Institutional Quality and Climate Vulnerability: Empirical Evidence from GCC Economies

Abstract

In recent years, the intersection of institutional quality and climate vulnerability has emerged as a critical yet largely untapped area of study, despite its profound implications for understanding resilience to environmental challenges and sustainable development. The purpose of this paper is to establish a relationship between the quality of domestic institutions and climate vulnerability in the case of “Gulf Cooperation Council (GCC)” economies. Annual data spanning the period 2002–2021 were sourced from the “World Governance Indicators (WGI)”, “World Development Indicators (WDI)”, and the “Notre Dame Global Adaptation Initiative (ND-GAIN)”, providing a valuable trace of examined variables. We have applied several econometric techniques including the “Pooled Least Squares (PLS)”, “Fixed Effects (FET)”, “Feasible Generalized Least Squares (FGLS)”, and “Two Stages Least Squares (2SLS)” to estimate the specified models and extract results. Our findings indicate that enhanced institutional quality significantly reduces climate vulnerability in GCC economies. In other words, effective climate governance practices in GCC countries have successfully mitigated climate vulnerability across these economies. The causality analysis confirmed the one-way causality running from institutional quality towards climate vulnerability. On the other hand, increased income level, urbanization, and the degree of trade openness are the major threats as their impacts on climate vulnerability are positive and statistically significant. The results obtained offer valuable insights for policymakers in GCC economies seeking to formulate effective policies addressing climate vulnerability.

1. Introduction

Climate risk and vulnerabilities arise as a rational consequence of the aggravated pressures of achieving sustainable economic growth. This growing concern underscores the need for comprehensive strategies to mitigate these risks. In recent decades, climate vulnerability has received higher attention from both policymakers and researchers due to its enormous adverse effects on the global economy, humans, and nature. Climate vulnerability can be defined as “the degree to which a system is susceptible to and unable to cope with adverse effects of climate change including climate variability and extreme events while climate risk is the inclusive interplay of exposure, vulnerability and adaptive capacity” as endorsed by Bera et al. [1]. Climate risk is basically the outcome of environmental issues. On the other hand, climate vulnerability measures the consequences of environmental events on societies. Climate vulnerability has received increased attention due to its adverse effects on economies, societies, and ecosystems. If not addressed, it may lead to significant economic disruptions, food insecurity, and environmental degradation. Collier and Tirole [2] endorsed the notion that, in the absence of immediate collective efforts, climate change is expected to worsen the well-being of coming generations enormously. Moreover, Haas et al. [3] focused on the uncertainty and the transition towards a low-carbon economy and endorsed the notion that huge investment is needed to achieve the target of zero emissions.

The Gulf Cooperation Council (GCC) is no exception from other economies where substantial economic advancement came at the cost of higher carbon emissions levels and a more fragile climate standing. The threat of climate vulnerability is rising in the GCC region due to the increased CO₂ emissions stemming from the increased use of carbon-based energy. During the last couple of decades, energy use has increased in all GCC economies as pointed out by Mahmood [4]. The GCC economies in general are facing a multitude of environmental challenges including rising CO₂ emissions, pollution in coastal and marine areas, and biodiversity loss. In the GCC economies, some new environmental issues have emerged recently. These problems include demolition debris, construction, and deforestation, as endorsed by Ansari et al. [5]. Al-Maamary et al. [6] documented that the consequences of climate change have appeared in the GCC region in several forms including the rise in temperature with a severe decrease in precipitation. All these developments in the GCC region are the major drivers of increased climate vulnerability.

Climate vulnerability is a dynamic, complicated, and multi-faceted phenomenon. Hence, climate vulnerability may be influenced by several factors. Cardona et al. [7] highlighted the same point and endorsed the notion that vulnerability and exposure respond to several factors including social, economic, institutional, environmental, and cultural factors. However, the institutional determinant of climate vulnerability has received relatively less attention as far as the empirical literature is concerned. Moreover, Dilling et al. [8] endorsed the notion that the quality of domestic institutions ultimately determines resource distribution in societies. It is a fact that the distribution of resources can either improve or worsen climate vulnerability levels. Improvement in institutional quality provides an effective platform for all key stakeholders to play their optimum role in uplifting the quality of the environment by taking visible collective steps.

The GCC economies are experiencing relatively stable political, and economic systems and are taking serious steps to advance institutional quality. These economies have demonstrated a sincere commitment to environmental stewardship by implementing solid long-term plans toward achieving lower carbon emissions. Saudi Arabia introduced the “National Program for the Circular Carbon Economy” in 2020 composed of four strategies: “reduction, reuse, rotation, and removal”, which represents a viable and economically sustainable way to control emissions. The Saudi Green Initiatives set an ambitious multidimensional plan that aims to reach net zero carbon emissions by 2060, exhibiting a strong commitment toward the global environment. Other GCC countries have set their national climate targets. For instance, Bahrain has established a “Joint National Committee on Climate Change”, Kuwait has developed a “National Adaptation Plan”, Oman has implemented a “National Strategy for Climate Change Adaptation and Mitigation”, Qatar has outlined its goals in the “Qatar National Vision”, and the UAE has formulated its “National Climate Change Plan”. The purpose of these initiatives is to address the challenges posed by climate change through comprehensive planning and action. However, achieving environmental goals extends beyond relying solely on shifting towards green energy and renewable resources; it also necessitates a thorough evaluation of climate governance, which heavily depends on a country’s institutional quality.

The aspect of institutional quality plays a pivotal role in measuring a country’s advancement and commitment toward addressing climate challenges. According to North [9], “Institutions are the rules of the game in society or, more formally, are the humanly devised constraints that shape human interaction”. Institutional quality measures governance effectiveness, regulatory frameworks, and political stability in a specific country. Strong institutions can enhance adaptive capacity and promote sustainable resource management. In the context of the GCC countries, where economic activities are closely tied to environmental factors, understanding the impact of institutional quality on climate vulnerability is essential. This study aims to provide empirical evidence on how robust institutional frameworks can help these economies navigate the complex interplay between economic development and climate resilience.

