Evaluating Growth and Crisis Risk Dynamics of Sustainable Climate Exchange-Traded Funds

Abstract

This study evaluates the dynamic risk spillovers and interconnectedness of environmental, social, and governance exchange-traded funds (ESG-ETFs) markets during two significant geopolitical conflicts, the Israel–Palestine and the Russia–Ukraine conflicts, alongside an extended analysis of the full period from July 2020 to October 2024. We investigate how crises transmit risks to the market by using the Total Connectedness Index (TCI) and net spillover measures. Our findings reveal a consistently high level of market interdependence. TCI values rose from 65.71% during the Israel–Palestine conflict to 67.28% in the full sample, indicating intensified risk sharing among markets as crises evolve. The markets “Deka MSCI World Climate Change ESG UCITS ETF (D6RP)” and “Amundi MSCI World SRI Climate Net Zero Ambition PAB UCITS ETF EUR Acc (XAMB)” emerge as prominent risk transmitters across all periods, actively spreading volatility throughout the system in both the crisis. In contrast, the markets “Amundi MSCI World Climate Transition CTB—UCITS ETF DR—EUR-C (LWCR)” and “Franklin STOXX Europe 600 Paris Aligned Climate UCITS ETF (PARI)” are primary risk receivers, absorbing a substantial portion of the instability in the Israel–Palestine and Russia–Ukraine conflicts. These dynamics underscore the shifting roles of financial markets during prolonged geopolitical tensions. These findings highlight the necessity of monitoring global markets, particularly during geopolitical shocks, to mitigate systemic risk and effectively navigate financial instability.

1. Introduction

The growing urgency of climate change and its devastating impact on global ecosystems have accelerated the demand for sustainable investment solutions and have been observed in several studies [1,2]. Climate-related crises have become more frequent and severe, as highlighted by the environmental shocks witnessed during the Russia–Ukraine conflict and the ongoing Israel–Palestine conflict [3]. These geopolitical tensions have not only disrupted traditional energy markets, but also amplified concerns over energy security, carbon emissions, and environmental sustainability. Sustainable climate exchange-traded funds (ETFs) have emerged as vital financial instruments designed to address environmental, social, and governance (ESG) concerns while offering investors opportunities for financial returns, and have been analyzed in some studies [4,5,6].

Theoretically, based on the efficient market hypothesis and the random walk theory, it is impossible to predict future prices [7]. Furthermore, arbitrage and contemporary portfolio theory highlight the advantages of diversification [8,9]. Since a shift to a green economy would necessitate financial inflows from investors and states, financial markets are crucial [10]. Sustainable development encompasses social, environmental, and economic facets, and sustainable investment and financing tackle these long-term issues [11,12]. This study discusses theories regarding the connections between sustainable finance, investment (SFI) tactics, and financial results [13].

This study evaluates the growth and crisis risk dynamics of several climate-focused ETFs that prioritize sustainable investments aligned with climate goals. These sustainable climate ETFs represent different regional markets and are tailored to respond to climate change by reducing carbon emissions, supporting low-carbon technologies, and meeting regulatory standards, such as the Paris Agreement goals. This study intends to provide investors and policymakers with important insights by illuminating the performance, growth, and risk dynamics of these ETFs in both normal conditions and crises. Sustainable investing has become a major focus for both institutional and retail investors as the financial landscape has changed dramatically in recent years. Nonetheless, an overview of exchange-traded funds was provided in [14]. Sustainable ETFs in relation to investment and climate change have been the subject of several studies [5,15]. Sustainable ETFs, especially those that concentrate on climate change, have grown as investors try to match their portfolios to climate objectives such as the Paris Agreement, which aims to keep global warming below 2 °C. Along with exposing investors to stocks that stand to win from the world’s shift to a low-carbon economy, these funds also provide them with a chance to protect themselves against climate risks [16,17]. Several factors have driven the development of sustainable ETFs. First, there is increasing recognition of the financial risks associated with climate change, including physical risks (e.g., natural disasters) and transition risks (e.g., policy changes). Second, regulatory pressures, particularly in Europe, have intensified, with governments mandating greater transparency and accountability in corporate environmental performance. Third, rising social awareness of ESG issues has prompted investors to prioritize sustainability in their investment strategies.

However, some new studies also explore diverse ETFs market dynamics in recent crises [3,18,19] and under climate policy uncertainty [20]. Despite the growing importance of climate ETFs, a gap remains in the literature on how these funds perform under geopolitical conflicts that disrupt energy markets and global supply chains. For instance, the Russia–Ukraine conflict in 2022 sent shockwaves through the energy sector, driving up oil and gas prices and raising concerns about energy independence in Europe. Similarly, the Israel–Palestine conflict has exacerbated instability in the Middle East, a region that plays a critical role in global energy supply. This study emphasizes climate-specific ETFs within the ESG category. The risk dynamics of climate ETFs are especially important to comprehend during crises because ESG and green ETFs are directly impacted by changes in climate policy and geopolitical tensions, which have distinctive effects on markets related to climate change. These conflicts underscore the interconnectedness between geopolitical risks, climate change, and sustainable investments. While previous studies have analyzed the financial performance of ESG and climate-related ETFs, few have delved into their behavior during periods of COVID-19 [21,22] and the heterogeneity of ETFs’ diversification [23]. Understanding the risk dynamics associated with these funds will provide valuable insights into their long-term sustainability and attractiveness as investments.

The novel contributions of this study are threefold. First, we analyze the crisis risk dynamics of sustainable climate ETFs during the Russia–Ukraine and Israel–Palestine conflicts using time-varying parameter vector autoregression (TVP-VAR), which has been adopted in various studies [24,25]. Some studies have focused on ESG ETFs during COVID-19 [6] and sustainable ETFs during the Ukrainian war [3]. However, this study is backward-looking and explores how these crises impact the performance and volatility of climate ETFs, offering insights into their resilience in fresh geopolitical conflicts. Second, the study offers a comparative analysis of climate ETFs across various regions (Europe, the US, and emerging markets) and crises for SFI investors and speculators to forecast future investment strategies. Some studies have focused on real estate [26], European ETFs [27], South African ETFs [28], and global ETFs [5]. Nonetheless, this study examines how regional factors and market conditions influence the growth and risk dynamics of climate-focused investment. Third, by focusing on ETFs that emphasize carbon reduction and climate transition, this study contributes to a growing body of research on sustainable finance. Some studies have explored the significance of energy and clean-energy ETFs [15,16,17,18,20]. However, by addressing the growth and crisis risk dynamics of climate-focused ETFs, this study adds to the growing literature on sustainable finance and climate resilience by offering new insights into how financial markets can contribute to global efforts to combat climate change. Additionally, this study enhances recent SFI literature by revealing novel insights into the changing risk-sharing patterns of ETFs during the Israel–Palestine and Russia–Ukraine conflicts.

This study’s findings illustrate that during the Israel–Palestine conflict, the TCI reached 65.71%, showing heightened market interconnectedness, with D6RP and XAMB acting as the dominant risk transmitters. Market trends of D6RP transferring risks and LWCR absorbing more shocks were observed in the Russia–Ukraine war, where the TCI increased marginally to 66.01%. Across both crises, the roles of the LWCR and PARI as risk receivers intensified, absorbing significant volatility. Nevertheless, this study is noteworthy in several respects. This study provides instant insights into the performance of climate ETFs during the Russia–Ukraine and Israel–Palestine conflicts, helping stakeholders make informed decisions in an increasingly shocking environment. Further, it reveals insights into sustainable investments across different markets and regions, highlighting the significance of growth. Furthermore, it reveals the role of finance in the context of sustainable climate issues, with significant implications for investors, regulators, and policymakers.

