Econometric Analysis of the Impact of Climate Change on the Performance of Egypt’s Fish Foreign Trade

Abstract

This study examines the impact of climate change on the performance of Egypt’s fish foreign trade during the period from 1995 to 2022. The analysis incorporates a set of climate indicators, including average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions, in addition to fish trade indicators represented by exports, imports, total trade volume, trade balance, and export-to-import coverage ratio. The study employs the Autoregressive Distributed Lag (ARDL) model to investigate both the short-run and long-run relationships between climate change variables and fish foreign trade performance in Egypt. Unit root tests confirmed that the variables were integrated at mixed orders I(0) and I(1), supporting the suitability of the ARDL methodology. The findings reveal the existence of a statistically significant long-run equilibrium relationship between climate change indicators and Egyptian fish exports. In particular, nitrous oxide emissions exerted a significant negative effect on fish exports in the long run, while rainfall showed a positive short-run effect. The results also indicate that approximately 57% of short-run disequilibria are corrected annually toward the long-run equilibrium. In contrast, no long-run cointegration relationship was found between climate variables and fish imports, total fish trade volume, or the fish trade balance, indicating that climate impacts on these indicators are mainly short-term in nature. The study concludes that climate change represents an important determinant of Egypt’s fish trade performance through its effects on productivity, environmental quality, and trade competitiveness. The findings highlight the need for integrated adaptation and mitigation policies to strengthen the sustainability and resilience of Egypt’s fisheries sector under changing climatic conditions.

1. Introduction

In recent decades, global attention to climate change has increased markedly, as it is now recognized as one of the most critical challenges facing sustainable development and food security at both regional and international levels. These changes have contributed to profound disruptions in ecological systems through rising temperatures, fluctuations in precipitation patterns, and increasing water salinity as a result of sea level rise. Such shifts directly affect marine and inland ecosystems and, consequently, the productivity and sustainability of fish resources [1]. The impacts of climate change are no longer confined to production alone but have extended to all stages of the value chain, including processing, storage, transportation, and international trade. This expansion has intensified uncertainty in global fish markets and contributed to fluctuations in supply and demand [2,3].

In Egypt, this issue holds particular significance due to the structural characteristics of the fisheries sector, which relies heavily on aquaculture as the primary source of production. Egypt contributes more than 80 percent of total fish production in Africa and ranks among the leading producers globally in this field [4]. However, the sector faces increasing challenges, including declining water quality and rising salinity levels in the northern lakes, the effects of higher temperatures on growth and productivity rates, and the growing incidence of diseases in intensive aquaculture systems [5,6]. Moreover, the heavy dependence on freshwater resources, particularly those linked to the Nile River, heightens the sector’s vulnerability to climate variability, which affects water availability, quality, and distribution [7,8]. As a lower-middle-income country characterized by increasing food demand, limited freshwater resources, and growing dependence on climate-sensitive production systems, Egypt represents a particularly important case for examining the interaction between climate change and fish foreign trade in developing economies.

The implications of climate change extend beyond production to directly influence the performance of Egypt’s fish foreign trade. Export and import volumes, prices, and the costs of production, transportation, and storage are all affected, leading to a restructuring of the fish trade system. This has resulted in increased pressure on the balance of payments due to rising imports needed to cover the domestic food gap, alongside a decline in the competitiveness of some exports due to higher costs or reduced quality. Consequently, instability in the sector’s trade performance has intensified [9,10]. Furthermore, climate-driven shifts in the global distribution of fish resources may lead to a reconfiguration of international markets and changing demand patterns, thereby influencing the market share of Egyptian exports [2,4]. In addition, the water dimension emerges as a critical factor in analyzing this relationship, particularly within the framework of the virtual water concept. The fish trade is closely linked to the efficiency and cost of water use, as well as the degree of reliance on non-domestic water resources. Accordingly, climate change may reshape export and import decisions by affecting water productivity and the costs associated with its use. This underscores the need for integrated policies that connect water resource management with foreign trade strategies [11,12,13].

Despite the growing body of literature addressing climate change and fisheries sustainability, most previous studies have primarily focused on the effects of climate change on fish production, aquaculture systems, food security, and ecosystem dynamics, while limited attention has been directed toward its implications for fish foreign trade performance, particularly in developing countries and the Egyptian context. Moreover, few empirical studies have simultaneously examined the relationship between climate indicators and fish trade indicators within an integrated econometric framework capable of distinguishing between short-run and long-run effects.

Accordingly, the research problem of the current study centers on understanding how climate change affects the performance of Egypt’s fish foreign trade sector and whether long-run equilibrium relationships exist between climate change indicators and fish trade indicators. In this context, the study seeks to answer the following research questions: To what extent do climate change indicators affect Egyptian fish exports and imports? How do climate variables influence total fish trade volume and fish trade balance performance? Do long-run equilibrium relationships exist between climate indicators and fish foreign trade indicators in Egypt?

Despite the growing body of literature addressing climate change and fisheries production, limited empirical studies have examined the direct relationship between climate change indicators and fish foreign trade performance, particularly in developing countries and within the Egyptian context. In light of the above, there is a clear need to analyze the combined effects of climate change on the performance of Egypt’s fish foreign trade. This requires a comprehensive understanding of both direct and indirect transmission channels and an evaluation of their implications for domestic production, trade costs, and competitiveness. Such analysis can support the formulation of effective adaptation policies that enhance the sustainability of the fisheries sector and strengthen its position in international markets under changing climatic conditions.

2. Review of the Literature

Climate change has become one of the most significant environmental challenges affecting fisheries production and trade worldwide. The existing literature indicates that climate change influences fisheries systems through multiple physical, ecological, economic, and institutional pathways. Rising sea surface temperatures, ocean acidification, sea-level rise, altered ocean circulation, and changes in nutrient dynamics have been identified as major factors affecting fish stock productivity, species distribution, and ecosystem stability [14,15].

Several studies have emphasized that increases in sea surface temperature and changing marine conditions directly affect fish growth, recruitment, habitat suitability, and catch composition. Tropical fisheries are considered particularly vulnerable to climate change because of their narrow thermal tolerance ranges and strong dependence on stable environmental conditions [15]. Other studies argued that climate-driven ecosystem disruptions may contribute to fisheries collapse unless ecosystem-based fisheries management approaches are adopted [16]. Similarly, climate warming may reduce maximum sustainable yields and alter optimal fisheries management strategies, although adaptive management approaches may partially mitigate these negative effects [17].

The literature also documented the economic consequences of climate change on fisheries production and profitability. Climate change scenarios were found to significantly reduce fisheries production and landed catches in several coastal regions, with substantial regional disparities in economic impacts and adaptation capacities [18]. In addition, warming and ecological competition were shown to negatively affect fisheries productivity and long-term sustainability, particularly in small-scale coastal fisheries [19]. Other studies highlighted that climate-adaptive fisheries reforms may improve resilience and reduce future economic losses under moderate climate change scenarios [14].

Another important theme in the literature concerns adaptation and governance mechanisms. Previous studies consistently emphasized the importance of climate-aware and ecosystem-based fisheries management systems. Ecosystem-based fisheries management frameworks can reduce the risks of climate-driven fisheries collapse, although their effectiveness depends on the magnitude of warming and governance capacity [16]. Effective adaptation also requires integrated approaches that combine environmental management, economic efficiency, and institutional flexibility [17,19].

The literature has also highlighted the growing role of aquaculture as a complementary strategy to offset declines in capture fisheries production. Aquaculture expansion may contribute to food security and fisheries resilience; however, the sector itself remains vulnerable to climate-related stressors such as temperature increases, disease outbreaks, and water quality deterioration [20]. Likewise, the sustainability of aquaculture depends on technological innovation, environmental regulation, and adaptive management practices [19].

At the regional level, Mediterranean and North African studies have demonstrated that climate change poses significant risks to fisheries systems and coastal economies. Climate change increases interconnected risks related to water resources, ecosystems, and food security across the Mediterranean region [21]. Rising temperatures and salinity changes negatively affect fisheries yields and economic performance in Mediterranean countries [22]. Mediterranean aquaculture systems are also increasingly threatened by warming temperatures, sea-level rise, acidification, and disease risks [23].

In the Egyptian context, several studies addressed the vulnerability of fisheries production systems and coastal zones to climate change. The Nile Delta has been identified as one of the most vulnerable coastal regions to sea-level rise, storm surges, and coastal erosion, threatening fisheries infrastructure and aquaculture activities [24]. Other studies developed inundation models for the northern coastal zone of the Nile Delta and revealed substantial future risks to fish farms and coastal fisheries due to sea-level rise and land subsidence [25].

Climate-related pressures affecting aquaculture systems in Egypt have also received increasing attention. Climate change negatively affects fish farming through rising water temperatures, increasing salinity levels, declining dissolved oxygen concentrations, and higher disease incidence [26]. Furthermore, climate change threatens aquaculture activities and food security in the Nile Delta through inundation risks and environmental degradation [27]. Other studies discussed the implications of climate-related degradation of marine and coastal resources for Egypt’s blue economy and fisheries sustainability [28].

Despite the growing body of literature on climate change and fisheries production, several research gaps remain evident. Most previous studies focused primarily on fisheries production, ecological impacts, aquaculture systems, and adaptation policies, while limited attention has been devoted to the implications of climate change for fish foreign trade indicators such as exports, imports, trade balance, total trade volume, and export competitiveness. In addition, empirical studies examining the relationship between climate change and fish trade in developing countries remain relatively scarce, particularly in the Egyptian context. Furthermore, limited studies have applied econometric approaches capable of distinguishing between short-run and long-run effects of climate variables on fisheries trade performance.

Accordingly, the current study seeks to address these gaps by analyzing the impact of climate change on Egypt’s fish foreign trade using the Autoregressive Distributed Lag (ARDL) model. The study extends the existing literature by integrating climate indicators with fish trade indicators within a unified econometric framework and by examining both the short-run and long-run dynamics between climate change and fish foreign trade performance in Egypt.

2.1. Study Contribution

The current study contributes to the existing literature by extending the analysis of climate change impacts beyond fisheries production to include fish foreign trade performance in Egypt. While previous studies mainly focused on the environmental, ecological, and production-related effects of climate change on fisheries and aquaculture systems, limited empirical attention has been devoted to analyzing its implications for fish exports, imports, trade balance, total trade volume, and export coverage ratio, particularly in the Egyptian context.

The study also contributes by providing an empirical assessment of the relationship between climate change indicators and fish foreign trade indicators in Egypt during the period of 1995–2022. In this regard, the study integrates climate variables, including average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions, with fish trade indicators within a unified analytical framework.