There are logical arguments underpinning the relationship between institutional quality and climate vulnerability. The role of institutional quality is important as it accounts for certain important challenges and opportunities for local authorities in terms of building adaptive capacity for managing climate change as demonstrated by Glass et al. [10]. Improved institutional quality protects the quality of the environment by formulating and implementing certain environment-specific rules and regulations. For instance, imposing carbon taxes or ensuring the use of environmentally friendly technologies through a strict monitoring system will improve the quality of the environment and hence the likelihood that climate vulnerability could be reduced. The resource punishment hypothesis predicts that legislators tend to reduce carbon emissions during election campaigns as voters may hold them responsible for environmental degradation, as shown by research by Stef and Jabeur [11]. To put it differently, an improved political system, which is also one of the main aspects of institutional quality, could significantly improve the quality of the environment. Both a formal and an informal institutional setup can shape the three distinct components of vulnerability including exposure, sensitivity, and adaptive capacity, as endorsed by Lebel et al. [12]. Finally, using data from 23 emerging economies from 1996 to 2014, Omri and Hadj [13] empirically demonstrated that improved governance quality and technological innovation reduces carbon emissions.

Earlier work on climate vulnerability highlighted the influence of excessive carbon emissions on human health, natural resources, and temperature fluctuations, as well as on various socio-economic aspects and business activities. Omri and Kahia [14] demonstrated that resources have a positive impact on the standard of living, and institutional quality plays a crucial role in augmenting this positive effect. However, empirical research evidence is not very extensive on the true and explicit impact of institutional quality on climate vulnerability and this area remains largely untapped. In a recent paper, Birkmann et al. [15] correctly pointed out that less attention has been paid to the assessment of vulnerability and embedded social, economic, and historical conditions that foster the vulnerability of societies. Hassan et al. [16] also pointed out that the role of institutions from the perspective of the environment is rarely considered in the literature. In other words, the previous literature is silent on the factors that cause vulnerability. This means that the institution–climate relationship is largely unexplored despite its importance. This research paper seeks to contribute to bridging this gap in the literature by examining the impact of institutional quality on climate vulnerability, a recent and intriguing yet relatively underexplored area in empirical research.

Despite these challenges, limited empirical research has examined the role of institutional quality in shaping climate vulnerability, particularly in the context of the GCC economies. This study seeks to bridge this gap by establishing a relationship between institutional quality and climate vulnerability. This research paper contributes to the existing literature in several ways. Firstly, the current study attempts to assess the impact of institutional quality on climate vulnerability by focusing on the GCC economies. The existing literature has not paid sufficient attention to figuring out the responsiveness of climate change to enhanced institutional quality. Secondly, instead of using CO₂ emissions as an indicator of climate vulnerability, we are using a comprehensive index of climate vulnerability which is based on three main aspects such as exposure, sensitivity, and adaptive capacity. Therefore, the use of a comprehensive index instead of CO₂ emissions is expected to offer more in-depth insights. Thirdly, the current study is also interested in finding out the direction of the relationship between institutional quality and climate vulnerability by running advanced causality testing. In other words, our study is interested in determining whether better institutional quality shapes climate vulnerability or whether climate vulnerability alters institutional quality. The available literature, to the best of the knowledge of the author, is silent on the issue of causality between institutional quality and climate vulnerability. The current study is therefore differentiated from the previous literature in terms of the sample, methods, and variable measurement, particularly regarding climate vulnerability. We anticipate that the findings of the current study will prove to be highly valuable for the GCC member economies. Consequently, policymakers could use the findings of the current study while formulating policies related to climate vulnerability and institutional improvement.

This research article is divided into seven interconnected sections. Commentaries on the available relevant literature and important relevant statistics are articulated in section two to provide a comprehensive picture of the problem to potential readers. The third section includes the specification of the model while the fourth section includes results. The penultimate section includes discussion on the regression and causality results. The final section includes brief remarks about the findings, implications, and limitations.

2. Literature Review

2.1. Empirical Literature

Owing to the adverse consequences of climate vulnerability, researchers have attempted to figure out its true determinants. Several factors could explain the severity of climate vulnerability. Cardona et al. [7] highlighted the same point and endorsed the notion that vulnerability and exposure respond to several factors including social, economic, institutional, environmental, and cultural factors. Trade openness, urbanization, and the income level of the population have been consistently blamed for the worsened environmental quality and climate change over the years as mentioned by researchers such as Burki and Tahir [17], Tahir et al. [18], and Tahir et al. [19].

The institutional aspects of a country could also be an important determinant of climate vulnerability. The institutional theory provides a solid framework by which a relationship between institutional factors and climate vulnerability can be studied. Bartlett et al. [20] endorsed that institutional theory is relevant in explaining the nexus between institutions and environment. Institutional theory explains how certain changes in rules and regulations impact the ultimate decision-making process towards sustainable development, as in Godefroit-Winkel [21]. This implies that strong and effective institutional factors could improve the overall quality of the environment and will reduce climate vulnerability. Thus, it seems logical to think of there being a significant relationship between the quality of institutions and climate vulnerability. However, the institution–environment relationship is basically an empirical research question due to the complex nature of both environmental and institutional factors. Therefore, the institution–environment relationship is mostly studied by researchers empirically.

On the empirical front, there are some research studies on the relationship between institutional factors and climate vulnerability. In other words, the relationship between climate vulnerability has become an empirical phenomenon instead of a theoretical phenomenon. For instance, Brooks et al. [22] conducted a comprehensive analysis and showed that the quality of institutions is predominantly connected with adaptive capacity. They further endorsed the notion that adaptive capacity is dependent on several aspects of institutional quality including civil and political rights, governance, and literacy. Lebel et al. [12] also indicated the important role of institutional quality in mitigating emissions and reducing the risk of climate vulnerability. They utilized data for 23 emerging economies and demonstrated that technological innovation and an improved quality of governance significantly reduce carbon emissions.