The literature review is presented in Section 2, the data and methods are covered in Section 3, the findings are discussed in Section 4, and the study is divided into basic sections. Section 5 offers insights and implications.

2. Literature Review

Advances in exchange-traded funds (ETFs) have remained the center of research for sustainable investment stakeholders during the Russia–Ukraine crisis and the financial crisis, especially in the context of risk–return analysis [4,6,18]. Some studies have identified that ETF spillover dynamics can change significantly during crises, affecting risk transfer and hedging diversification effectiveness [17]. However, according to another study, clean energy ETFs, including PBW, QCLN, SMOG, and TAN, become net volatility transmitters during shocks, although OVX acts as a net risk spillover receiver [16]. Further, in the context of clean energy risk management, the investigation highlights the larger hedging return of VXXLE over OVX and the conditional connection between implied volatility ETFs and clean energy ETFs during negative shocks. According to Goodwin, Kanuri [26], compared to S&P 500 ETFs, real estate exchange-traded funds (REETFs) confirmed larger volatility with greater monthly returns. REETFs did not perform as well as traditional ETFs such as IVV and RSP at higher risk levels. In due course, REETFs accumulated large losses and increased risk volatility during the global financial crisis. However, a study observed the market development of ETFs in the Asia-Pacific region [29].

Conversely, some studies have also investigated ETF performance during crises and the emergent significance of governance, social, and environmental (ESG) distress [6,30]. Similarly, clean energy stocks can act as hedging instruments against their duller counterparts in the face of increasing climate-related risks. Furthermore, climate risks have negative effects on the spillover dynamic linkages between dirty and alternative energy stocks [31]. Furthermore, Lin and Swain [21] identified that ESG markets with negative screening produce positive investor surpluses during crises and shocks. These indices indicate robustness even when the market is unstable. Another study reviewed exchange-traded products, covering their taxonomy, risks, and mitigation strategies [32]. However, all these previous studies show the importance of ESG considerations when selecting investment plans, particularly during economic uncertainty when traditional market theories are challenged. Substantial study evidence suggests that sustainable asset classes frequently carry certain risks during market crises [3,30].

Notably, recent research has also noted intensifying consequences of ETF performance during the COVID-19 pandemic [6,19,33]. Meanwhile, Joshi and Dash [34] and bibliometric analysis disclosed a growing interest in research on ETFs, especially the global shocks between 2020 and 2022—the COVID-19 epidemic crisis—by exposing significant trends, such as the rise of machine learning, artificial intelligence (AI), ESG requirements, and ETFs with a sustainability focus. The research findings illustrate how crises can offset innovation and changes in investment strategies. Understanding these trends can assist in making critical choices regarding the future evolution and their appropriate placement in investment portfolios as the market shifts. Further, this study investigates why investing in sustainability and European ETFs adoption factors into sustainable investment [35].

Recent research has also focused on the relationship between energy and financial exchange-traded funds, indicating that spillover effects are more prominent during crises [36,37]. However, a recent study showed unidirectional causation from financial ETF futures to S&P 500 futures and bidirectional causality between financial ETF futures and crude oil futures [38]. A short-term spillover from financial ETF futures to natural gas futures was observed, whereas a long-term spillover from S&P 500 futures to natural gas futures was observed. This demonstrates the importance of understanding how different asset classes interact, as this can substantially affect risk management and investment strategies, particularly in the face of market volatility brought on by crises. According to one study, do climate threats impact the dynamic correlations between dirty and clean energy stock prices? This study observed significant outcomes [31].

Ji and Naeem [17] illustrate how different ETF types become intertwined in times of crisis. Energy ETFs exhibit greater pass-through effects on gold, whereas clean energy ETFs exhibit greater volatility spillovers to the real estate market. In addition to highlighting the complex dynamics between these linkages, peak connections during the COVID-19 pandemic offer crucial information for investors who wish to adjust their portfolios in the case of market shocks. Additionally, D’Ecclesia and Morelli [15] scrutinized the performance of clean energy ETFs in comparison to fossil fuel ETFs in significant events such as the Russia–Ukraine conflict and the Paris Agreement. Research findings reveal that clean energy ETFs often offer a better risk–return trade-off, especially in crises. However, although they do not always perform better than their fossil fuel equivalents, these events signify the strong determinants of portfolio performance. Further, the study explored the tail dependency of green bonds and energy markets, time-varying optimal copulas, and portfolio consequences in a comparable analysis [39]. However, differences across news regimes in ETF liquidity spillovers have been observed [40].

In addition to the critical behavior of definite ETFs, the extensive consequences of international market interdependence in times of crisis are crucial. Yavas, Rezayat [41] examined interdependence between the European, Japanese, and US equity markets and pointed out that international diversification is better in market turbulence when correlations are stronger. This result challenges the idea of market efficiency during the crisis and is consistent with another study that examined the effectiveness of pairs-trading methods within ETFs. Another study evaluated the significance of strategically allocating assets during crises. A well-planned pairs-trading strategy may create steady profits, particularly in down markets [42]. Another study assessed the performance of equity investments in markets for sustainable eco-friendly practices [43]. Further studies have assessed the liquidity responsiveness of South African exchange-traded funds to the implications of national risk [28].

Predominantly, new studies have also examined the interdependence and spillover dynamics of energy and climate with other cross-markets in the Russia–Ukraine conflict [24,25]. However, another study underscores the interconnectedness between sustainable ETFs and geopolitical risks in both crises. We show a link between price declines for sustainable assets and increases in geopolitical concerns. Their findings highlight the substantial effects of political changes on market performance [3]. Further research evaluates how climate risk and policy uncertainty may intensify the dynamics of connectedness between clean and brown energy ETFs during crises [20,44]. Clean energy ETFs can act as hedges for brown energy assets against rising climate risks. This study also highlights the importance of considering climate risk when designing investment strategies for crises.

Finally, the advent of decentralized financing (DeFi) has made scrutinizing ETF performance in times of crisis much more challenging [45]. During the COVID-19 pandemic and the Russia–Ukraine conflict, contagion effects and volatility spillovers were discovered between the G7 ETFs and highly traded DeFi assets. Further, the DeFi and ETF pairs continue to show a weak positive correlation, indicating that DeFi assets can improve risk-managing methods in shocks by providing diversification benefits inside investment portfolios. Several studies have explored the spillover dynamics and interconnectedness of DeFi with Islamic, global equity, and banking markets [46,47].

However, the SFI field faces challenges, including under-theorization and a predominantly short-term financial perspective [13]. Furthermore, sustainable investment and financing tackle long-term ESG issues [11,12]. Several studies have focused on different sectors and areas of ETFs, such as basic ETFs [14], global ETFs [5], energy and clean energy ETFs [15,16,17,18,20], real estate ETFs [26], European ETFs [27], South African ETFs [28], ESG ETFs in COVID-19 [6], and sustainable ETFs in the Ukrainian war [3]. However, there is a significant gap in the literature regarding the performance dynamics of sustainable climate ESG ETFs during geopolitical crises, such as the Russia–Ukraine and Israel–Palestine conflicts. This study aims to address this gap by analyzing the crisis risk dynamics of climate-focused ESG ETFs in turbulent market environments.