The theoretical foundation of this study is based on the assumption that climate change affects fish foreign trade through multiple direct and indirect transmission channels. Rising temperatures, changes in rainfall patterns, and increasing greenhouse gas emissions influence fish productivity, water quality, production costs, and the sustainability of fisheries resources. These environmental changes subsequently affect export competitiveness, import dependence, trade balance performance, and the overall stability of fish foreign trade. Accordingly, the empirical framework of the current study incorporates climate indicators and fish trade indicators within an integrated ARDL model to evaluate both short-run and long-run relationships.

Methodologically, the study applies the Autoregressive Distributed Lag (ARDL) model to investigate both short-run and long-run relationships between climate change variables and fish foreign trade performance. The use of the ARDL approach allows the estimation of dynamic relationships among variables with different orders of integration and enables the identification of long-run equilibrium relationships between climate indicators and fish trade variables.

In addition, the study contributes to the sustainability and fisheries economics literature by providing evidence on how climate change may influence the performance and competitiveness of the fish trade sector in Egypt. The findings may support policymakers in designing climate-adaptive strategies aimed at improving fisheries sustainability, strengthening export performance, reducing trade vulnerabilities, and enhancing food security under changing climatic conditions.

2.2. Research Significance

The significance of this study lies in its examination of the interaction between climate change and Egypt’s fish foreign trade, a critical nexus where environmental, economic, and strategic dimensions intersect. From an environmental perspective, the fisheries sector is among the most vulnerable to changes in temperature, water salinity, and sea level rise, all of which directly affect the availability and quality of fish stocks [29]. Economically, fish foreign trade represents an important source of foreign exchange earnings and contributes to narrowing the food gap. However, climate variability may weaken export competitiveness and increase dependence on imports [1]. At the strategic level, the importance of this research is reinforced by its alignment with Egypt’s Vision 2030, which places food security and the sustainability of natural resources at the core of its development agenda. Accordingly, analyzing this relationship is essential for guiding future policy formulation [4].

2.3. Research Problem

The research problem centers on examining how climate change affects the performance of Egypt’s fish foreign trade. Although the fisheries sector constitutes a key pillar of food security and an important component of Egypt’s balance of payments, it faces increasing challenges due to climate change. Climatic phenomena such as rising temperatures, increasing water salinity, and changes in ocean current patterns have led to a decline in fish productivity in certain coastal areas. This has negatively affected domestic supply and weakened Egypt’s competitiveness in international markets [30]. In addition, these changes have increased reliance on fish imports to meet the food gap, thereby exerting additional pressure on the trade balance [1]. Accordingly, the central research question can be formulated as follows: How does climate change affect Egypt’s fish foreign trade?

2.4. Research Objectives

This study aims to achieve the following objectives:

To measure the impact of climate change on the performance of fish foreign trade in Egypt.

To examine and analyze the development of the following variables: average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, nitrous oxide emissions, the value of Egyptian fish exports, the value of fish imports, total trade volume, the fish trade balance, and the export to import coverage ratio over the period from 1995 to 2022.

To conduct an econometric analysis of the relationship between climate change and fish foreign trade in Egypt during the period of 1995 to 2022 using the Autoregressive Distributed Lag model.

To estimate and analyze the responsiveness of fish foreign trade in Egypt to climate change, and to determine whether a long-run relationship exists between them.

3. Methodology

3.1. Data Sources and Variables

This study examines the impact of climate change on the performance of Egypt’s fish foreign trade during the period of 1995–2022. The analysis is based on annual time-series data collected from several national and international sources. Data on fish exports, fish imports, total fish trade volume, and fish trade balance were obtained from the Central Agency for Public Mobilization and Statistics (CAPMAS) and the General Authority for Fish Resources Development in Egypt. Climate-related variables, including average surface air temperature, relative humidity, and rainfall, were obtained from the World Bank Climate Knowledge Portal [31]. Data on carbon dioxide emissions, methane emissions, and nitrous oxide emissions were obtained from the Food and Agriculture Organization of the United Nations (FAO). EViews 12 software was used to conduct the econometric and statistical analyses of the study.

The study incorporates four fish foreign trade indicators as dependent variables, namely fish exports (FEX), fish imports (FIM), total fish trade volume (FTV), and fish trade balance (FTB). The explanatory variables include average surface air temperature (TEMP), relative humidity (HUM), rainfall (RAIN), carbon dioxide emissions (CO₂), methane emissions (CH₄), and nitrous oxide emissions (N₂O).

3.2. Model Specification

To examine the relationship between climate change and fish foreign trade performance in Egypt, the study employs the Autoregressive Distributed Lag (ARDL) approach developed by Pesaran et al. The ARDL methodology is appropriate because it can be applied to variables integrated at mixed orders, I(0) and I(1), while also providing reliable estimates in small sample sizes. Moreover, the ARDL approach is particularly suitable for the current study because the analysis relies on a relatively small annual time-series sample covering the period of 1995–2022 and simultaneously examines both short-run dynamics and long-run equilibrium relationships between climate variables and fish foreign trade indicators. In addition, the ARDL framework allows simultaneous estimation of short-run dynamics and long-run equilibrium relationships among variables.

The functional forms of the estimated models are expressed as follows:

$$FE{X}_t = f(TEM{P}_t, HU{M}_t, RAI{N}_t, C{O}_{2t}, C{H}_{4t}, N_2{O}_t)$$

(1)

$$FI{M}_t=f(TEM{P}_t, HU{M}_t, RAI{N}_t, C{O}_{2t}, C{H}_{4t}, N_2{O}_t)$$

(2)

$$FT{V}_t=f(TEM{P}_t, HU{M}_t, RAI{N}_t, C{O}_{2t}, C{H}_{4t}, N_2{O}_t)$$

(3)

$$FT{B}_t=f(TEM{P}_t, HU{M}_t, RAI{N}_t, C{O}_{2t}, C{H}_{4t}, N_2{O}_t)$$

(4)

The general ARDL specification can be expressed as follows:

$$Y_t={\alpha }_0+{\displaystyle \sum _{i=1}^p} {\alpha }_i{Y}_{t-i}+{\displaystyle \sum _{j=0}^q} {\beta }_j{X}_{t-j}+{\epsilon }_t$$

(5)

where Y_t represents the fish foreign trade indicator, X_t denotes the climate change variables, α₀ is the intercept term, α_i and β_j are the estimated coefficients, p and q indicate the optimal lag lengths, and ε_t represents the random error term.

To estimate the short-run adjustments toward long-run equilibrium, the Error Correction Model (ECM) associated with the ARDL framework is expressed as follows:

$$\Delta Y_t={\alpha }_0+{\displaystyle \sum _{i=1}^p} {\alpha }_i\Delta Y_{t-i}+{\displaystyle \sum _{j=0}^q} {\beta }_j\Delta X_{t-j}+\lambda EC{M}_{t-1}+{\epsilon }_t$$

(6)

where ECM_t−1 represents the lagged error correction term and λ measures the speed of adjustment toward the long-run equilibrium.

3.3. ARDL Estimation Procedure

The empirical analysis was conducted in several stages. First, the stationarity properties of the variables were examined using the Augmented Dickey–Fuller (ADF) unit root test to determine the order of integration of each variable. Second, the optimal lag length structure was selected using Vector Autoregression (VAR) lag selection criteria. Third, the ARDL Bounds Testing approach was applied to investigate the existence of long-run cointegration relationships between climate change variables and fish foreign trade indicators.

Following the confirmation of cointegration relationships, the long-run coefficients and short-run dynamic relationships were estimated within the ARDL framework. The Error Correction Model (ECM) was subsequently estimated to evaluate the speed of adjustment from short-run disequilibrium toward long-run equilibrium. Finally, several diagnostic and stability tests, including serial correlation, heteroskedasticity, and model stability tests, were conducted to evaluate the robustness and reliability of the estimated models.

4. Results and Discussion

4.1. Trend Analysis of Climate Change and Egypt’s Fish Foreign Trade During the Period from 1995 to 2022

Figure 1 illustrates the temporal evolution of the main climate change indicators in Egypt during the period of 1995–2022. The results reveal a statistically significant upward trend in average surface air temperature, which increased at an annual rate of approximately 0.034 °C, with an average value of 23.26 °C during the study period. The estimated trend was statistically significant at the 1% level, with an (R²) value of 0.359, indicating that approximately 35.9% of the variation in temperature was explained by the time trend. In contrast, relative humidity exhibited a statistically significant declining trend, with an estimated annual decrease of 0.067%, reflecting increasingly arid climatic conditions and rising environmental stress on aquatic ecosystems.

Rainfall showed a weak but statistically significant upward trend, increasing annually by approximately 0.194 mm, although Egypt continued to maintain its predominantly arid climate characteristics. In addition, carbon dioxide emissions resulting from energy use in agriculture recorded a statistically significant upward trend, increasing annually by approximately 171.942 kilotons, with an average value of 6026.99 kilotons during the study period. Methane and nitrous oxide emissions also displayed positive trends over time, although these trends were not statistically significant. Overall, these findings highlight the increasing environmental pressures associated with climate change and their potential implications for the sustainability and long-term resilience of Egypt’s fisheries sector.

Figure 2 illustrates the temporal trends in Egypt’s fish foreign trade indicators during the period of 1995–2022. The results reveal a statistically significant upward trend in the value of Egyptian fish exports, which recorded an average value of approximately 16.39 million EGP and increased annually by about 1.668 million EGP. The estimated trend model was highly significant at the 1% level, with an (R²) value of 0.712, indicating that approximately 71.2% of the variation in fish exports was explained by the time trend. This upward trend reflects the growing competitiveness of Egyptian fish products in international markets and the expansion of aquaculture production and export-oriented activities.

The results also indicate a statistically significant upward trend in the coverage ratio (exports to imports ratio), which averaged approximately 12.36% during the study period and increased annually by about 0.784%. The estimated trend equation was statistically significant at the 1% level, with an (R²) value of 0.554, suggesting a gradual improvement in the trade efficiency of Egypt’s fisheries sector. This improvement reflects the increasing growth of fish exports relative to imports and highlights the sector’s growing capacity to adapt to economic and climatic changes through production expansion and modernization efforts.

In contrast, fish imports, total trade volume, and fish trade balance did not exhibit statistically significant trends over the study period, despite observable fluctuations across different years. Overall, the findings indicate that climate-related pressures coexist with gradual structural improvements in Egypt’s fisheries trade sector, particularly through export growth and improved coverage ratios, which may contribute to strengthening sector resilience and sustainability under changing climatic conditions.

4.2. Results of the Econometric Model Estimation for Measuring the Economic Effects of Climate Change on Egyptian Fish Exports

4.2.1. Estimation of the Cointegration Model Using the ARDL Approach

The Autoregressive Distributed Lag model was estimated using the value of Egyptian fish exports as the dependent variable, while climate change indicators were included as independent variables. These indicators comprise average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. The ARDL model is employed to test for the existence of a long-run equilibrium relationship among the variables, that is, to determine whether cointegration exists between them.