Stef and Jabeur [11] highlighted the political aspect of institutional quality and demonstrated that environmental quality improves during election time. Similarly, by utilizing data on the Indian economy and employing the general liner modeling, Saikia and Mahanta [23] showed that the quality of institutions has played an enormous role in reducing climate vulnerability. Omri and Mabrouk [24] focused on economies belonging to the MENA region and showed that institutional quality is positively linked with the three main pillars of the sustainable development goals. Javaid et al. [25] focused on 114 economies and empirically demonstrated that improved institutional quality reduces CO₂ emissions and hence the risk of climate change and climate variability decrease. Moreover, in the context of the BRIC member economies, Hussain and Dogan [26] highlighted the important role of institutional quality and demonstrated that improvement in institutional quality protects the quality of the environment by reducing CO₂ emissions. This empirical evidence reflects that the problem of climate vulnerability could be tackled effectively by focusing on improving the quality of domestic institutions.

Amid the recent development in institutional quality in the GCC economies, some researchers have carried out empirical studies to assess the responsiveness of climate vulnerability to advanced environmental quality. For instance, Awan [27] focused on the GCC economies to assess the influence of institutional quality on environmental quality. His results show that democratic accountability, corruption, and ethnic conflicts increase CO₂ emissions in the case of the GCC economies. The study further demonstrated that government stability improves environmental quality owing to its adverse impact on CO₂ emissions. On the other hand, Farooq et al. [28] displayed that FDI, growth, and domestic investment are mainly responsible for worsened environmental quality in the GCC economies. Zmami and Ben-Salha [29] also blamed energy use and FDI for environmental degradation in the GCC economies. Kasem and Alawin [30] have also focused on the GCC economies and reported that renewable energy use has a negative impact, improving environmental quality, and electricity use degrades the quality of the environment. The GCC region is indeed very vulnerable to several ecosystem- and climate-related issues as documented by Al Ansari [31]. The GCC economies need to take the problem of climate vulnerability seriously due to its multiple adverse consequences. Satrovic et al. [32] suggested, based on their findings, that the GCC region should prioritize environmental protection and the SDGs as it would help the region on the path to climate change mitigation. Moreover, it should be noted that all these studies have focused on CO₂ emissions and have ignored the concept of climate vulnerability which is more relevant in capturing the overall environmental assessment.

It could be concluded that the relationship between institutional quality and climate vulnerability is largely unexplored. However, looking at the adverse effects of climate vulnerability and the importance of institutional quality, more empirical research studies are needed to uncover the true relationship between institutional quality and climate vulnerability. Specifically, in the context of the GCC economies, researchers have not paid the required amount of attention to seeing how institutional quality impacts climate vulnerability. Meanwhile, it is also a fact that the GCC region is faced with severe environmental threats including climate vulnerability. Furthermore, institutional quality has also been improved in recent years in the GCC region. Therefore, to fill this gap, we in this paper carry out a comprehensive empirical analysis to assess the influence of institutional quality on climate vulnerability in the context of the GCC economies. We expect that the outcome of our study will draw the attention of policymakers and government authorities in the GCC economies.

2.2. Highlights on Climate Vulnerability and Institutional Quality Indices

This section presents an overview of climate vulnerability and institutional quality indices and discusses the historical behavior of these two areas in the context of the GCC countries. The climate vulnerability index was developed by “Notre Dame Research” [33] (NDR, hereafter), and it consists of three main factors including “exposure, sensitivity and adaptive capacity”. Climate exposure, according to NDR [33], refers to “the degree to which a system is exposed to significant climate change from a biophysical perspective”. Similarly, climate sensitivity could be defined as the “Extent to which a country is dependent upon a sector negatively affected by climate hazard, or the proportion of the population particularly susceptible to a climate change hazard”. On the other hand, adaptive capacity is defined as “the availability of social resources for sector-specific adaptation”. In some cases, these capacities reflect sustainable adaptation solutions. The climate vulnerability index ranges from 0 to 1, where higher values reflect a higher climate vulnerability and vice versa.

The institutional quality index developed by World Governance Indicators (WGIs) is composed of six different dimensions to measure the quality of governance within a country including political stability, government effectiveness, the regularity of law, and the control of corruption. The Institutional Quality Index ranges from (−2.5–2.5), where higher values represent better institutional quality. We have used the average value of the different aspects of institutional quality developed by World Governance Indicators (WGIs) to institutional quality and estimated its effect on climate vulnerability.

Table 1. Key Statistics of Climate Vulnerability and Institutional Quality Indices for GCC countries.

Variables	Description	2002	2006	2011	2016	2021	% Change (2002–2021)
Climate Vulnerability	“Climate Vulnerability Index (0–1) Where higher values reflection higher climate vulnerability”	0.409	0.402	0.395	0.392	0.370	−9.530%
Institutional Quality	“Institutional Quality Index (−2.5–2.5) Where higher values represent better institutional quality”	0.508	0.388	0.353	0.366	0.983	93.503%

Note: Authors’ calculation using data from “World Governance Indicators” and “Notre Dame Research”.

presents an overview of climate vulnerability and institutional quality for the GCC countries between 2002 and 2021. The values reported in Table 1 show that climate vulnerability has decreased by more than nine percent during the study period. On the other hand, during the study period, institutional quality increased over the years significantly for the GCC region. The statistics indicate that overall institutional quality has increased by approximately 93 percent. The observed decline in climate vulnerability and improvement in institutional quality has motivated us to think of a potential relationship between institutional quality and climate vulnerability.