3. Data and Methods

3.1. Data and Basic Analysis

In this study, we selected the higher capitalization of the top nine sustainable climate ESG ETFs markets to investigate climate sustainability risk dynamics and growth during the fresh crisis. First, we searched these specific sustainable climate ETF markets and chose smaller-, mid-, and higher-capitalization markets for research generalizability. Second, we selected nine higher-capitalization climate ETF markets based on data availability from July 2020 to October 2024. We also clarified that we excluded ETFs with insufficient data to ensure a robust analysis, thus enhancing the representativeness of our selection within the climate-focused ETF landscape. This study covers the specific time frame of the full sample (16 July 2020–10 October 2024), and the sub-samples are the RU-conflict sample (1 February 2022–30 September 2023) and the Israel–Palestine (IP-conflict) sample (1 October 2023–10 October 2024), as in previous studies [24,47]. Furthermore, we excluded the COVID-19 pandemic period from the analysis because our dataset (July 2020–October 2024) starts after COVID-19’s initial onset, and the pandemic has already been extensively covered in prior studies, making recent geopolitical crises more relevant to our focus. All datasets were extracted from investing.com. The information on the selected study’s climate sustainability ESG ETFs indices from Europe, the US, emerging markets, and the world is given in

Table 1. Climate sustainability ETFs index.

S No.	Ticker	Markets Details Information
1	EMCR	Xtrackers Emerging Markets Carbon Reduction and Climate Improvers ETF (EMCR)
2	D6RR	Deka MSCI Europe Climate Change ESG UCITS ETF (D6RR)
3	D6RP	Deka MSCI World Climate Change ESG UCITS ETF (D6RP)
4	D6RQ	Deka MSCI USA Climate Change ESG UCITS ETF (D6RQ)
5	LWCR	Amundi MSCI World Climate Transition CTB—UCITS ETF DR—EUR-C (LWCR)
6	FLX5	Franklin S&P 500 Paris Aligned Climate UCITS ETF (FLX5)
7	PARI	Franklin STOXX Europe 600 Paris Aligned Climate UCITS ETF (PARI)
8	FLXP	Franklin STOXX Europe 600 Paris Aligned Climate UCITS ETF (FLXP)
9	XAMB	Amundi MSCI World SRI Climate Net Zero Ambition PAB UCITS ETF EUR Acc (XAMB)

.

In

Table 2. Descriptive statistics of the sustainable climate ETF markets.

	EMCR	D6RR	D6RP	D6RQ	LWCR	FLX5	PARI	FLXP	XAMB
Mean	0.025279	0.036786	0.059924	0.071211	0.061121	0.068752	0.044235	0.0448	0.050305
Median	0.00	0.069396	0.112952	0.121322	0.11484	0.120919	0.067317	0.083195	0.127
Maximum	8.570413	4.74732	3.230654	3.899371	3.423295	3.660024	4.240283	5.934613	3.443526
Minimum	−5.19172	−4.28044	−4.4505	−4.82098	−4.26382	−7.84569	−4.51157	−3.80687	−4.78146
Std. Dev.	1.127852	0.933127	0.964109	1.067154	0.917551	1.021503	0.927689	0.937904	0.899945
Skewness	0.285556	−0.20973	−0.31995	−0.26694	−0.32719	−0.69071	−0.2922	−0.04182	−0.37973
Kurtosis	6.618602	5.271583	4.63623	4.4369	4.699013	7.416689	5.543997	5.813466	4.903563
Jarque-Bera	607.8343	241.6779	139.8029	106.4221	150.1357	969.9416	308.5919	358.8272	190.2394
Probability	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00	0.00
ERS	−6.87 ***	−10.52 ***	−6.570 ***	−7.473 ***	−3.395 ***	−13.84 ***	−3.550 ***	−7.172 ***	−9.529 ***
ARCH RESID(−1)^2	0.057 ***	0.122 ***	0.099 ***	0.098 ***	0.115 ***	0.180 ***	0.027 ***	0.136 ***	0.129 ***
Sum	27.47869	39.98661	65.1375	77.40647	66.43847	74.73361	48.08395	48.69717	54.68169
Sum Sq. Dev.	1381.447	945.6077	1009.444	1236.757	914.3026	1133.206	934.6188	955.3142	879.5534

Note: J.B. stands for the Jarque–Bera test statistics for normality. Elliott et al. (1996), the ERS is a unit root test with constants and ARCH used for heteroscedasticity. Statistical significance is indicated by the symbol *** at 1% level.

, the data present summary statistics for nine variables, with positive mean values indicating upward trends. EMCR has the lowest mean (0.025), reflecting modest returns, while D6RQ has the highest (0.071), showing a stronger performance. The maximum values show high variability, especially for EMCR (8.57), but the negative minimum values highlight significant downturns. D6RP (0.060), LWCR (0.061), and FLX5 (0.069) also exhibit strong growth, while PARI (0.044), FLXP (0.045), and XAMB (0.050) show moderate returns. Standard deviations suggest moderate volatility, and most variables have negative skewness, indicating a left-tailed distribution. High kurtosis values indicate heavy tails, particularly for LWCR and FLX5. The Jarque–Bera test rejects normality for all variables, indicating non-normal distribution. Significant volatility clustering is shown using the ERS test. The significant ARCH coefficients indicate that past volatility affects current volatility. The data indicate a notable amount of volatility, non-normality, and volatility clustering in asset returns. Overall, these findings suggest volatility with occasional extreme values.

Table 3. Covariance of sustainable climate ETF market return series.

	EMCR	D6RR	D6RP	D6RQ	LWCR	FLX5	PARI	FLXP	XAMB
EMCR	1.270881	0.02383	0.014179	0.009234	0.010376	2.21 × 10−5	0.056005	0.040543	0.021366
D6RR	0.02383	0.869924	0.714198	0.696535	0.091647	0.666282	−0.01444	0.829975	0.694933
D6RP	0.014179	0.714198	0.928651	0.999076	0.093235	0.896769	−0.02618	0.694265	0.830032
D6RQ	0.009234	0.696535	0.999076	1.137771	0.09946	0.988854	−0.03113	0.672773	0.88676
LWCR	0.010376	0.091647	0.093235	0.09946	0.841125	0.076033	0.119501	0.063917	0.082841
FLX5	2.21 × 10−5	0.666282	0.896769	0.988854	0.076033	1.042508	−0.03677	0.682951	0.820609
PARI	0.056005	−0.01444	−0.02618	−0.03113	0.119501	−0.03677	0.859815	−0.01113	−0.03252
FLXP	0.040543	0.829975	0.694265	0.672773	0.063917	0.682951	−0.01113	0.878854	0.680888
XAMB	0.021366	0.694933	0.830032	0.88676	0.082841	0.820609	−0.03252	0.680888	0.809157

Note: The covariance matrix shows the degree to which the returns of different ETFs markets move together (positive/negative), indicating their co-movement or interdependence.

illustrates the covariance between the nine variables (EMCR, D6RR, D6RP, etc.), where positive values indicate that two variables move together, whereas negative values suggest inverse relationships. EMCR has a weak covariance with other variables, particularly close to zero with FLX5 (0.00002), indicating minimal co-movement. D6RP and D6RQ have the strongest positive covariance (0.999), meaning that they move almost identically. FLX5 also showed a strong covariance with D6RP (0.897) and D6RQ (0.989), highlighting similar performance trends. LWCR has low covariance with most variables, reflecting limited co-movement, whereas PARI shows a negative covariance with most, indicating inverse relationships with variables such as FLX5 and D6RP. FLXP and XAMB have moderate positive covariances with several variables, such as D6RP and D6RQ, suggesting some interconnectedness. The table shows the varying degrees of co-movement, with D6RP, D6RQ, and FLX5 having the most significant positive relationships. To diversify risk, combining EMCR, PARI, and LWCR with more correlated assets such as D6RP, D6RQ, and FLX5 could create a more balanced portfolio. Conversely, focusing heavily on D6RP, D6RQ, and FLX5 may offer higher returns, but at the cost of increased volatility.