4.2.2. Steps for Applying the ARDL Model

Determining the Stationarity of the Series

The Augmented Dickey–Fuller unit root test was conducted for all variables, including the value of Egyptian fish exports as the dependent variable and the climate variables as independent variables. The results indicate that the time series are stationary at the level for the value of fish exports, average surface air temperature, relative humidity, and rainfall. In contrast, the remaining variables become stationary after first differencing. These results confirm that the variables are integrated of different orders, which justifies the application of the ARDL model.

Table 1. Presents the results of the unit root test for the stationarity of the time series.

Degree of Integration	1st Difference			Level
	none	trend & intercept	intercept	none	trend & intercept	intercept
I(0)	−9.58 **	−9.49 **	−9.86 **	−3.39 **	−1.80	−3.66 *	Total value of fish exports
I(0)	−10.82 **	−7.19 **	−7.19 **	0.89	−7.06 **	−4.56 **	Average surface air temperature
I(0)	−9.33 **	−6.01 **	−9.18 **	−0.49	−4.58 **	−3.86 **	Relative humidity%
I(0)	−4.58 **	−5.23 **	−4.64 **	0.91	−4.12 *	−3.74 **	Rain
I(1)	−4.90 **	−5.02 **	−4.89 **	0.09	−1.28	−1.69	Carbon gas emission
I(1)	−4.81 **	−4.88 **	−4.73 **	−0.31	−1.27	−1.60	Methane emission
I(1)	−4.78 **	−4.77 **	−4.69 **	−0.56	−1.39	−1.61	Emission of nitrous gas

Source: Calculated using EViews 12. Note: Critical values at the 1% significance level are −3.75 (Intercept), −4.44 (Trend and Intercept), and −2.67 (None). Critical values at the 5% significance level are −2.99 (Intercept), −3.63 (Trend and Intercept), and −1.96 (None). * Significant at the 5% level. ** Significant at the 1% level.

.

Determination of the Optimal Lag Length

A Vector Autoregression (VAR) model was estimated to determine the optimal lag length between the dependent variable and the independent variables. The results indicate that the optimal lag length for the model is two periods.

Table 2. Results of lag length selection criteria.

Lag	LogL	LR	FPE	AIC	SC	HQ
0	−360.23	NA	4375.61	28.25	28.59	28.35
1	−242.71	162.73	25.96	22.98	25.69	23.76
2	−161.91	68.36 *	6.05 *	20.53 *	25.61 *	21.99 *

Source: Calculated using EViews 12. Note: * indicates the optimal lag length selected according to the corresponding information criterion. Abbreviations: LR, sequential modified likelihood ratio test statistic; FPE, final prediction error; AIC, Akaike information criterion; SC, Schwarz information criterion; HQ, Hannan–Quinn information criterion.

presents the results of the lag length selection criteria.

ARDL Bounds Test

This step involves estimating the ARDL model to test for cointegration among the variables. The results presented in

Table 3. Results of the cointegration test using the ARDL methodology and the ARDL Bounds Test.

F-Bounds Test		Null Hypothesis: No Level Relationship
Test Statistic	Value	Signif.	I(0)	I(1)
			Asymptotic: n = 1000
F-statistic	3.594382	10%	1.99	2.94
K	6	5%	2.27	3.28
		2.5%	2.55	3.61
		1%	2.88	3.99
Actual Sample Size	26		Finite Sample: n = 35
		10%	2.254	3.388
		5%	2.685	3.96
		1%	3.713	5.326
			Finite Sample: n = 30
		10%	2.334	3.515
		5%	2.794	4.148
		1%	3.976	5.691

Source: Calculated using EViews 12.

indicate that the calculated F-statistic is 3.59, which exceeds the upper critical bound at the one percent significance level. Accordingly, the null hypothesis of no cointegration is rejected.

This finding confirms the existence of a long-run equilibrium relationship between the value of Egyptian fish exports and the climate change variables, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. In other words, cointegration among these variables is established at the one percent significance level.

Estimation of the Long Run Relationship in the ARDL Model

The relationship between climate change variables and Egypt’s fish foreign trade is interpreted over both the long run and short run during the period from 1995 to 2022, as presented in

Table 4. Results of long-run equation estimation according to the ARDL model (2, 1, 0, 2, 0, 0, 0).

Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	−259.2396	181.5593	−1.427851	0.1753
$\mathrm{T}\mathrm{O}\mathrm{T}\mathrm{A}\mathrm{L}\_\mathrm{V}\mathrm{A}\mathrm{L}\mathrm{U}\mathrm{E}\_\mathrm{O}\mathrm{F}\_\mathrm{F}\mathrm{I}\mathrm{S}\mathrm{H}\_\mathrm{E}\mathrm{X}\mathrm{P}\mathrm{O}\mathrm{R}\mathrm{T}\mathrm{S}(-1)\ast$	−0.574901	0.293437	−1.959198	0.0703
$\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E}(-1)$	10.98625	6.286548	1.747581	0.1024
$\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_\ast \ast$	0.009367	2.124397	0.004409	0.9965
$\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}(-1)$	−0.254153	0.703159	−0.361445	0.7232
$\mathrm{C}\mathrm{A}\mathrm{R}\mathrm{B}\mathrm{O}\mathrm{N}\_\mathrm{G}\mathrm{A}\mathrm{S}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\ast \ast$	0.004876	0.005034	0.968474	0.3492
$\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\ast \ast$	66.42494	77.44166	0.857742	0.4055
$\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S}\ast \ast$	−25.92975	11.76825	−2.203364	0.0448
$\mathrm{D}(\mathrm{T}\mathrm{O}\mathrm{T}\mathrm{A}\mathrm{L}\_\mathrm{V}\mathrm{A}\mathrm{L}\mathrm{U}\mathrm{E}\_\mathrm{O}\mathrm{F}\_\mathrm{F}\mathrm{I}\mathrm{S}\mathrm{H}\_\mathrm{E}\mathrm{X}\mathrm{P}\mathrm{O}\mathrm{R}\mathrm{T}\mathrm{S}(-1\left)\right)$	−0.401234	0.260678	−1.539196	0.1460
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E})$	5.230312	3.939284	1.327732	0.2055
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right)$	0.031076	0.397620	0.078154	0.9388
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right(-1\left)\right)$	1.129651	0.382943	2.949922	0.0105

Source: Calculated using EViews software. Note: * and ** indicate significance at the 5% and 1% levels, respectively.

.

First: Long Run Interpretation

The long-run component of the model reflects the equilibrium relationship between climate variables and the value of Egyptian fish exports over time, that is, after short-term shocks have dissipated and the variables have adjusted toward their stable equilibrium levels. The estimated ARDL specification ARDL (2, 1, 0, 2, 0, 0, 0) indicates that the dependent variable entered the model with two lag periods, while rainfall also entered with two lag periods, reflecting the dynamic adjustment process between climate variables and fish export performance over time.

Nitrous Oxide Emissions (EMISSION_OF_NITROUS_GAS)

The results indicate that nitrous oxide emissions have a negative and statistically significant effect on Egyptian fish exports at the 0.0448 significance level. This suggests that increasing nitrous oxide emissions contributes to a decline in fish export values in the long run.

This negative relationship may be attributed to the adverse environmental effects associated with greenhouse gas emissions, particularly their role in deteriorating water quality and increasing environmental stress on aquatic ecosystems. Such conditions may reduce fish productivity and weaken the export competitiveness of the fisheries sector over time.

Other Climate Variables (Temperature, Relative Humidity, Rainfall, Carbon Dioxide Emissions, and Methane Emissions)

The remaining climate variables did not exhibit statistically significant long-run effects despite differences in the signs of their estimated coefficients. This indicates that their effects on Egypt’s fish foreign trade may be indirect or intertwined with other economic and production-related factors, including technological development, fisheries management systems, market demand conditions, and production efficiency.

Lagged Value of Fish Exports (TOTAL_VALUE_OF_FISH_EXPORTS(−1))

Consistent with the selected ARDL (2, 1, 0, 2, 0, 0, 0) specification, the lagged dependent variable entered the model with two lag periods to capture the dynamic adjustment behavior of fish exports over time. The first lag of fish exports showed a negative coefficient and was marginally significant at the 0.0703 probability level. This finding suggests the existence of a partial adjustment mechanism, whereby unusually high export levels in one period tend to be followed by corrective movements toward equilibrium levels in subsequent periods.

The ARDL methodology distinguishes between the long-run equilibrium relationship and the short-run dynamic adjustment process. Accordingly, Table 4 presents the long-run coefficients for all explanatory variables included in the estimated ARDL model, whereas

Table 5. Presents the results of the Error Correction Model for cointegration.

ECM Regression
Variable	Coefficient	Std. Error	t-Statistic	Prob.
$\mathrm{D}(\mathrm{T}\mathrm{O}\mathrm{T}\mathrm{A}\mathrm{L}\_\mathrm{V}\mathrm{A}\mathrm{L}\mathrm{U}\mathrm{E}\_\mathrm{O}\mathrm{F}\_\mathrm{F}\mathrm{I}\mathrm{S}\mathrm{H}\_\mathrm{E}\mathrm{X}\mathrm{P}\mathrm{O}\mathrm{R}\mathrm{T}\mathrm{S}(-1\left)\right)$	−0.401234	0.147364	−2.722743	0.0165
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E})$	5.230312	1.580431	3.309421	0.0052
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right)$	0.031076	0.183630	0.169230	0.8680
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right(-1\left)\right)$	1.129651	0.175909	6.421797	0.0000
$\mathrm{C}\mathrm{o}\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{E}\mathrm{q}(-1)$	−0.574901	0.087537	−6.567540	0.0000
$\mathrm{R}—\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}$	0.889676	Mean dependent var.		0.261538
$\mathrm{A}\mathrm{d}\mathrm{j}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{R}—\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}$	0.868662	S.D. dependent var.		10.91916
$\mathrm{S}.\mathrm{E}.\mathrm{o}\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}$	3.957173	Akaike info. Criterion		5.759978
$\mathrm{S}\mathrm{u}\mathrm{m}\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{i}\mathrm{d}$	328.8437	Schwarz criterion		6.001920
$\mathrm{L}\mathrm{o}\mathrm{g}\mathrm{l}\mathrm{i}\mathrm{k}\mathrm{e}\mathrm{l}\mathrm{i}\mathrm{h}\mathrm{o}\mathrm{o}\mathrm{d}$	−69.87972	Hannan–Quinn criter.		5.829649
$\mathrm{D}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{n}—\mathrm{W}\mathrm{a}\mathrm{t}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{t}$	2.264833		F-statistic	3.594382

Source: Calculated using EViews 12.

reports the Error Correction Model (ECM) results representing the short-run dynamics. Therefore, only the differenced variables and lagged adjustment terms generated by the selected lag structure appear in the short-run specification. Accordingly, the short-run equation includes only the variables that exhibited dynamic short-term effects within the selected ARDL specification.