3. Methodology

3.1. Theoretical Underpinning and Model Design

To rigorously examine the relationship between institutional quality and climate vulnerability, we have developed a model framework that considers both primary variables and relevant control variables to capture the multifaceted nature of climate vulnerability. Institutional quality and climate vulnerability are the main independent and dependent variables, respectively. The relationship between institutions and climate vulnerability is based on institutional theory. Institutional theory is relevant in explaining the relationship between institutions and the environment, as mentioned by Tina et al. [34]. Institutional theory provides a solid framework by which a relationship between institutional factors and climate vulnerability can be studied. For instance, Bartlett et al. [20] documented that institutional theory is relevant to explaining the institution–environment relationship. Institutional theory explains how certain changes in rules and regulations impact the ultimate decision-making process towards sustainable development as discussed by Godefroit-Winkel [21]. It implies that strong and effective institutional factors could improve the overall quality of the environment and will reduce climate vulnerability. However, keeping in mind the complex nature of climate vulnerability, we have also included some control variables into the specified model. These include trade openness, urbanization, income level, and energy use. The previous literature has provided evidence that such control variables have a significant impact and could explain observed variabilities in climate vulnerability as pointed out by Burki and Tahir [17], Tahir et al. [18], and Afridi et al. [35]. Afridi et al. [35] provided sound theoretical inputs about the relationship between income, trade, urbanization, energy use, and climate vulnerability. Thus, the following function containing all considered variables could be expressed in the following way:

CLV = F (TOPEN, URB, YPC, ENG, INSTQ)

(1)

Expression 1 indicates that climate vulnerability (CLV) is dependent upon the degree of openness ($\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}$) to trade, urbanization ($\mathrm{U}\mathrm{R}\mathrm{B}$), income level ($\mathrm{Y}\mathrm{P}\mathrm{C}$), energy use ($\mathrm{E}\mathrm{N}\mathrm{G}$), and the quality of domestic institutions ($\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}$). Considering the non-linearities among the variables, we have transformed expression using logarithmic transformation, as shown below.

$${\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\beta }}_2{\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\beta }}_3{\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\beta }}_4{\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\beta }}_5{\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}+{\mathrm{U}}_{\mathrm{i}\mathrm{t}}$$

(2)

To estimate the impact of institutional quality on climate vulnerability, this work utilizes two indices, namely ND-GAIN and WGI, which represent climate vulnerability and institutional quality, respectively. Trade openness, urbanization, income level, and energy use are considered control variables in this setting, used to investigate the research question.

3.2. Reasons for Variable Selection

Variable selection in research studies should be based on the theoretical literature or the empirical literature. The same procedure is followed by the current study for the selection of variables. For the relationship between trade openness and climate vulnerability, we followed the theoretical framework of Grossman and Krueger [36], who believe that trade openness impacts environmental vulnerability mainly through the channels of technique, composition, and scale. Several empirical studies, such as Afridi et al. [35] and Tahir et al. [18], have repeatedly assessed the impact of trade openness on the environment. Similarly, for the income–vulnerability relationship, we have followed the traditional “Environmental Kuznets Curve (EKC)” where the income level initially degrades the environment. The income–environment relationship is tested in several empirical research studies such as Tahir et al. [18] and Tahir et al. [19]. Similarly, regarding the inclusion of urbanization in our research model, the current study relied on the novel research of Fang and Wang [37] who endorsed the notion that urbanization is indeed a threat to the environment. The empirical study of Burki and Tahir [17] indicated that increased urbanization leads to worsened environmental outcomes and increased climate vulnerability. Moreover, the energy–environment relationship has been extensively researched by researchers as energy use is the primary engine of the growth and development process of nations. Studies such as Tahir et al. [18] and Afridi et al. [35] have shown that traditional energy sources directly enhance the likelihood of climate vulnerability due to their positive impact on CO₂ emissions. Therefore, in the light of the prior literature, we have included energy use as one of the explanatory variables. Finally, regarding institutional quality as a determinant of climate vulnerability, several empirical studies are available. For example, Cardona et al. [7], Brooks et al. [22], and Saikia and Mahanta [23] have provided significant evidence regarding the institutional impact on climate outcomes and climate vulnerability. Therefore, based on the predictions of the mentioned studies, we have included institutions as a determinant of climate vulnerability in our model.

4. Estimation Techniques

Panel data require specific econometric techniques for the purpose of estimation. Panel data have many advantages. Among other benefits, panel data provide an increased degree of freedom, more sample variability and further control the consequences of omitted variables, as documented by Hsiao [38]. Due to these benefits, panel data have been utilized extensively. For the estimation of panel data, “Fixed Effects (FET)” and “Random Effects (RET)” are two powerful and effective tools. FET is suitable for models with no correlation between the error component and regressors. However, FET normally suffers from the dummy variable trap due to the presence of time-invariant factors in the model. On the other hand, RET is assumed to work well in terms of taking care of time-invariant factors among independent variables. However, RET is unable to address problem correlation. Borenstein et al. [39] documented that both FET and RET may sometimes produce similar estimates for parameters. However, both FET and RET are fundamentally different from each other having different assumptions.

In the literature, the Hausman test [40] of specification has been utilized by researchers for choosing between FET and RET. The results displayed in

Table A1. “Hausman Test”.

“Test”	“Chi. Sq. Value”	“Chi. Sq. D. F”	“Probability”	“Decision”
“Cross-section random”	769.558	5	0.000	FET

(Appendix A) show that FET is suitable. Therefore, we have estimated our models using FET modeling. Similarly, we have also examined the possibility of “Cross-Sectional Dependency (CD)” by using the “Pesaran CD test”. The results presented in

Table A2. “Residual Cross-Section Dependence Test”.

“Test”	“Statistic”	“D.F”	“Probability”	“Conclusion”
“Pesaran CD Test”	−0.4293	10	0.6677	“Cross-sections are independent”

confirm that the sampled countries are independent of each other, which is desirable. Moreover, the study also carried out the Chow Test to decide between the “Pooled Least Squares (PLS)” and FET. The results presented in

Table A3. “Chow Test”.

