies.

Second, the mechanism analysis shows that geopolitical risk indirectly weakens green economic efficiency through two channels: it increases the share of fossil fuel consumption, thereby hindering the transition of the energy structure toward renewable energy, and it causes exchange rate depreciation, which increases uncertainty in trade and investment and thereby amplifies the adverse impact on green economic efficiency. Regarding fossil fuel consumption, this mainly reflects countries’ concerns about energy security under geopolitical risk. When a country relies on external energy sources, it becomes more vulnerable in the event of a geopolitical crisis. Regarding the exchange rate, when a country is confronted with a geopolitical crisis, its currency faces pressure and international capital circulation is hindered.

Third, in terms of the moderating effect, foreign direct investment (FDI) plays a buffering role. In countries with high levels of FDI inflows, the inhibitory effect of geopolitical risk on green economic efficiency is significantly weakened. FDI helps to introduce advanced capital and technology, enhance economic resilience, and thus partially offset the impact of geopolitical uncertainty.

5.2. Discussion

The empirical results of this study indicate that geopolitical risks significantly reduce green economic efficiency, and recent global events provide concrete evidence of the practical significance of our findings. For example, at the beginning of 2022, after the outbreak of conflict between Russia and Ukraine, Russia’s gas supply to European countries was suddenly interrupted, forcing many European countries to reactivate coal-fired power plants, expand fossil fuel reserves, and delay the deployment of renewable energy. This event made energy security the top priority for countries with relatively limited resource endowments. These emergency measures support our mechanism that geopolitical risks will increase the stockpiling of fossil fuels to address potential energy security issues, thereby undermining green economic efficiency.

In addition, the recurring geopolitical tensions and domestic instability in Turkey have led to a significant depreciation of its currency. Since renewable energy equipment and clean technology components largely depend on imports, the depreciation of the currency has significantly increased the cost of green investments. This aligns with the empirical mechanism identified in this study, which shows that geopolitical risks can reduce the efficiency of the green economy by increasing exchange rate volatility and raising the costs of green production inputs.

These real-world cases highlight the broader practical implications of our research findings: geopolitical shocks can hinder energy security, foreign investment, and the advancement of sustainable policies. Understanding these mechanisms is crucial for formulating climate and energy strategies that can remain resilient under geopolitical uncertainty.

5.3. Policy Recommendations

Based on the research findings, we provide several recommendations for policymakers and managers to better promote sustainable development, improve energy utilization efficiency, and achieve green economic growth.

First, the country should accelerate the diversification of energy sources to reduce dependence on a single energy supplier, which can effectively avoid energy crises caused by sudden deterioration in bilateral relations. At the same time, countries should strengthen their green finance system to ensure stable funding support for renewable energy and green infrastructure during external shocks. In addition, by enhancing the flexibility of the power grid and strategic reserve capabilities, developed countries can avoid the passive situation of “reverting to high-carbon energy” during crises, thereby maintaining the continuity of long-term emission reduction goals.

Moreover, emerging economies are more susceptible to geopolitical risks in terms of technology, capital, and institutions capacity. Therefore, they need to enhance their foundational capabilities for green development through institutional building and attracting foreign investment. Improving institutional quality and enhancing policy transparency and stability can help reduce investors’ concerns about geopolitical risks.

Overall, geopolitical risks are an important factor affecting the green transition, and countries should take them into account when formulating climate policies and energy strategies. Whether developed economies or emerging economies, there is a need to seek a balance between energy security, financial stability, and institutional development to ensure that green development can continue to progress steadily in the context of rising global uncertainty.

5.4. Limitation

We acknowledge that this study still has some limitations. For example, there are limitations in the sample range and research design. The sample of this study comes from the G20 countries. Future studies could extend the scope of analysis to more developing countries or a longer time span to examine whether the impact of geopolitical risks on green development is consistent across national and regional contexts. Future work could examine micro-level transmission channels and other potential drivers—such as domestic political stability and the characteristics of regional conflicts—to deepen our understanding of how geopolitical uncertainty relates to green economic efficiency. Moreover, due to data availability and limitations, in using the Super-SBM model to calculate green economic efficiency (GEE), an indicator related to forest coverage lacks an effective substitute variable in standard sources. In future research, it may be possible to replace this indicator with a more precise variable to measure green economic efficiency (GEE) more accurately.

Overall, our findings suggest that the international geopolitical environment has a measurable influence on green and high-quality development, and they offer empirical evidence that may inform policies aimed at advancing green transformation and sustainable development under external turbulence.