Second: Short Run Interpretation

The short-run component of the ARDL model captures the immediate responses of fish export values to short-term changes in climate conditions and environmental variables.

Lagged Change in Rainfall D(RAIN(−1))

Consistent with the selected lag structure of the ARDL model, rainfall entered the model with two lag periods. The lagged change in rainfall exhibited a positive and statistically significant effect at the 0.0105 significance level. This implies that increases in rainfall during previous periods contribute positively to fish export values in the short run.

This positive effect may be explained by the improvement in environmental and ecological conditions associated with increased rainfall, including improved water availability and enhanced productivity in fisheries and aquaculture systems.

Other Variables (Temperature, Gas Emissions, and Relative Humidity)

The remaining explanatory variables did not exhibit statistically significant short-run effects. This suggests that short-term fluctuations in climate conditions may not directly translate into immediate changes in fish foreign trade performance, but rather require time to affect production systems, market supply conditions, and export capacity.

Third: General Conclusion

Based on the estimated ARDL results, the findings confirm the existence of a relatively weak long-run relationship between climate variables and Egyptian fish exports, with nitrous oxide emissions representing the most significant negative environmental factor affecting export performance. In the short run, rainfall emerged as the most important positive climatic variable supporting fish export growth. Overall, the model indicates that climate variables affect fish foreign trade primarily through indirect channels related to environmental quality, fisheries productivity, and production sustainability rather than through immediate direct effects on trade performance.

Estimation of the Error Correction Model According to the ARDL Framework

After completing the previous steps, which form the basis for the subsequent analysis, the Error Correction Model was estimated. The value of the error correction term is expected to lie between zero and a negative value, reflecting the speed of adjustment toward the long-run equilibrium.

• Economic Evaluation of the Cointegration Model Between Egyptian Fish Exports and Climate Change Variables

The results confirm the existence of a long-run relationship between the value of Egyptian fish exports and the climate variables, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. This is supported by the error correction term, which has a value of −0.574901.

This indicates that approximately 57 percent of short-term deviations from equilibrium are corrected in the long run. In other words, the system exhibits a relatively high speed of adjustment, confirming the presence of cointegration among the variables. Table 5 presents the results of the Error Correction Model.

• Statistical Evaluation of the Estimated Model

The results demonstrate that both the value of Egyptian fish exports and the error correction term are statistically significant. The coefficient of determination indicates that 88.9 percent of the variation in fish export values is explained by the included explanatory variables, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. This finding is consistent with the economic interpretation of the model. Accordingly, approximately 11.1 percent of the variation is attributed to other factors not captured within the model. Furthermore, the overall model is statistically significant, as indicated by the F-statistic.

Diagnostic Tests for Model Validity

• Normality and Heteroskedasticity Test

Table 6. The results of the Breusch–Pagan–Godfrey test for heteroskedasticity.

Heteroskedasticity Test: Breusch–Pagan–Godfrey
Null Hypothesis: Homoskedasticity
F-statistic	2.492931	Prob. F (11,14)	0.0552
Obs × R-squared	17.21245	Prob. Chi-Square (11)	0.1017
Scaled explained SS	4.881755	Prob. Chi-Square (11)	0.9368

Source: Calculated using EViews 12.

indicates that all probability values exceed 0.05, suggesting that there is no statistically significant evidence of heteroskedasticity. Accordingly, the null hypothesis of homoskedasticity cannot be rejected, implying that the variance of the residuals is constant.

This result confirms that one of the key assumptions of the econometric model is satisfied, thereby supporting the reliability and efficiency of the estimated coefficients.

• Serial Correlation Test

Since the probability value exceeds 0.05, the null hypothesis is not rejected at the five percent significance level. This indicates that there is no evidence of serial correlation among the residuals.

These results suggest that the estimated model does not suffer from autocorrelation problems in the error terms, thereby supporting the validity of the model’s specification.

Table 7. The results of the Breusch–Godfrey serial correlation LM test.

Breusch–Godfrey Serial Correlation LM Test:
F-statistic	0.818042	Prob. F (2,12)	0.4645
Obs × R-squared	3.119532	Prob. Chi-Square (2)	0.2102

Source: Calculated using EViews 12.

.

• Results of Estimating the Econometric Model for Measuring the Economic Effects of Climate Change and Their Impact on Egyptian Fish Exports

Stability Tests of Model Coefficients (CUSUM and CUSUM of Squares)

“The CUSUM and CUSUM of Squares tests were employed to examine the structural stability of the model coefficients over the study period. The results of the CUSUM test indicate that the model exhibits an acceptable degree of stability, as the test statistic remained within the critical bounds at the 5% significance level. In contrast, the CUSUM of Squares test revealed noticeable fluctuations and deviations in the test curve, with the statistic approaching the critical bounds during several periods and rising markedly toward the end of the time series. This may reflect certain signs of instability or the possibility of partial structural changes in the model over the study period. Accordingly, the model results should be interpreted with relative caution, particularly with respect to the long-run relationships.” Figure 3.

5. Results of the Econometric Model Estimation for Measuring the Economic Effects of Climate Change on Egyptian Fish Imports

5.1. Estimation of the Cointegration Model Using the ARDL Approach

The Autoregressive Distributed Lag model was estimated using the value of Egyptian fish imports as the dependent variable, while climate change indicators were included as independent variables. These variables include average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. The ARDL model is applied to test for the existence of a long-run equilibrium relationship among the variables, that is, to determine whether cointegration exists.

5.2. Steps for Applying the ARDL Model

5.2.1. Determining the Stationarity of the Series

The Augmented Dickey–Fuller unit root test was conducted for all variables, including the value of Egyptian fish imports as the dependent variable and the climate variables as independent variables. The results indicate that the time series are stationary at the level for fish imports, average surface air temperature, relative humidity, and rainfall. The remaining variables become stationary after first differencing, as presented in

Table 8. The results of the unit root test for the stationarity of the time series.

Degree of Integration	1st Difference			Level
	none	trend & intercept	intercept	none	trend & intercept	intercept
I(0)	−7.47 **	−7.26 **	−7.33 **	−1.56	−2.73	−2.79 *	Total value of fish imports
I(0)	−10.82 **	−7.19 **	−7.19 **	0.89	−7.06 **	−4.56 **	Average surface air temperature
I(0)	−9.33 **	−6.01 **	−9.18 **	−0.49	−4.58 **	−3.86 **	Relative humidity%
I(0)	−4.58 **	−5.23 **	−4.64 **	0.91	−4.12 *	−3.74 **	Rain
I(1)	−4.90 **	−5.02 **	−4.89 **	0.09	−1.28	−1.69	Carbon gas emission
I(1)	−4.81 **	−4.88 **	−4.73 **	−0.31	−1.27	−1.6	Methane emission
I(1)	−4.78 **	−4.77 **	−4.69 **	−0.56	−1.39	−1.61	Emission of nitrous gas

Source: Calculated using EViews 12. Note: Critical values at the 1% significance level are −3.75 (Intercept), −4.44 (Trend and Intercept), and −2.67 (None). Critical values at the 5% significance level are −2.99 (Intercept), −3.63 (Trend and Intercept), and −1.96 (None). * Significant at the 5% level. ** Significant at the 1% level.

. These findings confirm that the variables are integrated of different orders, which supports the suitability of applying the ARDL methodology.

5.2.2. Determination of the Optimal Lag Length

A Vector Autoregression model was estimated to identify the optimal lag length between the dependent variable and the independent variables. The results indicate that the optimal lag length for the model is two periods.

Table 9. Presents the results of the lag length selection criteria.

Lag	LogL	LR	FPE	AIC	SC	HQ
0	−424.366	NA	607,511.4	33.182	33.52072	33.27954
1	−318.473	146.621	8819.39	28.80562	31.51536	29.58593
2	−236.0538	69.73934 *	1812.720 *	26.23491 *	31.31568 *	27.69799 *

Source: Calculated using EViews 12. * indicates lag order selected by the criterion. LR: sequential modified LR test statistic (each test at 5% level). FPE: Final prediction error. AIC: Akaike information criterion. SC: Schwarz information criterion. HQ: Hannan–Quinn information criterion.

.

5.2.3. ARDL Bounds Test

This step involves estimating the ARDL model to test for cointegration among the variables. The results presented in

Table 10. Results of the ARDL Bounds Test for cointegration.

F-Bounds Test		Null Hypothesis: No Level Relationship
Test Statistic	Value	Sign.	I(0)	I(1)
			Asymptotic: n = 1000
F-statistic	2.045184	10%	1.99	2.94
k	6	5%	2.27	3.28
		2.5%	2.55	3.61
		1%	2.88	3.99
Actual Sample Size	26		Finite Sample: n = 35
		10%	2.254	3.388
		5%	2.685	3.96
		1%	3.713	5.326
			Finite Sample: n = 30
		10%	2.334	3.515
		5%	2.794	4.148
		1%	3.976	5.691

Source: Calculated using EViews 12.

indicate that the calculated F-statistic is 2.05, which is lower than the upper critical bound at the one percent significance level. Accordingly, the null hypothesis of no cointegration is accepted.

This finding implies that there is no long-run equilibrium relationship between the value of Egyptian fish imports and the climate change variables, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. In other words, no cointegration exists among these variables at the one percent significance level.

Therefore, the analysis is limited to short-run dynamics only, as the variables do not move together in the long run. Based on this result, the ARDL model can be interpreted within a short-run framework without incorporating an error correction term.

5.2.4. Estimation of the ARDL Model and Interpretation of Short Run Dynamics

Given the absence of a long-run relationship, the analysis focuses on the short-run dynamics between climate change variables and Egyptian fish imports over the period from 1995 to 2022, as presented in

Table 11. Presents the results of the ARDL model estimation for the short run dynamics under the specification ARDL (1, 2, 2, 2, 1, 2, 2).