“Effects Test”	“Statistic”	“d.f.”	“Prob.”
“Cross-section F”	230.345	(5, 43)	0.000
“Cross-section Chi-square”	219.415	5	0.000
“Period F”	6.023	(12, 43)	0.000
“Period Chi-square”	65.085	12	0.000
“Cross-Section/Period F”	99.136	(17, 43)	0.000
“Cross-Section/Period Chi-square”	243.784	17	0.000

confirmed the superiority of FET.

Before moving onto FET estimation, we also utilized the PLS estimator. The purpose behind using the PLS is that it provides firsthand information as endorsed by Chen and Gupta [41] and Tahir and Alam [42]. For robustness and sensitivity, the present study utilized “Feasible Generalized Least Squares (FGLS)”, which is a standard procedure as documented by Tahir and Khan [43]. Lastly, to address the endogeneity problem, the present study employed the “Two Stages Lest Squares (2SLS)” estimator. In the 2SLS estimation, we used the lagged values of variables as instruments, which is supported by Tahir and Alam [42]. We skipped GMM as the number of cross-sections is less than the number of years (N < T). Therefore, we followed the previous literature, such as Tahir and Alam (2023), and utilized the 2SLS estimator to address the potential endogeneity issue.

5. Results

5.1. Data and Descriptive Statistics

This paper utilizes the panel data of the GCC countries to assess the response of climate vulnerability to enhanced institutions. Data were collected for all six GCC member economies. Our sample data cover the period 2002–2021. Data on the climate vulnerability index were taken from the “Notre Dame Global Adaptation Initiative”. Similarly, data on institutional quality were obtained from the WGI. Moreover, data on all control variables, such as trade openness, urbanization, income level, and energy use, were sourced from “World Development Indicators (WDI)”. “Total trade as a percentage of GDP” was used to measure the “degree of trade openness” while “urban population as a percentage of total population” was used to approximate the urbanization level. Energy use was measured by taking “energy use (kg of oil equivalent per capita)”. Details about the data and variables are shown in the Appendix A in

Table A4. “Variables and Data Sources”.

“Variable”	“Description”	“Source”
${\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$	“Climate Vulnerability Index (0–1) Where higher values reflect higher climate vulnerability and vice versa”	“Notre Dame Global Adaptation Initiative”
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	“Trade openness (trade as % of GDP)”	“World Development Indicators”
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	“Urban population as % of total population”	“World Development Indicators”
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	“Per capita GDP (Constant US $)”	“World Development Indicators”
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	“Energy Use (kg of oil equivalent per capita)”	“World Development Indicators”
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	“Institutional Quality Index (−2.5–2.5) Where higher values represent better institutional quality and vice versa”	“World Governance Indicators”

.

Descriptive statistics are presented in

Table 2. Descriptive Statistics.

	${\mathbf{C}\mathbf{L}\mathbf{V}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{I}\mathbf{N}\mathbf{S}\mathbf{T}\mathbf{Q}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{T}\mathbf{O}\mathbf{P}\mathbf{E}\mathbf{N}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{U}\mathbf{R}\mathbf{B}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{N}\mathbf{E}\mathbf{N}\mathbf{G}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{Y}\mathbf{P}\mathbf{C}}_{\mathbf{i}\mathbf{t}}$
“Mean”	0.397	0.429	109.378	87.691	9997.415	34,399.310
“Maximum”	0.455	1.070	191.872	100.000	21,420.630	73,493.270
“Minimum”	0.353	−0.260	64.592	71.509	3358.637	15,561.480
“Std. Dev”.	0.030	0.339	28.578	9.134	4527.734	17,414.040

Note: Authors’ own calculations using data from WDI, WGI, and the Gain Index.

. Climate vulnerability, which is our main variable of interest, takes an average value of 0.397 with a standard deviation of 0.030. The highest value of climate vulnerability is observed for Bahrain for the year 2002. The minimum value is seen for Kuwait in 2018. In general, these values are quite high and hence all economies belonging to the GCC region face severe climate vulnerability. Similarly, the average value of institutional quality is 0.429, which is moderate as this institutional quality index ranges from “−2.5 to +2.5”, where higher values are the reflection of better quality institutions. The highest value of institutional quality 1.070 is observed for Qatar for the year 2009, while the lowest value of −0.26 is recorded for Saudi Arabia for the year 2004. The overall performance of the GCC economies is reasonable.

The statistics of trade openness show that GCC economies are remarkable. The “trade to GDP ratio”, which is the main indicator of openness, is 109.378 percent on average for the GCC economies. The highest value is recorded for Bahrain for the year 2013, while lowest value is observed for the Saudi Arabia in 2002. Similarly, the urbanization statistics show all of the economies are highly globalized. The economy of Kuwait is completely globalized, while the lowest value of urbanization is recorded for Oman for the year 2003.

In terms of energy use, the average value is 9997.415 “(kg of oil equivalent per capita)”, which is reasonably high. The maximum and minimum values of energy use were recorded for Qatar and Oman for the years 2004 and 2002, respectively. Finally, in terms of income level, the statistics show that the GCC economies are rich and hence the populations of these economies enjoy relatively high levels of income. The maximum value of the income level is recorded for Qatar for the year 2011. The minimum income level is seen for Saudi Arabia for the year 2002.

5.2. Preliminary Tests

We conduct bivariate analysis by creating the correlation matrix among the examined variables shown in

Table A5. “Correlation Matrix”.