Figure 1 shows upward trends (growth) for most variables (D6RP, D6RQ, LWCR, FLX5, FLXP, and XAMB), indicating general growth over time, particularly since 2022. EMCR and D6RR exhibit some volatility, with periods of rise and fall before stabilization. The PARI followed a similar upward trend, but with notable fluctuations, suggesting moderate growth with intermittent corrections. In Figure 2, the return series trends across all variables (EMCR, D6RR, D6RP, etc.) show high volatility with frequent fluctuations around zero, as well as short-term volatility in returns over time. EMCR and FLXP highlight the periods of sudden significant gains or losses.

3.2. Econometric Analysis a TVP-VAR Approach

Several prior studies on interconnectedness dynamics have utilized a conventional time-series model to examine connections. However, these models do not completely demonstrate the impact of price change on interdependence. Following the spillover methods of Diebold and Yilmaz [48] and Diebold and Yılmaz [49], we utilize TVP-VAR, which eliminates the requirement for a rolling window of arbitrary size. This removes the possibility of inaccurate observations and inconsistent or unchanged variables, enabling researchers to study interactive relationships using limited time-series data [50]. The benefit of the TVP-VAR method over other approaches (such as GARCH) is that it eliminates the weight of the often-arbitrary rolling window size, which can lead to flattened or extremely unpredictable parameters and the loss of significant observations, as discussed in [47]. Furthermore, TVP-VARs incorporate stochastic volatility for time-varying error variances and enable time-varying coefficients to reflect nonlinearity in macroeconomic time series. The fast TVP-VAR model estimates score-driven volatilities [51]. The initial step includes generating a categorized series from the low- and high-frequency segments using the TVP-VAR model, as illustrated below:

$${\mathrm{y}}_{\mathrm{t}}={\mathsf{\beta }}_{\mathrm{t}}{\mathrm{z}}_{\mathrm{t}-1}+{\in }_{\mathrm{t}};{\in }_{\mathrm{t}}\left|{\mathrm{F}}_{\mathrm{t}-1}\right.~\mathrm{N}(0,{\mathrm{S}}_{\mathrm{t}})$$

(1)

$$\mathrm{v}\mathrm{e}\mathrm{c}\left({\mathsf{\beta }}_{\mathrm{t}}\right)=\mathrm{v}\mathrm{e}\mathrm{c}\left({\mathsf{\beta }}_{\mathrm{t}-1}\right)+{\mathrm{v}}_{\mathrm{t}};{\mathrm{v}}_{\mathrm{t}}\left|{\mathrm{F}}_{\mathrm{t}-1}\right.~\mathrm{N}(0,{\mathrm{R}}_{\mathrm{t}})$$

(2)

Here, ${\mathrm{y}}_{\mathrm{t}}$ and ${\mathrm{z}}_{\mathrm{t}}={[{\mathrm{y}}_{\mathrm{t}-1},\dots ,{\mathrm{y}}_{\mathrm{t}}-\mathrm{p}]}^{\prime }$ represent $\mathrm{N}\times 1$ and $\mathrm{P}\times 1$ dimensional vectors, respectively. ${\mathsf{\beta }}_{\mathrm{t}}$ is an $\mathrm{N}\times {\mathrm{N}}_{\mathrm{p}}$ dimensional time-varying coefficient matrix and $\mathrm{N}\times {\mathrm{N}}_{\mathrm{p}}$ dimensional error disturbance vector with an $\mathrm{N}\times \mathrm{N}$ time-varying variance-covariance matrix ${\mathrm{S}}_{\mathrm{t}},\mathrm{v}\mathrm{e}\mathrm{c}\left({\mathsf{\beta }}_{\mathrm{t}}\right)$. Meanwhile, ${\mathrm{v}}_{\mathrm{t}}$ represents ${\mathrm{N}}_{\mathrm{p}}^2\times 1$ dimensional vectors, and ${\mathrm{R}}_{\mathrm{t}}$ is an ${\mathrm{N}}_{\mathrm{p}}^2$ × ${\mathrm{N}}_{\mathrm{p}}^2$ dimensional matrix.

Afterwards, the VAR system is converted into its vector moving average form, and the generalized impulse response functions (GIRF) and generalized forecast error variance decomposition (GFEVD) are computed according to [52,53]:

$${\mathrm{y}}_{\mathrm{t}}={\sum }_{\mathrm{j}=0}^{\mathrm{\infty }}{\mathrm{L}}^{\prime }{\mathrm{W}}_{\mathrm{t}}^{\mathrm{j}}{\mathrm{L}\in }_{\mathrm{t}-\mathrm{j}}$$

(3)

$${\mathrm{y}}_{\mathrm{t}}={\sum }_{\mathrm{j}=0}^{\mathrm{\infty }}{\mathrm{A}}_{\mathrm{i}\mathrm{t}}{\in }_{\mathrm{t}-\mathrm{j}}$$

(4)

where $={[{\mathrm{I}}_{\mathrm{N}},\dots ,0_{\mathrm{p}}]}^{\prime }$ is an ${\mathrm{N}}_{\mathrm{p}}\times \mathrm{N}$ dimensional matrix, $\mathrm{W}=[\mathsf{\beta }\mathrm{t};{\mathrm{I}}_{\mathrm{N}(\mathrm{p}-1),}{0}_{\mathrm{N}(\mathrm{p}-1)\times \mathrm{N}}]$ is an ${\mathrm{N}}_{\mathrm{p}}\times {\mathrm{N}}_{\mathrm{p}}$ dimensional matrix, and ${\mathrm{A}}_{\mathrm{i}\mathrm{t}}$ is an $\mathrm{N}\times \mathrm{N}$ dimensional matrix. GIRFs illustrate how all variables react to a shock in variable i. An absence of a structural model results in the duplication of calculations for a J-step-ahead forecast, first when variable i is shocked and then when it is not. The variation is attributed to a disturbance in variable i, and is subsequently calculated.

$${\mathrm{G}\mathrm{I}\mathrm{R}\mathrm{F}}_{\mathrm{t}}\left(\mathrm{K},{\mathsf{\delta }}_{\mathrm{j},\mathrm{t}}{\mathrm{F}}_{\mathrm{t}-1}\right)=\mathrm{E}\left({\mathrm{y}}_{\mathrm{t}+\mathrm{K}}\left|{\in }_{\mathrm{j},\mathrm{t}}\right.={\mathsf{\delta }}_{\mathrm{j},\mathrm{t}}{\mathrm{F}}_{\mathrm{t}-1}\right)-\mathrm{E}\left({\mathrm{Y}}_{\mathrm{t}+\mathrm{K}}\left|{\mathrm{F}}_{\mathrm{t}-1}\right.\right)$$