Conditional Error Correction Regression
Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	14,066.09	10,789.40	1.303695	0.2336
$\mathrm{T}\mathrm{O}\mathrm{T}\mathrm{A}\mathrm{L}\_\mathrm{V}\mathrm{A}\mathrm{L}\mathrm{U}\mathrm{E}\_\mathrm{O}\mathrm{F}\_\mathrm{F}\mathrm{I}\mathrm{S}\mathrm{H}\_\mathrm{I}\mathrm{M}\mathrm{P}\mathrm{O}\mathrm{R}\mathrm{T}\mathrm{S}(-1)$	−1.069951	0.291811	−3.666586	0.0080
$\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E}(-1)$	−348.1558	282.7214	−1.231445	0.2579
$\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_(-1)$	−174.3838	143.4616	−1.215543	0.2636
$\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}(-1)$	63.22520	21.72164	2.910702	0.0226
$\mathrm{C}\mathrm{A}\mathrm{R}\mathrm{B}\mathrm{O}\mathrm{N}\_\mathrm{G}\mathrm{A}\mathrm{S}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}(-1)$	−0.237045	0.111079	−2.134011	0.0703
$\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}(-1)$	8974.023	4220.833	2.126126	0.0711
$\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S}(-1)$	−706.7503	467.2396	−1.512608	0.1741
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E})$	−143.5782	102.4073	−1.402031	0.2037
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E}(-1\left)\right)$	148.4343	109.0522	1.361131	0.2157
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_)$	−194.0797	85.19677	−2.278017	0.0568
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_(-1\left)\right)$	105.6740	68.66430	1.538995	0.1677
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right)$	49.64275	15.35942	3.232072	0.0144
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right(-1\left)\right)$	−36.27805	15.19809	−2.387013	0.0484
$\mathrm{D}(\mathrm{C}\mathrm{A}\mathrm{R}\mathrm{B}\mathrm{O}\mathrm{N}\_\mathrm{G}\mathrm{A}\mathrm{S}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N})$	−0.389746	0.161282	−2.416555	0.0463
$\mathrm{D}(\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N})$	6210.239	2431.217	2.554374	0.0379
$\mathrm{D}(\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}(-1\left)\right)$	−6362.747	3756.237	−1.693915	0.1341
$\mathrm{D}(\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S})$	247.6623	359.7264	0.688474	0.5133
$\mathrm{D}(\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S}(-1\left)\right)$	965.8001	587.9696	1.642602	0.1445

Source: Calculated using EViews 12.

.

Short Run Interpretation

The short-run component reflects the immediate responses of fish import values to changes in climate variables, that is, the short-term effects observed before any potential long-run equilibrium is reached.

Current Change in Rainfall D (RAIN)

This variable shows a positive and statistically significant effect at the 0.0144 level, indicating that an increase in rainfall in the short run leads to higher fish imports. This may be explained by the temporary disruption of domestic fishing activities or internal supply chains during periods of heavy rainfall, prompting an increase in imports to meet local demand.

Lagged Change in Rainfall D (RAIN (−1))

The effect is negative and statistically significant at the 0.0484 level, suggesting that higher rainfall in previous periods contributes to a reduction in fish imports in subsequent periods. This may be due to improved fishery productivity or stabilization of domestic supply, thereby reducing the need for imports.

Change in Carbon Dioxide Emissions D (CARBON_GAS_EMISSION)

This variable has a negative and statistically significant effect at the 0.0463 level. This indicates that an increase in carbon emissions is associated with a decline in fish imports in the short run, possibly due to economic slowdowns or increased energy and transportation costs that constrain external trade.

Change in Methane Emissions D (METHANE_EMISSION)

The effect is positive and statistically significant at the 0.0379 level, suggesting that a sudden increase in methane emissions may lead to a short-term rise in fish imports. This could reflect increased consumption demand or temporary shifts in domestic production patterns.

Change in Relative Humidity D (RELATIVE_HUMIDITY)

This variable exhibits a negative effect and is marginally significant at the 0.0568 level. This suggests that higher humidity levels may reduce fish imports in the short run, possibly due to improved environmental conditions for domestic production or reduced post-harvest losses during storage and transportation.

General Conclusion

The short-run analysis reveals that the relationship between climate change and Egyptian fish imports is complex and multidimensional. Climate variables influence imports both directly and indirectly through their effects on domestic production and market demand. Rainfall and greenhouse gas emissions emerge as the most responsive variables in the short run, showing immediate impacts on import decisions.

These findings highlight the sensitivity of Egypt’s fish market to sudden climatic fluctuations. Moreover, the results suggest the presence of a self-correcting mechanism within the fish trade system, indicating that the sector tends to stabilize after experiencing climatic or economic shocks.

5.2.5. Estimation of the Error Correction Model According to the ARDL Approach

After completing the previous steps, which constitute the basis for conducting the subsequent analysis, the Error Correction Model (ECM) was estimated. The error correction coefficient is expected to take a value between zero and negative one.

Economic Evaluation of the Cointegration Model Between the Value of Egyptian Fish Imports and Climate Change Variables

The results confirm the existence of a long-run relationship between the value of Egyptian fish imports and the explanatory variables represented by average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. This conclusion is supported by the estimated error correction coefficient, which reached (−1.069951). This indicates that approximately 106% of the short-run disequilibrium errors are corrected in the long run, confirming the presence of a cointegration relationship among the variables.

Table 12. Results of the Error Correction Model for cointegration.

ECM Regression
Variable	Coefficient	Std. Error	t-Statistic	Prob.
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E})$	−143.5782	44.75094	−3.208383	0.0149
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E}(-1\left)\right)$	148.4343	39.80410	3.729121	0.0074
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_)$	−194.0797	35.67648	−5.439990	0.0010
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_(-1\left)\right)$	105.6740	26.49213	3.988882	0.0053
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right)$	49.64275	7.840668	6.331444	0.0004
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right(-1\left)\right)$	−36.27805	7.279528	−4.983571	0.0016
$\mathrm{D}(\mathrm{C}\mathrm{A}\mathrm{R}\mathrm{B}\mathrm{O}\mathrm{N}\_\mathrm{G}\mathrm{A}\mathrm{S}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N})$	−0.389746	0.087384	−4.460140	0.0029
$\mathrm{D}(\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N})$	6210.239	1438.810	4.316233	0.0035
$\mathrm{D}(\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}(-1\left)\right)$	−6362.747	1552.264	−4.099010	0.0046
$\mathrm{D}(\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S})$	247.6623	183.4357	1.350131	0.2190
$\mathrm{D}(\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S}(-1\left)\right)$	965.8001	238.8048	4.044308	0.0049
$\mathrm{C}\mathrm{o}\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{E}\mathrm{q}(-1)$	−1.069951	0.187041	−5.720397	0.0007
$\mathrm{R}—\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}$	0.770168	Mean dependent var.		−6.688462
$\mathrm{A}\mathrm{d}\mathrm{j}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{R}—\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}$	0.589586	S.D. dependent var.		97.06194
$\mathrm{S}.\mathrm{E}.\mathrm{o}\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}$	62.18132	Akaike info. Criterion		11.40202
$\mathrm{S}\mathrm{u}\mathrm{m}\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{i}\mathrm{d}$	54,131.23	Schwarz criterion		11.98268
$\mathrm{L}\mathrm{o}\mathrm{g}\mathrm{l}\mathrm{i}\mathrm{k}\mathrm{e}\mathrm{l}\mathrm{i}\mathrm{h}\mathrm{o}\mathrm{o}\mathrm{d}$	−136.2263	Hannan–Quinn criter.		11.56923
$\mathrm{D}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{n}—\mathrm{W}\mathrm{a}\mathrm{t}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{t}$	2.562451

Source: Calculated using EViews 12.

presents the results of the Error Correction Model.

Statistical Evaluation of the Estimated Model

The statistical results demonstrate the significance of both the value of Egyptian fish imports and the error correction coefficient. In addition, the coefficient of determination indicates that 77.01% of the variation in the value of Egyptian fish imports is explained by the independent variables included in the model, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. This finding is consistent with the economic evaluation of the model, while the remaining 22.99% is attributed to other factors not captured by the model. Furthermore, the estimated model was found to be statistically significant based on the value of the F-statistic.

5.2.6. Diagnostic Tests for Model Validity

Normality and Heteroskedasticity Test

Table 13. Results of the Breusch–Pagan–Godfrey Test for Heteroskedasticity.

Test Statistic	Value	Probability Statistic	Prob.
F-statistic	0.426579	Prob. F (18,7)	0.9312
Obs × R-squared	13.60085	Prob. Chi-Square (18)	0.7547
Scaled explained SS	1.042397	Prob. Chi-Square (18)	1.0000

Null Hypothesis: Homoskedasticity. Source: Calculated using EViews 12.

shows that all probability values exceed 0.05, indicating that there is no statistically significant evidence of heteroskedasticity. Accordingly, the null hypothesis of homoskedasticity cannot be rejected, implying that the variance of the residuals is constant. This result confirms that one of the fundamental assumptions of the econometric model is satisfied, thereby supporting the reliability and consistency of the estimated results.

Serial Correlation Test

Since the probability value exceeds 0.05, the null hypothesis is not rejected at the five percent significance level. This indicates that there is no evidence of serial correlation among the residuals. These results confirm that the estimated model does not suffer from autocorrelation problems in the error terms, which supports the validity and robustness of the model

Table 14. Results of the Breusch–Godfrey Serial Correlation LM Test.

Test Statistic	Value	Probability Statistic	Prob.
F-statistic	2.314255	Prob. F (2,5)	0.1943
Obs × R-squared	12.49843	Prob. Chi-Square (2)	0.0019

Null Hypothesis: No serial correlation. Source: Calculated using EViews 12.

.

Results of Estimating the Econometric Model for Measuring the Economic Effects of Climate Change and Their Impact on Egyptian Fish Imports

Stability Tests of Model Coefficients (CUSUM and CUSUM of Squares)

To examine the structural stability of the estimated model coefficients and assess their consistency over time, the CUSUM and CUSUM of Squares tests were conducted. The figures indicate that the plots of both tests remained entirely within the critical bounds at the 5% significance level, confirming the absence of any structural breaks during the study period. These findings demonstrate that the estimated coefficients are stable and consistent over time, thereby supporting the reliability of the estimated model and the validity of the derived results for economic analysis and long-run interpretation. The results are illustrated in Figure 4.

6. Results of the Econometric Model Estimation for Measuring the Economic Effects of Climate Change on Total Fish Trade in Egypt

6.1. Estimation of the Cointegration Model Using the ARDL Approach

The Autoregressive Distributed Lag model was estimated using the total value of fish trade in Egypt as the dependent variable, while climate change indicators were included as independent variables. These variables include average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. The ARDL model is applied to test for the existence of a long-run equilibrium relationship among the variables, that is, to determine whether cointegration exists.

6.2. Steps for Applying the ARDL Model

6.2.1. Determining the Stationarity of the Series

The Augmented Dickey–Fuller unit root test was conducted for all variables, including total fish trade as the dependent variable and the climate variables as independent variables. The results indicate that the time series are stationary at the level for total fish trade, average surface air temperature, relative humidity, and rainfall. The remaining variables become stationary after first differencing, as shown in

Table 15. The results of the unit root test for the stationarity of the time series.