	${\mathbf{C}\mathbf{L}\mathbf{V}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{L}\mathbf{N}\mathbf{I}\mathbf{N}\mathbf{S}\mathbf{T}\mathbf{Q}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{L}\mathbf{N}\mathbf{T}\mathbf{O}\mathbf{P}\mathbf{E}\mathbf{N}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{L}\mathbf{N}\mathbf{U}\mathbf{R}\mathbf{B}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{L}\mathbf{N}\mathbf{N}\mathbf{E}\mathbf{N}\mathbf{G}}_{\mathbf{i}\mathbf{t}}$	${\mathbf{L}\mathbf{N}\mathbf{Y}\mathbf{P}\mathbf{C}}_{\mathbf{i}\mathbf{t}}$
${\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$	1	0.358	−0.309	0.657	0.460	0.687
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	0.358	1	0.235	0.108	0.445	0.694
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	−0.309	0.235	1	−0.091	0.0005	−0.085
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	0.657	0.108	−0.091	1	0.759	0.552
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	0.460	0.445	0.0005	0.759	1	0.790
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	0.687	0.694	−0.085	0.552	0.790	1

 Appendix A. The correlation results show that variables are correlated in a moderate manner. The highest correlation of 0.79 is observed between income level and energy use. The lowest correlation is recorded for energy openness. All other variables have a modest correlation with each other. Looking into the nature and strength of the correlations among the variables, we also carried out the “Variance Inflation Factor (VIF)” test to explore the possibility of multicollinearity. The VIF test displayed in

Table A6. “VIF Test Results”.

“Variables”	“VIF”	“Conclusion”
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	1.982536	“No Multicollinearity”
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	4.243955
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	3.206247
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	3.096526
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	2.603754

 Appendix A rejected the possibility of multicollinearity as all values are below five. The possible reasons, among others, for the absence of multicollinearity despite the high correlation between energy and income is that multicollinearity happens when multiple regressors are strongly correlated.

Moreover, we also conducted unit root testing. The results of the unit root are shown in Appendix A (

Table A7. Unit Root Analysis.

“Variables”	“Level”	“F.D”	“Decision”
${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$	−4.116 ***	−4.894 ***	I(0)
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	−2.085 **	−8.934 ***	I(0)
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	−3.098 ***	−5.740 ***	I(0)
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	−2.803 **	−5.627 ***	I(0)
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	−2.309 **	−11.433 ***	I(0)
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	−2.214 **	−3.897 ***	I(0)

Note: The asterisks (***) and (**) show 1 and 5 percent levels of significance.

). The results rejected the presence of unit root and hence the possibility of spurious regression was not accepted. Therefore, the stationarity of the variables is accepted.

5.3. Main Regression Results and Causality Results

The results obtained using the PLS are presented in the penultimate column of

Table 3. Regression Results.

Variables	PLS	FET
	Coefficient	Coefficient
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	−0.024 *** (0.006)	0.007 * (0.004)
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	0.211 *** (0.026)	0.230 *** (0.028)
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	0.060 *** (0.005)	0.008 ** (0.004)
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	−0.061 *** (0.007)	−0.004 (0.003)
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	−0.002 (0.002)	−0.002 *** (0.0009)
Constant (${\mathbf{B}}_{\mathbf{0}}$)	−0.291 (0.091)	−0.515 (0.126)
Diagnostics Checking	“R-Squared”: 0.860 “Adjusted R-Squared”: 0.848 “S.E. Regression”: 0.011 “F-Test”: 73.316 ***	“R-Squared”: 0.988 “Adjusted R-Squared”: 0.965 “S.E. Regression”: 0.003 “F-Test”: 583.512 ***

Note: The dependent variable is climate vulnerability. The asterisks (***), (**), and (*) indicate the significance level at the 1, 5, and 10 percent levels. In the parenthesis, we have reported standard errors.

, while the final column includes results extracted using FET. The PLS-based results show that energy use and trade openness are negatively associated with climate vulnerability. On the other hand, income levels cast a positive influence on climate vulnerability. However, we do not place much emphasis on the PLS-based results as the Chow test indicates the use of FET.

Similarly, the results based on FET show that institutional quality is negatively and significantly related to climate vulnerability. On the other hand, the coefficient of trade openness is positive and different from zero in terms of statistical significance. Moreover, urbanization and income level have also increased climate vulnerability in the GCC economies as their coefficients are positive and significant. Finally, energy use is unable to explain climate vulnerability in the GCC economies as its coefficient is insignificant statistically.

The findings of the robust analysis are shown in

Table 4. Robustness Results.

Variables	FGLS	2SLS
	Coefficients	Coefficients
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	0.006 * (0.003)	0.011 ** (0.004)
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	0.229 *** (0.022)	0.221 *** (0.025)
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	0.005 * (0.003)	0.009 ** (0.004)
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	−0.002 (0.003)	−0.010 *** (0.003)
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	−0.002 *** (0.0004)	−0.002 ** (0.0007)
Constant (${\mathbf{B}}_{\mathbf{0}}$)	−0.485 (0.098)	−0.451 (0.131)
Diagnostics Checking	“R-Squared”: 0.988 “Adjusted R-Squared”: 0.962 “S.E. Regression”: 0.003 “F-Test”: 580.136 ***	“R-Squared”: 0.918 “Adjusted R-Squared”: “S.E. Regression”: 0.002 “F-Test”: 866.269 ***

Note: The dependent variable is climate vulnerability. The asterisks (***), (**), and (*) indicate the significance level at the 1, 5, and 10 percent levels. In the parenthesis, we have reported standard errors.

. Two estimators, namely, the FGLS and TSLS are used. Both the estimators showed that institutional quality is negatively associated with climate vulnerability. Similarly, the results further demonstrated that climate vulnerability responds positively to increased income level, urbanization, and trade openness.

Finally, the findings of causality analysis conducted to examine the exact direction of the relationship among the variables are shown in

Table 5. Causality Testing.