(5)

$${\mathsf{\psi }}_{\mathrm{j},\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)={\displaystyle \frac{{\mathrm{A}}_{\mathrm{K},\mathrm{t}}{\mathrm{S}}_{\mathrm{t}}{\in }_{\mathrm{j},\mathrm{t}}}{\sqrt{{\mathrm{S}}_{\mathrm{j},\mathrm{j},\mathrm{t}}}}}{\displaystyle \frac{{\mathsf{\delta }}_{\mathrm{j},\mathrm{t}}}{\sqrt{{\mathrm{S}}_{\mathrm{j},\mathrm{j},\mathrm{t}}}}}{\mathsf{\delta }}_{\mathrm{j},\mathrm{t}}=\sqrt{{\mathrm{S}}_{\mathrm{j},\mathrm{j},\mathrm{t}}}$$

(6)

$${\mathsf{\psi }}_{\mathrm{j},\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)={\displaystyle \frac{{\mathrm{A}}_{\mathrm{K},\mathrm{t}}{\mathrm{S}}_{\mathrm{t}}{\in }_{\mathrm{j},\mathrm{t}}}{\sqrt{{\mathrm{S}}_{\mathrm{j},\mathrm{j},\mathrm{t}}}}}$$

(7)

where ${\mathsf{\psi }}_{\mathrm{j},\mathrm{t}}^{\mathrm{g}}$ GIRFs represent the j-th variable and K represents the forecast horizon. In the meantime, ${\mathsf{\delta }}_{\mathrm{j},\mathrm{t}}$ denotes the selection vector with 1 in the jth position and 0 otherwise, and ${\mathrm{F}}_{\mathrm{t}-1}$ indicates the information set until t − 1. Afterwards, the GFEVD, which measures the variance shares of one variable on others, can be calculated in the following way:

$$\tilde{\mathsf{\Phi }}{}_{\mathrm{i}\mathrm{j},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)={\displaystyle \frac{{\sum }_{\mathrm{t}=1}^{\mathrm{K}-1}{\mathsf{\psi }}_{\mathrm{j},\mathrm{t}}^{2,\mathrm{g}}}{{\sum }_{\mathrm{j}=1}^{\mathrm{N}}{\sum }_{\mathrm{t}=1}^{\mathrm{K}-1}{\mathsf{\psi }}_{\mathrm{j},\mathrm{t}}^{2,\mathrm{g}}}}; {\sum }_{\mathrm{j}=1}^{\mathrm{N}}\tilde{\mathsf{\Phi }}{}_{\mathrm{i}\mathrm{j},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)=1\mathrm{a}\mathrm{n}\mathrm{d}{\sum }_{\mathrm{j}=1}^{\mathrm{N}}{\mathrm{N}}_{\mathrm{i}\mathrm{j},\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)=\mathrm{N}$$

(8)

Utilizing Equation (8), an investigation into the interconnectedness of sustainable climate ESG-ETF prices can be conducted through the construction of the total connectedness index (TCI).

$${\mathrm{C}}_{\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)={\displaystyle \frac{{\sum }_{\mathrm{i},\mathrm{j}=1\mathrm{i}\ne \mathrm{j}}^{\mathrm{N}}\tilde{\mathsf{\Phi }}{}_{\mathrm{i}\mathrm{j},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)}{\mathrm{N}}}\ast 100$$

(9)

Investigating directional connectedness is fascinating. The technique considered in this study considers three facets of this approach. The initial aspect was the overall level of connectivity with others in a specific direction.

$${\mathrm{C}}_{\mathrm{i}\to \mathrm{j},\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)={\displaystyle \frac{{\sum }_{\mathrm{i},\mathrm{j}=1\mathrm{i}\ne \mathrm{j}}^{\mathrm{N}}\tilde{\mathsf{\Phi }}{}_{\mathrm{i}\mathrm{j},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)}{{\sum }_{\mathrm{i},\mathrm{j}=1}^{\mathrm{N}}\tilde{\mathsf{\Phi }}{}_{\mathrm{i}\mathrm{j},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)}}\ast 100$$

(10)

The next factor is the overall level of connectedness in a specific direction from that of the other individuals, represented as:

$${\mathrm{C}}_{\mathrm{i}\leftarrow \mathrm{j},\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)={\displaystyle \frac{{\sum }_{\mathrm{i},\mathrm{j}=1\mathrm{i}\ne \mathrm{j}}^{\mathrm{N}}\tilde{\mathsf{\Phi }}{}_{\mathrm{i}\mathrm{j},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)}{{\sum }_{\mathrm{j}=1}^{\mathrm{N}}\tilde{\mathsf{\Phi }}{}_{\mathrm{i}\mathrm{j},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)}}\ast 100$$

(11)

Finally, we subtract Equation (11) from Equation (10) to find the total net directional value interconnectedness as follows:

$${\mathrm{C}}_{\mathrm{i},\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)={\mathrm{C}}_{\mathrm{i}\to \mathrm{j},\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)-{\mathrm{C}}_{\mathrm{i}\leftarrow \mathrm{j},\mathrm{t}}^{\mathrm{g}}\left(\mathrm{K}\right)$$

(12)

It should be noted that Equation (12) shows how prices in the ith category affect the analyzed network. Therefore, if Equation (12) is positive, prices in the ith have a greater impact on the network, while a negative value signifies that the network has a stronger effect on prices in the ith. Finally, bidirectional connections are further analyzed by calculating the net pairwise directional connectedness (NPDC) as follows:

$${\mathrm{N}\mathrm{P}\mathrm{D}\mathrm{C}}_{\mathrm{i},\mathrm{j}}\left(\mathrm{K}\right)=\tilde{\mathsf{\Phi }}{}_{\mathrm{i}\mathrm{j},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)-\tilde{\mathsf{\Phi }}{}_{\mathrm{j}\mathrm{i},\mathrm{t}}{}^{\mathrm{g}}\left(\mathrm{K}\right)$$

(13)

Based on Equation (13), a positive NPDC value shows that prices in j are overshadowed by prices in i, whereas a negative NPDC value shows that prices in j overshadow prices in i.

4. Findings and Discussion

4.1. Dynamic Total Connectedness Index Analysis

This total connectedness (

Table 4. Static and dynamic connectedness of sustainable climate ETF markets.