Degree of Integration	1st Difference			Level
	none	trend & intercept	intercept	none	trend & intercept	intercept
I(0)	−7.36 **	−7.12 **	−7.21 **	−1.45	−2.69	−2.73 *	Total trade volume
I(0)	−10.82 **	−7.19 **	−7.19 **	0.89	−7.06 **	−4.56 **	Average surface air temperature
I(0)	−9.33 **	−6.01 **	−9.18 **	−0.49	−4.58 **	−3.86 **	Relative humidity%
I(0)	−4.58 **	−5.23 **	−4.64 **	0.91	−4.12 *	−3.74 **	Rain
I(1)	−4.90 **	−5.02 **	−4.89 **	0.09	−1.28	−1.69	Carbon gas emission
I(1)	−4.81 **	−4.88 **	−4.73 **	−0.31	−1.27	−1.6	Methane emission
I(1)	−4.78 **	−4.77 **	−4.69 **	−0.56	−1.39	−1.61	Emission of nitrous gas

Source: Calculated using EViews 12. Note: Critical values at the 1% significance level are −3.75 (Intercept), −4.44 (Trend and Intercept), and −2.67 (None). Critical values at the 5% significance level are −2.99 (Intercept), −3.63 (Trend and Intercept), and −1.96 (None). * Significant at the 5% level. ** Significant at the 1% level.

. These findings confirm that the variables are integrated of different orders, which supports the appropriateness of applying the ARDL methodology.

6.2.2. Determination of the Optimal Lag Length

A Vector Autoregression model was estimated to determine the optimal lag length between the dependent variable and the independent variables. The results indicate that the optimal lag length for the model is two periods.

Table 16. The results of the lag length selection criteria.

Lag	LogL	LR	FPE	AIC	SC	HQ
0	−424.9979	NA	637,769.2	33.23061	33.56932	33.32814
1	−318.4917	147.4701	8832.06	28.80705	31.5168	29.58736
2	−236.5262	69.35543 *	1879.800 *	26.27125 *	31.35202 *	27.73432 *

Source: Calculated using EViews 12. * indicates lag order selected by the criterion. LR: sequential modified LR test statistic (each test at 5% level). FPE: Final prediction error. AIC: Akaike information criterion. SC: Schwarz information criterion. HQ: Hannan–Quinn information criterion.

.

6.2.3. ARDL Bounds Test

This step involves estimating the ARDL model to examine the existence of cointegration among the variables. The results presented in

Table 17. The results of the ARDL Bounds Test for cointegration.

F-Bounds Test		Null Hypothesis: No Level Relationship
Test Statistic	Value	Signif.	I(0)	I(1)
			Asymptotic: n = 1000
F-statistic	2.075951	10%	1.99	2.94
k	6	5%	2.27	3.28
		2.5%	2.55	3.61
		1%	2.88	3.99
Actual Sample Size	26		Finite Sample: n = 35
		10%	2.254	3.388
		5%	2.685	3.96
		1%	3.713	5.326
			Finite Sample: n = 30
		10%	2.334	3.515
		5%	2.794	4.148
		1%	3.976	5.691

Source: Calculated using EViews 12.

indicate that the calculated F-statistic is 2.076, which is lower than the upper critical bound at the one percent significance level. Accordingly, the null hypothesis of no cointegration is accepted.

This finding indicates that there is no long-run equilibrium relationship between total fish trade in Egypt and the climate change variables, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. In other words, no cointegration exists among these variables at the one percent significance level. Therefore, the analysis is restricted to short-run dynamics, as the variables do not move together in the long run. Based on this result, the ARDL model can be interpreted within a short-run framework without including an error correction term.

6.2.4. Estimation of the ARDL Model and Short Run Interpretation

In light of the absence of a long-run relationship, the analysis focuses on the short-run dynamics between climate change variables and total fish trade in Egypt during the period from 1995 to 2022, as presented in

Table 18. The results of the ARDL model estimation for the short run dynamics under the specification ARDL (1, 2, 2, 2, 1, 2, 2).

Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	13,952.90	10,868.35	1.283811	0.2401
$\mathrm{T}\mathrm{O}\mathrm{T}\mathrm{A}\mathrm{L}\_\mathrm{T}\mathrm{R}\mathrm{A}\mathrm{D}\mathrm{E}\_\mathrm{V}\mathrm{O}\mathrm{L}\mathrm{U}\mathrm{M}\mathrm{E}(-1)$	−1.081959	0.290917	−3.719135	0.0075
$\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E}(-1)$	−345.6195	284.7080	−1.213943	0.2641
$\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_(-1)$	−174.2910	144.8625	−1.203148	0.2680
$\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}(-1)$	64.69307	22.05343	2.933469	0.0219
$\mathrm{C}\mathrm{A}\mathrm{R}\mathrm{B}\mathrm{O}\mathrm{N}\_\mathrm{G}\mathrm{A}\mathrm{S}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}(-1)$	−0.230035	0.110486	−2.082037	0.0759
$\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}(-1)$	9327.963	4243.838	2.198002	0.0639
$\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S}(-1)$	−779.5709	475.7147	−1.638736	0.1453
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E})$	−139.1895	103.6689	−1.342635	0.2213
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E}(-1\left)\right)$	153.0368	109.8415	1.393251	0.2062
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_)$	−198.6763	85.92834	−2.312116	0.0540
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_(-1\left)\right)$	104.4947	69.54541	1.502539	0.1767
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right)$	50.76111	15.56997	3.260193	0.0139
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right(-1\left)\right)$	−35.99933	15.41242	−2.335735	0.0522
$\mathrm{D}(\mathrm{C}\mathrm{A}\mathrm{R}\mathrm{B}\mathrm{O}\mathrm{N}\_\mathrm{G}\mathrm{A}\mathrm{S}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N})$	−0.392463	0.163210	−2.404650	0.0471
$\mathrm{D}(\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N})$	6340.281	2462.697	2.574527	0.0368
$\mathrm{D}(\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}(-1\left)\right)$	−6678.029	3791.051	−1.761524	0.1215
$\mathrm{D}(\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S})$	241.6155	364.5489	0.662779	0.5287
$\mathrm{D}(\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S}(-1\left)\right)$	1013.121	593.5218	1.706965	0.1316

Source: Calculated using EViews 12.

.

Short Run Interpretation

Change in Rainfall D(RAIN)

This variable has a positive and statistically significant effect at the five percent level, confirming the continued importance of rainfall in influencing fish trade in the short run.

Change in Relative Humidity D (RELATIVE_HUMIDITY)

This variable shows a negative and marginally significant effect at the five percent level, suggesting that sudden increases in humidity may adversely affect activities related to fish transportation or storage.

Change in Carbon Dioxide Emissions D (CARBON_GAS_EMISSION)

This variable has a negative and statistically significant effect at the five percent level, supporting the hypothesis that increased pollution leads to deterioration in fish production and constrains trade performance.

Change in Methane Emissions D (METHANE_EMISSION)

This variable exhibits a positive and statistically significant effect at the five percent level, indicating that increased productive activity in aquaculture systems, which may be associated with methane emissions, contributes to a temporary increase in total fish trade.

6.2.5. Diagnostic Tests for Model Validity

Normality and Heteroskedasticity Test

Table 19. Results of the Breusch–Pagan–Godfrey Test for Heteroskedasticity.

Test Statistic	Value	Probability Statistic	Prob.
F-statistic	0.377481	Prob. F (18,7)	0.9547
Obs × R-squared	12.80649	Prob. Chi-Square (18)	0.8029
Scaled explained SS	0.991599	Prob. Chi-Square (18)	1.0000

Null Hypothesis: Homoskedasticity. Source: Calculated using EViews 12.

indicates that all probability values exceed 0.05, suggesting that there is no statistically significant evidence of heteroskedasticity. Accordingly, the null hypothesis of homoskedasticity cannot be rejected, implying that the residuals have constant variance.

This result confirms that one of the fundamental assumptions of the econometric model is satisfied, thereby supporting the reliability and consistency of the estimated results.

Serial Correlation Test

Since the probability value of the F-statistic exceeds 0.05, the null hypothesis is not rejected at the five percent significance level. This indicates that there is no evidence of serial correlation among the residuals. These results confirm that the estimated model does not suffer from autocorrelation in the error terms, which supports the robustness and validity of the model.

Table 20. Results of the Breusch–Godfrey Serial Correlation LM Test.

Test Statistic	Value	Probability Statistic	Prob.
F-statistic	1.903119	Prob. F (2,5)	0.2429
Obs × R-squared	11.23774	Prob. Chi-Square (2)	0.0036

Null Hypothesis: No serial correlation. Source: Calculated using EViews 12.

.

After completing the previous steps, which constitute the foundation for conducting the subsequent analysis, the Error Correction Model (ECM) was estimated. The error correction coefficient is theoretically expected to lie between zero and negative one.

6.2.6. Economic Evaluation of the Cointegration Model Between the Total Value of Egyptian Fish Trade and Climate Change Variables

The findings confirm the existence of a long-run relationship between the total value of Egyptian fish trade and the explanatory variables represented by average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. This conclusion is supported by the estimated error correction coefficient, which reached (−1.081959). This indicates that approximately 108% of the short-run disequilibrium errors are corrected in the long run, confirming the existence of a cointegration relationship among the variables.

Table 21. Results of the Error Correction Model for cointegration.

ECM Regression
Variable	Coefficient	Std. Error	t-Statistic	Prob.
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E})$	−139.1895	45.45906	−3.061864	0.0183
$\mathrm{D}(\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E}(-1\left)\right)$	153.0368	39.43120	3.881109	0.0060
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_)$	−198.6763	35.91034	−5.532566	0.0009
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_(-1\left)\right)$	104.4947	26.49253	3.944309	0.0056
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right)$	50.76111	7.878010	6.443393	0.0004
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right(-1\left)\right)$	−35.99933	7.414866	−4.855020	0.0018
$\mathrm{D}(\mathrm{C}\mathrm{A}\mathrm{R}\mathrm{B}\mathrm{O}\mathrm{N}\_\mathrm{G}\mathrm{A}\mathrm{S}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N})$	−0.392463	0.088033	−4.458124	0.0029
$\mathrm{D}(\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N})$	6340.281	1450.324	4.371630	0.0033
$\mathrm{D}(\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}(-1\left)\right)$	−6678.029	1587.359	−4.207006	0.0040
$\mathrm{D}(\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S})$	241.6155	185.9986	1.299018	0.2351
$\mathrm{D}(\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S}(-1\left)\right)$	1013.121	244.4672	4.144200	0.0043
$\mathrm{C}\mathrm{o}\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{E}\mathrm{q}(-1)$	−1.081959	0.187734	−5.763265	0.0007
$\mathrm{R}—\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}$	0.778436	Mean dependent var.		−6.426923
$\mathrm{A}\mathrm{d}\mathrm{j}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{R}—\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}$	0.604349	S.D. dependent var.		100.1484
$\mathrm{S}.\mathrm{E}.\mathrm{o}\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}$	62.99415	Akaike info. criterion		11.42800
$\mathrm{S}\mathrm{u}\mathrm{m}\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{i}\mathrm{d}$	55,555.69	Schwarz criterion		12.00866
$\mathrm{L}\mathrm{o}\mathrm{g}\mathrm{l}\mathrm{i}\mathrm{k}\mathrm{e}\mathrm{l}\mathrm{i}\mathrm{h}\mathrm{o}\mathrm{o}\mathrm{d}$	−136.5640	Hannan–Quinn criter.		11.59521
$\mathrm{D}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{n}—\mathrm{W}\mathrm{a}\mathrm{t}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{t}$	2.526292

Source: Calculated using EViews 12.

presents the results of the Error Correction Model.