Hypothesis (Null)	F-Statistic	Prob.
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$	3.215 *	0.076
${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$$\to$  ${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	0.462	0.498
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$	0.123	0.725
${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	0.038	0.845
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$	2.156	0.144
${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	10.478 ***	0.001
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$	1.147	0.286
${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	0.201	0.654
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$	1.843	0.178
${\mathrm{L}\mathrm{N}\mathrm{C}\mathrm{L}\mathrm{V}}_{\mathrm{i}\mathrm{t}}$  $\to {\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	0.0001	0.992
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	0.895	0.346
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$ $\to$ ${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	0.001	0.974
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	0.327	0.568
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}} \to$  ${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	0.028	0.865
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	1.963	0.164
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	0.575	0.450
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$	0.077	0.781
${\mathrm{L}\mathrm{N}\mathrm{I}\mathrm{N}\mathrm{S}\mathrm{T}\mathrm{Q}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	1.044	0.311
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	0.027	0.867
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	1.520	0.220
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	0.758	0.385
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	1.625	0.205
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}}$	0.0003	0.986
${\mathrm{L}\mathrm{N}\mathrm{T}\mathrm{O}\mathrm{P}\mathrm{E}\mathrm{N}}_{\mathrm{i}\mathrm{t}} \to$  ${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	0.814	0.370
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	0.647	0.422
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	0.487	0.486
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$	6.515 **	0.012
${\mathrm{L}\mathrm{N}\mathrm{U}\mathrm{R}\mathrm{B}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	0.395	0.531
${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$	2.498	0.118
${\mathrm{L}\mathrm{N}\mathrm{Y}\mathrm{P}\mathrm{C}}_{\mathrm{i}\mathrm{t}}$  $\to$  ${\mathrm{L}\mathrm{N}\mathrm{N}\mathrm{E}\mathrm{N}\mathrm{G}}_{\mathrm{i}\mathrm{t}}$	0.163	0.687

Note: The asterisks (***), (**), (*) show 1, 5, and 10 percent levels of significance.

. The causality analysis uncovered several causal relationships among the variables. Most importantly, the results showed that institutional quality is causing climate vulnerability unilaterally. Similarly, climate vulnerability is unilaterally connected with urbanization. Finally, a one-way causal relationship is witnessed between the use of energy and urbanization in the case of the GCC economies.

6. Discussion on Regression and Causality Findings

The results extracted using FET modeling showed that institutional quality is one of the dominant factors due to which the problem of climate vulnerability could be addressed. It implies that improved institutional quality is bound to address the problem of climate vulnerability as far as the economies of GCC are concerned. Saikia and Mahanta [23] endorsed the notion that institutional quality is indeed important for improving society, and specifically environmental problems. Further, improvement in institutional quality improves the adaptive capacity due to which the problem of climate vulnerability can be tackled. Based on global assessment, Javaid et al. [25] endorsed that climate risk can be significantly reduced by improving the quality of domestic institutions. Likewise, the climate vulnerability issue could be addressed effectively by focusing on bringing significant improvement to the quality of domestic institutions. Climate risk and vulnerability are extremely undesirable in the current globalized world as they are responsible for the destruction of ecosystems and human social systems, as pointed out by Song et al. [44]. Therefore, it is suggested that the GCC economies must take some visible steps to improve their institutional quality.

The results also revealed that trade openness is responsible for the severe nature of climate vulnerability. Industrial growth, which is a prerequisite of increased exports, also poses several threats to environmental quality in the form of increased emissions into the atmosphere, which further increases the chances of climate vulnerability. Barkat et al. [45] demonstrated a positive direct relationship between trade openness and environmental degradation. Further, they showed that indirectly, through the mediating effect of income, trade openness improves environmental quality. Dou et al. [46] provided evidence about an inverted U-shaped relationship which indicates that openness initially worsens the environment but ultimately improves it in the long run. Likewise, the GCC economies have not reached a turning point where trade openness could reduce the risk of climate vulnerability due to their small share in global trade. However, it is important to mention that exports are the key factor for securing long-term economic growth. Therefore, the production process should be based on cleaner and environmentally friendly technologies to boost the export sector without damaging the quality of the environment, which ultimately increases the risk of climate vulnerability.

Income level, which is one of the main factors behind all economic activities, also poses a major threat to a clean environment. In other words, the increased income level of the population also degrades the environmental quality due to irresponsible behavior due to which the severity of climate vulnerability increases enormously. Burki and Tahir [17] also showed that a rising income level degrades environmental quality. The framework of Grossman and Krueger [36] also supports the notion that an increased income level degrades the overall environment initially. Omri and Makhlouf [47] observed that the association between per capita income and environmental degradation exhibits an inverted U-shaped pattern. They observed that per capita GDP positively influences CO₂ emissions, whereas the squared GDP term negatively affects emissions in developed economies, thereby endorsing the conventional EKC hypothesis. Therefore, awareness among the population about the adverse consequences of climate vulnerability needs to be enhanced. Awareness among the population is the key to addressing the problem of climate vulnerability.

Moreover, it is found that urbanization is also harmful for environmental quality. Increased urbanization, according to the results, will destroy environmental quality and hence the problem of climate vulnerability will increase. The point estimate suggests that the impact of urbanization on climate vulnerability is highest compared to the other determinants included in the model. This implies that urbanization is the main threat to improving environmental quality as it is the primary cause of climate vulnerability. However, it is also a fact that increased urbanization is not only associated directly with deforestation processes but also increases CO₂ emissions in the atmosphere due to higher energy use. Therefore, the process of urbanization needs to be seriously and effectively monitored to reduce its adverse consequences.

Energy use entered into the estimated model negatively, which is an indication of its favorable impact on environmental quality. However, it is noteworthy that its coefficient is not significant. Burki and Tahir [17] also showed that increased energy use degrades environmental quality due to its positive impact on CO₂ emissions. Kahia et al. [48] also highlighted that the current renewable energy resources in Saudi Arabia are insufficient to meet the increasing demands for energy driven by economic growth, a gap that has an adverse effect on the environment. Thus, carbon emissions due to energy use are expected to remain a significant driver of climate vulnerability before achieving the net zero emissions targets in the GCC countries.

Finally, the fitted models were used to explain the variation in climate vulnerability. The values of R-squared are very high and they are an indication of the explanatory power of the estimated models. Furthermore, the F-test provided evidence for the fitness of models. Hence, our findings regarding the fitted models are valid and hence could be used by the authorities with confidence to solve problems.