Israel–Palestine Conflict Sample (1 October 2023–10 October 2024)
	EMCR	D6RR	D6RP	D6RQ	LWCR	FLX5	PARI	FLXP	XAMB	FROM
EMCR	94.58	0.53	0.79	0.85	0.18	0.97	0.48	0.73	0.88	5.42
D6RR	0.09	23.24	14.65	11.27	0.31	12.38	0.58	21.83	15.65	76.76
D6RP	0.23	12.34	19.62	18.64	0.15	18.44	0.22	11.75	18.6	80.38
D6RQ	0.28	10.19	20.07	21.12	0.17	19.61	0.22	9.56	18.78	78.88
LWCR	0.26	12.48	19.13	18.31	0.33	18.49	0.23	12.01	18.75	99.67
FLX5	0.35	10.93	19.34	19.08	0.14	20.56	0.18	10.55	18.86	79.44
PARI	0.42	2.46	2.16	1.92	0.14	2	85.64	2.85	2.41	14.36
FLXP	0.14	22.46	14.22	10.75	0.3	12.13	0.78	23.87	15.34	76.13
XAMB	0.29	13.16	18.57	17.43	0.19	17.95	0.21	12.63	19.57	80.43
TO	2.05	84.56	108.94	98.24	1.59	101.97	2.91	81.92	109.28	591.46
Inc.Own	96.64	107.8	128.56	119.36	1.92	122.54	88.55	105.78	128.85	cTCI/TCI
NET	−3.36	7.8	28.56	19.36	−98.08	22.54	−11.45	5.78	28.85	73.93/65.71
NPT	1	4	7	5	0	6	2	3	8
Russia–UkraineConflict Sample (1 February 2022–30 September 2023)
EMCR	80.8	1.77	2.92	2.74	2.32	3.26	1.1	1.94	3.14	19.2
D6RR	0.18	26.93	10.82	7.31	8.6	6.94	3.58	20.69	14.95	73.07
D6RP	0.31	8.4	20.79	19.44	12.85	14.69	1.73	6.87	14.92	79.21
D6RQ	0.33	6.17	21.03	22.51	13.14	16.03	1.38	4.95	14.46	77.49
LWCR	0.22	9.82	15.68	14.95	20.36	13.5	1.24	8.44	15.81	79.64
FLX5	0.65	6.3	16.97	17.09	12.53	24.4	1.83	6.75	13.48	75.6
PARI	1.05	8.44	5.09	3.67	3.53	5.35	59.18	7.92	5.78	40.82
FLXP	0.14	22.34	9.54	6.27	7.31	7.96	3.61	29.23	13.6	70.77
XAMB	0.33	12.15	15.39	13.8	12.06	12.17	1.98	10.42	21.71	78.29
TO	3.19	75.4	97.42	85.26	72.33	79.9	16.46	67.98	96.16	594.09
Inc.Own	84	102.33	118.21	107.77	92.69	104.3	75.64	97.21	117.87	cTCI/TCI
NET	−16	2.33	18.21	7.77	−7.31	4.3	−24.36	−2.79	17.87	74.26/66.01
NPT	0	4	8	6	2	5	1	3	7
Full Sample (16 July 2020–10 October 2024)
EMCR	88.23	1.25	1.65	1.54	0.92	1.68	1.87	1.14	1.73	11.77
D6RR	0.2	22.45	14.91	11.63	1.72	11.57	1.12	20.34	16.06	77.55
D6RP	0.16	12.71	19.61	18.55	2.01	16.41	0.61	11.9	18.05	80.39
D6RQ	0.15	10.67	20	21.08	2.03	17.5	0.52	9.92	18.13	78.92
LWCR	0.17	12.83	18.69	17.85	3.13	16.37	0.6	12.14	18.23	96.87
FLX5	0.26	10.77	17.96	17.82	2.02	22.17	0.48	11.37	17.15	77.83
PARI	1.18	3.88	3.73	3.46	0.85	3.56	74.95	4.33	4.06	25.05
FLXP	0.2	20.91	14.19	10.97	1.54	12.26	1.32	23.16	15.46	76.84
XAMB	0.17	13.73	18.15	16.91	1.94	15.76	0.68	12.96	19.7	80.3
TO	2.5	86.74	109.28	98.74	13.04	95.09	7.19	84.09	108.87	605.53
Inc.Own	90.72	109.19	128.89	119.81	16.16	117.26	82.14	107.25	128.57	cTCI/TCI
NET	−9.28	9.19	28.89	19.81	−83.84	17.26	−17.86	7.25	28.57	75.69/67.28
NPT	0	4	8	6	2	5	1	3	7

Note: The Total Connectedness Index measures overall spillover strength and risk interconnectedness among markets in a system. NET values help to identify whether a market is a risk spillover transmitter or receiver. The term “TO” describes a market subject to risk spillover from another market. The term “FROM” refers to the market causing risk spillover to another market.

) provides insights into the risk spillover dynamics among several markets during the Israel–Palestine conflict (October 2023–October 2024). The total connectedness index (TCI) is 65.71%, reflecting a significant level of interdependence between these markets, with approximately two-thirds of the risk shared across markets. TO values indicate how much risk is transmitted from one market to others, while FROM values show how much risk a market absorbs from others. NET values help to identify whether a market is a risk spillover transmitter or receiver. Markets such as D6RR (7.8%), D6RP (28.56%), and XAMB (28.85%) have positive NET values, signifying that they are net transmitters of risk during this period, contributing to heightened risk across other markets, which is consistent with previous studies [4,15]. This could signify to investors that these ETFs are more susceptible to market shocks and, as a result, are preferred for risk-taking or hedging strategies when volatility is high. Moreover, Ji and Naeem [17] illustrated how various interdependent ETF types become during times of crisis. On the other hand, markets such as EMCR (−3.36%) and LWCR (−98.08%) have negative NET values, marking them as net receivers and absorbing more risk than they transmit. Such ETFs may provide a stabilizing effect in a portfolio, appealing to risk-averse investors seeking defensive assets during crises. Regarding individual contributions to connectedness, D6RR and D6RP show substantial transmission with high TO values of 84.56 and 108.94, respectively, indicating their prominent role as risk disseminators. With a noteworthy FROM value of 99.67, the LWCR was identified as a significant risk receiver. These insights help us understand how risks circulated across these markets during the conflict, with certain markets acting as sources and others as destinations of financial instability.

The Russia–Ukraine conflict (February 2022–September 2023) shows a TCI of 66.01%, indicating that nearly two-thirds of the risk across these markets is interconnected. This suggests a moderate-to-high level of risk transmission and spillovers between the markets during this period of geopolitical tension and is similar to previous studies’ analyses [18,36]. Furthermore, D’Ecclesia and Morelli [15] examined how clean energy ETFs performed concerning fossil fuel ETFs throughout momentous occasions, such as the conflict in Ukraine. Markets such as D6RP (18.21%), XAMB (17.87%), and D6RQ (7.77%) are net transmitters, meaning that they spread more risk to other markets than they absorb. This indicates that ETFs may be more responsive to market shocks, making them suitable for risk-seeking or hedging strategies in times of high volatility. Conversely, EMCR (−16%), LWCR (−7.31%), and PARI (−24.36%) are net receivers, indicating that they absorb more risk than they transmit. Notably, the PARI shows a significantly negative NET value, making it a major risk absorber during conflict. These ETFs could have a stabilizing effect on a portfolio, making them attractive to risk-averse investors seeking defensive assets during crises. In terms of contributions to overall connectedness, D6RP, XAMB, and D6RQ had high TO values of 97.42, 96.16, and 85.26, respectively, indicating that they are key risk spreaders. However, EMCR has a high FROM value of 19.2, indicating that it is one of the largest recipients of risk spillover. Overall, the table highlights a complex risk-sharing dynamic, with some markets acting as major sources of financial instability, while others are more vulnerable to external shocks during the Russia–Ukraine conflict, similar to previous studies [1,3].