6.2.7. Statistical Evaluation of the Estimated Model

The statistical results confirm the significance of both the total value of the Egyptian fish trade and the error correction coefficient. In addition, the coefficient of determination indicates that 77.84% of the variation in the total value of Egyptian fish trade is explained by the independent variables included in the model, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. This finding is consistent with the economic evaluation of the model, while the remaining 22.16% is attributed to other factors not captured by the model. Furthermore, the estimated model was found to be statistically significant based on the value of the F-statistic.

6.2.8. Results of Estimating the Econometric Model for Measuring the Economic Effects of Climate Change and Their Impact on the Total Volume of Egyptian Fish Trade

Stability Tests of Model Coefficients (CUSUM and CUSUM of Squares)

To examine the structural stability of the estimated model coefficients and assess their consistency over time, the CUSUM and CUSUM of Squares tests were conducted. The figures indicate that the plots of both tests remained entirely within the critical bounds at the 5% significance level, confirming the absence of any structural breaks during the study period. These findings demonstrate that the estimated coefficients are stable and consistent over time, thereby supporting the reliability of the estimated model and the validity of the derived results for economic analysis and long-run interpretation. The results are illustrated in Figure 5.

7. Results of the Econometric Model Estimation for Measuring the Economic Effects of Climate Change on the Fish Trade Balance in Egypt

7.1. Estimation of the Cointegration Model Using the ARDL Approach

The Autoregressive Distributed Lag model was estimated using the fish trade balance in Egypt as the dependent variable, while climate change indicators were included as independent variables. These variables include average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. The ARDL model is applied to test for the existence of a long-run equilibrium relationship among the variables, that is, to determine whether cointegration exists.

7.2. Steps for Applying the ARDL Model

7.2.1. Determining the Stationarity of the Series

The Augmented Dickey–Fuller unit root test was conducted for all variables, including the fish trade balance as the dependent variable and the climate variables as independent variables. The results indicate that the time series are stationary at the level for the fish trade balance, average surface air temperature, relative humidity, and rainfall. The remaining variables become stationary after first differencing, as shown in

Table 22. The results of the unit root test for the stationarity of the time series.

Degree of Integration	1st Difference			Level
	none	trend & intercept	intercept	none	trend & intercept	intercept
I(0)	−7.60 **	−7.42 **	−7.48 **	−1.69 *	−2.79	−2.79 *	The value of the fish trade balance
I(0)	−10.82 **	−7.19 **	−7.19 **	0.89	−7.06 **	−4.56 *	Average surface air temperature
I(0)	−9.33 **	−6.01 **	−9.18 **	−0.49	−4.58 **	−3.86 **	Relative humidity%
I(0)	−4.58 **	−5.23 **	−4.64 **	0.91	−4.12 *	−3.74 **	Rain
I(1)	−4.90 **	−5.02 **	−4.89 **	0.09	−1.28	−1.69	Carbon gas emission
I(1)	−4.81 **	−4.88 **	−4.73 **	−0.31	−1.27	−1.6	Methane emission
I(1)	−4.78 **	−4.77 **	−4.69 **	0.56-	−1.39	−1.61	Emission of nitrous gas

Source: Calculated using EViews 12. Note: Critical values at the 1% significance level are −3.75 (Intercept), −4.44 (Trend and Intercept), and −2.67 (None). Critical values at the 5% significance level are −2.99 (Intercept), −3.63 (Trend and Intercept), and −1.96 (None). * Significant at the 5% level. ** Significant at the 1% level.

.

These findings confirm that the variables are integrated of different orders, which supports the suitability of applying the ARDL methodology.

7.2.2. Determination of the Optimal Lag Length

A Vector Autoregression model was estimated to determine the optimal lag length between the dependent variable and the independent variables. The results indicate that the optimal lag length for the model is two periods.

Table 23. The results of the lag length selection criteria.

Lag	LogL	LR	FPE	AIC	SC	HQ
0	−423.8974	NA	586,004.9	33.14596	33.48468	33.2435
1	−318.5765	145.829	8889.845	28.81358	31.52332	29.59388
2	−235.617	70.19649 *	1752.822 *	26.20131 *	31.28208 *	27.66439 *

Source: Calculated using EViews 12. * indicates lag order selected by the criterion. LR: sequential modified LR test statistic (each test at 5% level). FPE: Final prediction error. AIC: Akaike information criterion. SC: Schwarz information criterion. HQ: Hannan–Quinn information criterion.

.

7.2.3. ARDL Bounds Test

This step involves estimating the ARDL model to examine the presence of cointegration among the variables. The results presented in

Table 24. The results of the ARDL Bounds Test for cointegration.

F-Bounds Test		Null Hypothesis: No Level Relationship
Test Statistic	Value	Signif.	I(0)	I(1)
			Asymptotic: n = 1000
F-statistic	2.094225	10%	1.99	2.94
k	6	5%	2.27	3.28
		2.5%	2.55	3.61
		1%	2.88	3.99
Actual Sample Size	26		Finite Sample: n = 35
		10%	2.254	3.388
		5%	2.685	3.96
		1%	3.713	5.326
			Finite Sample: n = 30
		10%	2.334	3.515
		5%	2.794	4.148
		1%	3.976	5.691

Source: Calculated using EViews 12.

indicate that the calculated F-statistic is 2.094, which is lower than the upper critical bound at the one percent significance level. Accordingly, the null hypothesis of no cointegration is accepted.

This finding indicates that there is no long-run equilibrium relationship between the fish trade balance in Egypt and the climate change variables, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. In other words, no cointegration exists among these variables at the one percent significance level.

Therefore, the analysis is limited to short-run dynamics, as the variables do not move together in the long run. Based on this result, the ARDL model can be interpreted within a short-run framework without including an error correction term.

7.2.4. Estimation of the ARDL Model and Short Run Interpretation

Given the absence of a long-run relationship, the analysis focuses on the short-run dynamics between climate change variables and the Egyptian fish trade balance during the period from 1995 to 2022, as presented in

Table 25. Presents the results of the ARDL model estimation under the specification ARDL (1, 0, 1, 2, 0, 0, 0).

Variable	Coefficient	Std. Error	t-Statistic	Prob.
C	−11.52143	2312.495	−0.004982	0.9961
$\mathrm{T}\mathrm{H}\mathrm{E}\_\mathrm{V}\mathrm{A}\mathrm{L}\mathrm{U}\mathrm{E}\_\mathrm{O}\mathrm{F}\_\mathrm{T}\mathrm{H}\mathrm{E}\_\mathrm{F}\mathrm{I}\mathrm{S}\mathrm{H}\_\mathrm{T}\mathrm{R}\mathrm{A}\mathrm{D}\mathrm{E}\_\mathrm{B}\mathrm{A}\mathrm{L}\mathrm{A}\mathrm{N}\mathrm{C}\mathrm{E}(-1)$	−0.966719	0.248582	−3.888935	0.0015
$\mathrm{A}\mathrm{V}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{G}\mathrm{E}\_\mathrm{S}\mathrm{U}\mathrm{R}\mathrm{F}\mathrm{A}\mathrm{C}\mathrm{E}\_\mathrm{A}\mathrm{I}\mathrm{R}\_\mathrm{T}\mathrm{E}\mathrm{M}\mathrm{P}\mathrm{E}\mathrm{R}\mathrm{A}\mathrm{T}\mathrm{U}\mathrm{R}\mathrm{E}$	57.92532	53.10987	1.090670	0.2926
$\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_(-1)$	−23.29689	43.37617	−0.537090	0.5991
$\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}(-1)$	−34.20074	13.14704	−2.601402	0.0200
$\mathrm{C}\mathrm{A}\mathrm{R}\mathrm{B}\mathrm{O}\mathrm{N}\_\mathrm{G}\mathrm{A}\mathrm{S}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}$	0.101612	0.050313	2.019607	0.0617
$\mathrm{M}\mathrm{E}\mathrm{T}\mathrm{H}\mathrm{A}\mathrm{N}\mathrm{E}\_\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}$	−3974.988	1574.812	−2.524103	0.0234
$\mathrm{E}\mathrm{M}\mathrm{I}\mathrm{S}\mathrm{S}\mathrm{I}\mathrm{O}\mathrm{N}\_\mathrm{O}\mathrm{F}\_\mathrm{N}\mathrm{I}\mathrm{T}\mathrm{R}\mathrm{O}\mathrm{U}\mathrm{S}\_\mathrm{G}\mathrm{A}\mathrm{S}$	300.1074	162.6508	1.845103	0.0849
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_)$	39.41202	34.79914	1.132557	0.2752
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right)$	−22.40895	8.391256	−2.670512	0.0175
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right(-1\left)\right)$	11.78292	6.399114	1.841336	0.0854

Source: Calculated using EViews 12.

.

First: Short Run Interpretation

Current Change in Rainfall D (RAIN)

This variable has a negative and statistically significant effect at the 0.0175 level, indicating that sudden increases in rainfall may lead to temporary disruptions in production, transportation, or export activities. This, in turn, negatively affects the fish trade balance in the short run.

Lagged Change in Rainfall D (RAIN(−1))

This variable shows a positive but weakly significant effect at the 0.0854 level. This suggests that improved rainfall conditions in previous periods may subsequently enhance trade performance by supporting higher productivity or stabilizing fish supply.

Other Short Run Variables (Temperature, Humidity, Gas Emissions)

These variables do not exhibit statistically significant effects in the short run. This indicates that immediate climatic fluctuations do not directly translate into changes in the trade balance, but rather require time to influence production and trade dynamics.

Second: General Conclusion

The results suggest that carbon and nitrous oxide emissions may exhibit positive associations; however, this relationship likely reflects short-term economic activity rather than genuine improvements in trade efficiency or environmental performance. Overall, the findings confirm that the relationship between climate change and fish trade in Egypt is complex and multidimensional, influenced by a combination of environmental, production, and policy factors. This highlights the importance of integrating environmental considerations into fisheries and trade strategies to ensure sustainability and maintain competitiveness.

7.2.5. Estimation of the Error Correction Model According to the ARDL Approach

After completing the previous steps, which constitute the basis for conducting the subsequent analysis, the Error Correction Model (ECM) was estimated. The error correction coefficient is theoretically expected to take a value between zero and negative one.