The robust analysis also showed that indeed improved institutional quality is important for curbing the problem of climate vulnerability in the context of the GCC economies. Similarly, the results extracted using the 2SLS also confirmed the important role of improved institutional quality in addressing the problem of climate vulnerability. These results are consistent with our prior results which were obtained using FET modeling. Both in the FGLS and 2SLS estimations, most of the variables maintained their significance level as well as the sign of their coefficients. For instance, trade openness, urbanization, and income level appeared to be mainly responsible for the problem of climate vulnerability in the GCC economies. These results are in line with the results reported earlier. Similarly, the negative influence of energy use on climate vulnerability remained the same both in the FGLS and 2SLS estimations.

Overall, it could be summarized that the additional estimation tools, such as the FGLS and 2SLS estimators, provided significant support to the earlier results extracted using FET modeling. Therefore, we concluded that our results are robust compared to alternative estimation procedures. In other words, our results could be used by the policymakers of GCC economies effectively to addressing the problem of climate vulnerability.

We have also carried out causality testing to identify the exact direction of the relationship among the variables selected for the current study. The results of the causality testing are displayed in Table 5. Several one-way relationships among the variables are confirmed by the causality testing. For example, we found evidence of a one-way causal relationship running from institutional quality towards climate vulnerability. This implies that improved institutional quality is an important factor for addressing the problem of climate vulnerability. Thus, institutional development is indeed needed to curb the climate vulnerability problem in GCC economies. Agrawal et al. [49] rightly documented that both social and institutional factors significantly influence climate vulnerability. Similarly, climate vulnerability is unilaterally connected with urbanization. This means that climate vulnerability puts severe pressure on the urbanization process in economies and hence the quality of life worsens. In other worlds, climate vulnerability is undesirable from the perspective of quality of life. Finally, a one-way causal relationship was witnessed between the use of energy and urbanization in the case of the GCC economies. This indicates that energy use directly causes the urbanization process to flourish. Urbanized areas normally require more energy sources compared to less urbanized areas. Therefore, the urbanization process needs to be monitored and unplanned urbanization should be discouraged. Unplanned urbanization increases the risk of climate vulnerability.

7. Conclusions, Implications, and Limitations

7.1. Concluding Remarks

This paper aimed at investigating the relationship between the quality of institutions and climate vulnerability, which is an unexplored research area in the literature. The paper focused on the economies of the GCC and utilized panel data for the period 2002–2021. The models designed for the purpose of empirical analysis were estimated by employing four econometric tools (PLS, FET, FGLS, and TSLS).

The results indeed indicate that the quality of institutions matters in terms of addressing the problem of climate vulnerability as far as the GCC economies are concerned. The causality test also confirms the directional link running from institutional quality to climate vulnerability in the case of the GCC countries. This implies that institutional development could indeed help the GCC economies to address the problem of climate vulnerability. Thus, a rational policy would be to pay attention to improving institutional quality by taking some visible steps. Climate vulnerability is indeed a global problem owing to its worldwide consequences. This means that the global community needs to take visible steps to strengthen institutional frameworks despite the existing severe challenges including political resistance and government inefficiencies. In particular, GCC policymakers are encouraged to utilize their energies in shaping the quality of their existing institutions. The GCC economies have shown significant commitment to addressing environmental issues on an immediate basis. In such a scenario, they must focus more on the institutional aspects as these are directly linked with improvements in environmental quality. Therefore, a significant improvement in institutional quality would eradicate the problem of climate vulnerability, which is desirable. Similarly, the results provided significant evidence on the positive relationship between income level, openness to trade, urbanization, and climate vulnerability.

7.2. Policy Implications

The first and foremost policy suggestion would be that the authorities and policymakers of the GCC economies should focus on improving the quality of domestic institutions to alleviate climate vulnerability. Improved intuitional quality in light of the results obtained would bring enormous benefits in the form of addressing the problem of climate vulnerability, which is extremely desirable from the perspective of improved quality of life. However, it is a fact that improvements in institutional quality are not straightforward, amid political resistance and government inefficiencies. Also, institutional development is a long-term process and depends on several aspects including increased awareness, a strong will, consistency in policies, and a readiness to change for the better. Moreover, both the government sector and the private sectors need to join hands to work together on improvements in institutional quality. Furthermore, the use of greener and cleaner technologies for the purpose of production instead of traditional fossil fuels-based energy sources would significantly reduce the chances of climate vulnerability in the GCC region. Finally, awareness regarding the deadly consequences of climate vulnerability among the population will improve the overall environmental quality in the GCC economies due to their responsible behavior towards the environment.

7.3. Limitations of the Study and Future Research Directions

Like all research studies, this study has also some notable limitations. Firstly, the designed and estimated models include only a few variables due to the small cross-sectional dimensions of the panel. The current study did not include all of the determinants of climate vulnerability to avoid econometric problems such as the reduced degree of freedom. While climate vulnerability could be impacted by several other factors. Therefore, future research studies should include some more variables, such as natural resources and financial development, in their climate vulnerability models to provide more comprehensive evidence about the true determinants of climate vulnerability. Secondly, due to the small dimensions of the panel, the current study utilized only the traditional estimation techniques for panel data. Future research studies should use advanced panel data techniques including panel cointegration and generalized method of moment (GMM). The current study did not utilize the GMM estimator as the number of cross-sections was less than the number of years, as it is a fact that the GMM estimator requires the number of cross-sections to be greater than the number of years. Thirdly, the analysis carried out in the paper is based on data on the GCC economies. However, the GCC economies are unique in having some specific economic and political structures. Therefore, the results obtained are limited in terms of generalization ability. In this regard, the current study could be extended by considering other types of institutions such as the economic, political, and social. Lastly, future research studies on climate vulnerability should also focus on some other groups or regions, such as OECD, SAARC, and ASEAN, to test the reliability of our models. Future studies along the suggested lines will address the results’ generalization ability problem effectively.