The total connectedness (Table 4) for the full sample period (July 2020–October 2024) presents a TCI of 67.28%, indicating that nearly two-thirds of the total risk in the system is shared among the markets. This level of interdependence suggests that these markets are significantly exposed to spillovers, likely due to various global crises during this period, including the Russia–Ukraine and Israel–Palestine conflicts. However, one study evaluated equity investment performance in sustainable environmental markets [43]. Focusing on NET values, which highlight whether a market is a risk transmitter or receiver, we observe that D6RP (28.89%), XAMB (28.57%), and D6RQ (19.81%) are significant net transmitters of risk, indicating that they are more active in spreading risk across the system. This suggests that certain ETFs may be more sensitive to market shocks, making them good choices for hedging or risk-seeking strategies when volatility is high. Because these ETFs potentially increase portfolio risk exposure, stakeholders, particularly institutional investors and portfolio managers, need to closely monitor this response to shocks. Since increased volatility may affect both short-term returns and long-term investment objectives, it is crucial to comprehend this behavior in order to make well-informed judgments on asset allocation. To align their strategies with more general risk management goals, stakeholders may also consider the risk-transmitting tendencies of these ETFs in the context of overall market stability.

Conversely, LWCR (−83.84%), EMCR (−9.28%), and PARI (−17.86%) are net receivers, indicating that they absorb more risk than other markets. Notably, the LWCR has a highly negative NET value, making it a major risk absorber during this period. An ETF with negative NET values absorbs volatility, provides stability, and attracts risk-averse investors. These ETFs offer stakeholders, including institutional investors, protection from market volatility. However, during recuperation, their low response rate may impede progress. As stakeholders adjust asset allocation, they must strike a balance between risk mitigation and possible growth. In terms of contributions to overall connectedness, D6RP and XAMB exhibit high TO values of 109.28 and 108.87, respectively, positioning them as key risk spreaders within the system. In contrast, LWCR and PARI show high FROM values of 96.87 and 25.05, respectively, making them major recipients of risk spillover. Overall, this table provides a comprehensive view of risk dynamics, with certain markets acting as central nodes for transmitting instability, while others remain highly vulnerable to external shocks, consistent with previous studies [21,24]. However, some studies have also explored how policy uncertainty and climate risk affect ETF dynamics in times of crisis [20,44]. Furthermore, based on SFI theories, participants work together to drive positive social and environmental impacts through investment [13]. However, according to shareholder theory, SFI methods result in significant opportunity and engagement costs and are incompatible with maximizing shareholder wealth [54].

The TCI values across the three crises showed varying levels of market interconnectedness. During the Israel–Palestine conflict, the TCI was 65.71%, while in the Russia–Ukraine conflict, it rose slightly to 66.01%. Over the full sample period, the TCI reached 67.28%, reflecting a consistent increase in market connectedness over time as the crises evolve. This suggests that geopolitical conflicts have gradually intensified the interdependence between these markets, with the full period capturing the cumulative spillover effects. Looking at NET values, the behavior of individual markets also shifts between crises. D6RP and XAMB remained net transmitters across all periods, although their influence increased over the full sample (28.89% and 28.57%, respectively). However, the LWCR shifted from a minor net receiver in the Israel–Palestine conflict (−98.08%) to a significant absorber during the full period (−83.84%). However, contemporary portfolio theory suggests that nonfinancial screening is helpful for ideal financial outcomes [8]. This indicates that certain markets become more vulnerable as global crises go on, while others maintain their roles as primary risk disseminators during different geopolitical events. Expanding the practical implications to include targeted portfolio adjustments would offer SFI investors actionable insights into volatile geopolitical climates. SFI investors should also review their portfolio allocations on a regular basis, using the most recent TCI, TO, FROM, and NET values of specific climate ETFs that reflect changes in risk roles and market interconnection. Furthermore, as ESG-focused ETFs frequently demonstrate exceptional resilience, merging them with stable risk roles may align with sustainable investing objectives.

In Figure 3, The TCI graph shows relatively stable levels from 2021 to mid-2023, indicating consistent interconnectedness among the top nine high-capitalization sustainable climate ETFs. However, there is a significant rise in the index in 2024, which likely corresponds to increased market volatility and risk spillovers due to the Russia–Ukraine and Israel–Palestine conflicts. These geopolitical tensions have caused heightened interdependencies among markets, which is reflected in the sharp upward spike, suggesting that crises have increased cross-market connectedness, particularly in the context of ESG investments. However, some studies have explored ESG concerning ETF performance in times of crisis and found significant outcomes [6,30]. The upward trend in the Total Connectedness Index indicates heightened market correlation during geopolitical stress, which may diminish the effectiveness of diversification strategies. SFI investors should adjust their portfolios to mitigate risks and capitalize on potential opportunities arising from these connections among climate ETFs during turbulent periods, which is consistent with previous studies’ findings [35,39].

4.2. Net Connectedness Analysis

In Figure 4, the net connectedness index of climate ETFs during the Russia–Ukraine and Israel–Palestine conflicts reveals significant trends in market interdependence, which can be observed through the provided figures. During these conflicts, the graphs show times of increased market connectedness and risks as well as variations in the connectedness index for different ETFs. The Israel–Palestine conflict caused a visible surge in LWCR-ETF’s market connectedness, indicating a sharp investor interest in sustainable investments in the face of international conflicts. Investors may look for climate-resilient assets as a hedge against wider market volatility during times of crisis, as evidenced by the market D6RQ-ETF’s brief upward trend. There is a minor increase in connectivity in the EMCR-ETF market, which reflects increased links with global markets. However, in times of crisis, OVX acts as a net receiver, whereas clean energy exchange-traded funds (ETFs) such as PBW, QCLN, SMOG, and TAN become net transmitters of volatility [16].

The deteriorating pattern of the D6RR ETF markets suggests divergence from other ETF markets. When international tensions are high, investors have less faith in regional climate initiatives. The steadily declining connectedness index for the FLXP-ETF market indicates that external disputes have less of an impact on company performance, perhaps as a result of its emphasis on long-term sustainability. The D6RP-ETF market exhibits resilience in crisis because it signifies a steady index, suggesting that it may be less affected by market swings. The relatively flat FLX5-ETFs market connectedness index indicates that this fund sticks to its investment plan without experiencing any notable changes as a result of outside shocks. Similar to FLX5-ETFs, XAMB-ETFs markets exhibit little volatility, suggesting that despite geopolitical conflicts, climate risks are seen as reliable investments. The results show parallels with earlier research in the setting of connectedness during crises [3,32,38].

However, certain climate ESG ETFs show resilience or decoupling effects in crises, whereas others show stronger connectedness. These patterns simplify investors’ use of these approaches in response to the perceived risk of diversification and higher returns. These patterns might help direct future investment choices and explore how crucial adaptability is for climate-related portfolios in the face of geopolitical conflicts between Russia and Ukraine and Israel and Palestine. Furthermore, SFI investors can impact investments based on portfolio theory and SFI theories [8,13]. Therefore, the analysis of the net connectedness index provides important insights into how climate ESG ETFs respond to significant global events by clarifying the complex interaction between sustainability performance and ESG-ETF market spillover dynamics. We could solve this problem by providing more targeted and feasible alternatives for SFI investors navigating unstable geopolitical markets, thus broadening the practical consequences. For instance, investors might benefit from focusing on ETFs identified as consistent risk transmitters such as D6RP and XAMB, which could serve as hedging tools. Another option available to SFI investors is a dynamic rebalancing strategy that modifies exposure in response to current market interdependence patterns.

4.3. Network Connectedness Approach

A network graph displaying the spillover dynamics of climate ESG ETFs during the Israel–Palestine conflict is shown in Figure 5, along with a complete sample that centers on the LWCR ETF market as a transmitting node. Transmitting risk through spillover strong influence is demonstrated by the numerous relationships with other ETFs, such as D6RQ, EMCR, and XAMB, suggesting that these funds are greatly impacted by its performance. Because of their sensitivity to changes in the LWCR ETFs market throughout the conflict, risk-receiving spillover ETFs such as FLXP, PARI, and F5S are positioned as risk spillover-receiving nodes. According to the links, investors are focused on the LWCR ETFs market as geopolitical conflict intensifies, resulting in synchronous movements among the receiving nodes. Although Goodwin and Kanuri [26] displayed more net volatility, REETFs produced higher monthly returns than S&P 500 ETFs.

Further, network connectedness between several climate ESG ETFs in the Russia and Ukraine conflict emphasizes both transmitting and receiving nodes. As core nodes, EMCR-ETFs and D6RQ-ETFs, for instance, send out market signals that impact the performance of nearby ETF funds. As receiving nodes, ETFs, such as the LWCR and PARI ETFs markets, react to the fluctuations that transfer nodes start. The larger market dynamics brought about by geopolitical events impact their success. The degree of connectedness is represented by the thickness of the lines connecting nodes. ETFs are more likely to move in a crisis, which signifies higher and stronger connectedness. However, in times of crisis, ETF dynamics can change significantly, affecting risk transfer and hedging effectiveness [17].

Speculators and investors can more effectively control the risks associated with market volatility during crises by determining the ETFs that serve as transmitting or receiving nodes. The focus on climate-related assets indicates that interest in sustainable investing is persistent, even during times of conflict, which is consistent with a larger trend toward ESG considerations. Additionally, SFI investors can influence higher returns and sustainable impacts through investments, as suggested by SFI and portfolio theories [8,13]. All things considered, the diagram is a useful resource for comprehending the intricate interactions between climate ESG ETFs during the war between Russia and Ukraine, emphasizing the significance of network dynamics in investing choices. SFI investors may find it advantageous to concentrate on ETFs recognized as reliable risk transmitters, such as XAMB and D6RP, as they could be useful hedging instruments. Furthermore, ETFs such as LWCR, which move from receivers to absorbers, should assist investors in modifying their portfolios during protracted crises.

4.4. Total Connectedness Index of the Full Sample as a Robustness

For the top nine higher-capitalization climate ESG ETFs, Figure 6 shows the robustness of market interdependencies over time using the total connectedness index (TCI) with a window rolling size of 20. From 16 July 2020 to 10 October 2024, the TCI showed a high degree of connectedness and a normally consistent performance among these ETFs, indicating strong relationships. These ETFs are said to be robust to market shocks if their TCI is strong, retaining their interconnections even during conflict events. Speculators and investors can use TCI volatility for market prediction to evaluate the possible risks to their relevant sustainable climate-focused investments. This connects to a perspective on how strongly these links enable more effective risk management and smart asset allocation in climate ESG-ETF investments.

5. Conclusions

This study highlights how sustainable ESG-ETF financial markets become increasingly interconnected, as well as the spillover dynamics during fresh geopolitical conflicts. The stronger effects of several crises are underscored in the TCI values, which shows that the risk transmission between markets increases more intensely in conflict events, as revealed by looking backward. Climate ESG ETF markets, such as D6RP and XAMB, have continuously acted as risk transmitters, indicating that they are important causes of instability during the crisis. As geopolitical conflicts continue, some markets become more susceptible to external shocks, as evidenced by LWCR’s ETFs market transformation from a minor to a large risk spillover absorber. ETF markets, such as D6RP and XAMB, are considered essential in the risk spillover dynamics of the global financial system. This is highlighted by their persistent status as a net risk spillover transmitter throughout all subsamples. In contrast, the high-risk absorption by the LWCR ETFs market in the full sample indicates that market behaviors can be altered intensely as crises increase, underscoring the significance of observing how market roles change during protracted geopolitical events. This study gives SFI stakeholders important information for how to improve decision making by shedding light on how financial markets respond to various crises. The findings highlight insights for SFI market participants to modify their approaches in reaction to shifting risk patterns and market dynamics during periods of geopolitical conflict.

However, important information for speculators and investors is revealed by the covariance analysis of the nine ESG ETFs markets. ETFs markets D6RP, D6RQ, and FLX5 have substantial positive correlations, indicating higher returns and making the portfolio more volatile. Conversely, combining lower covariance indicators of ETF markets, such as EMCR, PARI, and LWCR, can aid in risk diversification and portfolio balance. Further, to reduce possible losses during market shocks, investors can consider a strategy that blends highly performing assets with less connected options. Speculators can profit from the trends in closely connected ETFs using strong co-movements that have been identified as opportunities for short-term trading techniques. Finally, managing these interrelated markets requires a well-informed sustainable investment strategy that strikes a balance between risk and return. This study supports the ongoing initiatives to increase the sustainability of financial systems. This study also provides a framework for directing and refining future SFI investment and research initiatives.

Certainly, the study demonstrates that market connectedness increases during times of crisis, and investors should consider integrating geopolitical risk considerations into their frameworks to make investment decisions. Losses can be minimized by shifting investments to less volatile markets and lowering exposure to risk transmitters such as the ETF markets D6RP and XAMB. Long-term investors should also concentrate on locating risk-absorbing ETF markets, such as LWCR, as these could present opportunities for safer investments when uncertainty is high. The results should be used by speculators to profit from transient volatility in risk-transmitting ETF markets such as D6RP and XAMB. The high TO values of these markets offer opportunities for higher speculative trading activities. However, during geopolitical crises, market roles can shift quickly, with some markets shifting to risk-absorbing positions. Consequently, SFI speculators should continue to be careful and flexible, as swift reallocation is crucial for optimizing higher returns.

Notably, the major risk-transmitting ETF markets, especially the D6RP and XAMB markets, have confirmed a continuous risk transmission of conflict instability as backward-looking. However, forward-looking SFI stakeholders can improve their risk management tactics in this study. Further, to minimize the effects of market shocks, the significant ETF market exposure to these markets and portfolio allocation flexibility should be diversified. This helps us to understand how some ETF markets, like LWCR, change from being minor to significant risk absorbers and how stakeholders can modify their strategies as crises using a prioritized approach. Furthermore, states and international policymakers should note the growing interconnection of markets to reduce the effects of spillover during geopolitical conflicts. However, to lower systemic risk, regulatory frameworks should concentrate on stabilizing important risk-transmitting ETF markets such as D6RP and XAMB. Additionally, to identify early indicators of ETF market vulnerability in markets, such as the LWCR, deterrent measures should be formulated and market surveillance improved.

Future research studies could concentrate on optimizing the approach to how green energy bonds might act with climate ESG ETFs during such disputes, which could improve sustainable investment tactics. Future research can also outline ESG ETFs in crises in developed and emerging markets. Additionally, this study’s dataset is limited to a specific time and can be extended for further transparency with specific analysis approaches.