Economic Evaluation of the Cointegration Model Between the Egyptian Fish Trade Balance and Climate Change Variables

The results confirm the existence of a long-run relationship between the Egyptian fish trade balance and the explanatory variables represented by average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. This conclusion is supported by the estimated error correction coefficient, which reached (−0.966719). This indicates that approximately 96% of the short-run disequilibrium errors are corrected in the long run, confirming the existence of a cointegration relationship among the variables.

Table 26. Results of the Error Correction Model for cointegration.

ECM Regression
Variable	Coefficient	Std. Error	t-Statistic	Prob.
$\mathrm{D}(\mathrm{R}\mathrm{E}\mathrm{L}\mathrm{A}\mathrm{T}\mathrm{I}\mathrm{V}\mathrm{E}\_\mathrm{H}\mathrm{U}\mathrm{M}\mathrm{I}\mathrm{D}\mathrm{I}\mathrm{T}\mathrm{Y}\_\_)$	39.41202	17.91860	2.199503	0.0439
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right)$	−22.40895	4.892279	−4.580473	0.0004
$\mathrm{D}\left(\mathrm{R}\mathrm{A}\mathrm{I}\mathrm{N}\right(-1\left)\right)$	11.78292	4.033687	2.921129	0.0105
$\mathrm{C}\mathrm{o}\mathrm{i}\mathrm{n}\mathrm{t}\mathrm{E}\mathrm{q}(-1)$	−0.966719	0.195019	−4.957039	0.0002
$\mathrm{R}—\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}$	0.566631	Mean dependent var.		6.950000
$\mathrm{A}\mathrm{d}\mathrm{j}\mathrm{u}\mathrm{s}\mathrm{t}\mathrm{e}\mathrm{d}\mathrm{R}—\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}$	0.507535	S.D. dependent var.		95.13561
$\mathrm{S}.\mathrm{E}.\mathrm{o}\mathrm{f}\mathrm{r}\mathrm{e}\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}$	66.76224	Akaike info. criterion		11.38079
$\mathrm{S}\mathrm{u}\mathrm{m}\mathrm{s}\mathrm{q}\mathrm{u}\mathrm{a}\mathrm{r}\mathrm{e}\mathrm{d}\mathrm{r}\mathrm{e}\mathrm{s}\mathrm{i}\mathrm{d}$	98058.32	Schwarz criterion		11.57434
$\mathrm{L}\mathrm{o}\mathrm{g}\mathrm{l}\mathrm{i}\mathrm{k}\mathrm{e}\mathrm{l}\mathrm{i}\mathrm{h}\mathrm{o}\mathrm{o}\mathrm{d}$	−143.9503	Hannan–Quinn criter.		11.43653
$\mathrm{D}\mathrm{u}\mathrm{r}\mathrm{b}\mathrm{i}\mathrm{n}—\mathrm{W}\mathrm{a}\mathrm{t}\mathrm{s}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{t}$	2.245382

Source: Calculated using EViews 12.

presents the results of the Error Correction Model.

Statistical Evaluation of the Estimated Model

The statistical results confirm the significance of both the Egyptian fish trade balance and the error correction coefficient. In addition, the coefficient of determination indicates that 56.66% of the variation in the Egyptian fish trade balance is explained by the independent variables included in the model, namely average surface air temperature, relative humidity, rainfall, carbon dioxide emissions, methane emissions, and nitrous oxide emissions. This finding is consistent with the economic evaluation of the model, while the remaining 43.34% is attributed to other factors not captured by the model. Furthermore, the estimated model was found to be statistically significant based on the value of the F-statistic.

7.2.6. Diagnostic Tests for Model Validity

Normality and Heteroskedasticity Test

Table 27. Results of the Breusch–Pagan–Godfrey Test for Heteroskedasticity.

Test Statistic	Value	Probability Statistic	Prob.
F-statistic	1.118326	Prob. F (10,15)	0.4093
Obs × R-squared	11.10499	Prob. Chi-Square (10)	0.3494
Scaled explained SS	8.950120	Prob. Chi-Square (10)	0.5368

Null Hypothesis: Homoskedasticity. Source: Calculated using EViews 12.

indicates that all probability values exceed 0.05, suggesting that there is no statistically significant evidence of heteroskedasticity. Accordingly, the null hypothesis of homoskedasticity cannot be rejected, implying that the residuals exhibit constant variance.

This result confirms that one of the fundamental assumptions of the econometric model is satisfied, thereby supporting the reliability and consistency of the estimated results.

Serial Correlation Test

Since the probability value of the F-statistic exceeds 0.05, the null hypothesis is not rejected at the five percent significance level. This indicates that there is no evidence of serial correlation among the residuals.

These results confirm that the estimated model does not suffer from autocorrelation in the error terms, thereby supporting the robustness and validity of the model.

Table 28. Results of the Breusch–Godfrey Serial Correlation LM Test.

Test Statistic	Value	Probability Statistic	Prob.
F-statistic	1.482611	Prob. F (2,13)	0.2630
Obs × R-squared	4.828982	Prob. Chi-Square (2)	0.0894

Null Hypothesis: No serial correlation. Source: Calculated using EViews 12.

.

7.2.7. Results of Estimating the Econometric Model for Measuring the Economic Effects of Climate Change and Their Impact on the Egyptian Fish Trade Balance

Stability Tests of Model Coefficients (CUSUM and CUSUM of Squares)

To examine the structural stability of the estimated model coefficients and assess their consistency over time, the CUSUM and CUSUM of Squares tests were conducted. The figures indicate that the plots of both tests remained entirely within the critical bounds at the 5% significance level, confirming the absence of any structural breaks during the study period. These findings demonstrate that the estimated coefficients are stable and consistent over time, thereby supporting the reliability of the estimated model and the validity of the derived results for economic analysis and long-run interpretation. The results are illustrated in Figure 6.

8. Discussion

The findings of the current study provide important empirical evidence regarding the relationship between climate change and the performance of Egypt’s fish foreign trade during the period of 1995–2022. The results generally indicate that climate variables affect fish trade performance both directly and indirectly through their influence on environmental quality, fisheries productivity, production costs, and market competitiveness. Overall, the estimated ARDL models reveal that climate impacts on Egypt’s fish trade sector are more pronounced in the short run, while limited evidence of long-run equilibrium relationships was identified, except in the fish export model.

The long-run estimation results for fish exports demonstrate that nitrous oxide emissions exert a statistically significant negative effect on export performance. This finding is consistent with previous studies emphasizing the adverse effects of environmental degradation and greenhouse gas emissions on fisheries productivity and trade competitiveness [19,23,24]. Higher greenhouse gas emissions contribute to water quality deterioration, ecological stress, and reduced sustainability of aquatic ecosystems, which may weaken the productive and export capacities of fisheries sectors over time. Similar conclusions were reported in Mediterranean and developing-country studies, which found that climate-related environmental stress negatively affects fisheries sustainability and export performance [26,27].

The results also indicate that rainfall represents the most influential short-run climatic variable affecting fish exports, fish imports, and total fish trade volume. The positive short-run effect of rainfall on exports may reflect improvements in environmental and hydrological conditions that support fisheries productivity and aquaculture activities. This finding is consistent with studies suggesting that favorable climatic conditions and improved water availability may enhance fish production and market supply in the short term [25,31]. At the same time, the short-run fluctuations observed in imports and trade balance indicators confirm that climate variability may generate temporary disturbances in domestic supply conditions, thereby affecting import demand and trade performance.

Another important finding concerns the absence of cointegration relationships in the fish imports, total trade volume, and fish trade balance models. This result suggests that climate change does not exert a stable long-run equilibrium effect on these trade indicators, but rather influences them through temporary and dynamic adjustments associated with production conditions, domestic market fluctuations, transportation costs, and short-term supply constraints. This interpretation partially agrees with previous studies indicating that climate variability affects food imports and trade systems indirectly through its effects on production efficiency, food availability, and market stability [23,26]. The findings also support the argument that developing countries remain highly vulnerable to climate-induced trade fluctuations due to their dependence on climate-sensitive production systems and limited adaptive capacities [12,27].

The results further highlight the particular importance of Egypt as a case study within the climate change and fisheries literature. Egypt’s fisheries sector depends heavily on aquaculture systems that are highly sensitive to water quality, temperature changes, salinity levels, and environmental degradation. At the same time, the fish trade has become increasingly important for food security, export diversification, and narrowing the domestic food gap. Consequently, climate-related environmental pressures may simultaneously affect both production systems and trade competitiveness. This finding aligns with previous studies emphasizing the vulnerability of Mediterranean and Nile Delta fisheries systems to climate-related environmental risks, including sea-level rise, warming temperatures, and water resource pressures [28,29,30,31,32].

From a sustainability perspective, the results indicate that the gradual improvement observed in fish exports and export coverage ratios may reflect the adaptive capacity of Egypt’s fisheries and aquaculture sector despite increasing climatic pressures. The expansion of aquaculture production, improvements in production technologies, and the development of export-oriented systems may have partially mitigated some negative climate effects. However, the continuing increase in greenhouse gas emissions and environmental pressures remains a significant long-term challenge for fisheries sustainability and trade resilience. These findings support the growing international literature emphasizing that climate change should not be viewed solely as an environmental issue, but also as a major economic and trade-related challenge affecting food security, resource sustainability, and external sector performance in developing economies.

9. Conclusions

This study examined the impact of climate change on the performance of Egypt’s fish foreign trade during the period of 1995–2022 using the Autoregressive Distributed Lag (ARDL) approach. The analysis incorporated major climate indicators, including temperature, rainfall, relative humidity, and greenhouse gas emissions, together with fish trade indicators represented by exports, imports, total trade volume, trade balance, and export coverage ratio.

The empirical findings indicate that climate change exerts heterogeneous effects on the different dimensions of Egypt’s fish foreign trade. The results confirmed the existence of a long-run equilibrium relationship only in the fish exports model, suggesting that export performance is structurally associated with climate and environmental conditions over time. In contrast, no long-run cointegration relationship was identified for fish imports, total trade volume, or fish trade balance, indicating that these indicators are mainly influenced by short-run market dynamics and temporary climatic fluctuations.

At the short-run level, rainfall emerged as the most influential climatic variable affecting fish trade indicators, while nitrous oxide emissions showed a statistically significant negative long-run effect on fish exports. These findings suggest that climate change affects fish trade indirectly through its influence on fisheries productivity, environmental quality, production efficiency, and trade competitiveness. The results also demonstrate that the fisheries sector in Egypt remains highly sensitive to environmental and climatic pressures despite the gradual improvements observed in export performance and coverage ratios during the study period.

The study contributes to the growing literature on climate change and fisheries economics by extending the analysis beyond fisheries production to examine fish foreign trade performance within an integrated econometric framework. The findings provide empirical evidence on how climate variables influence exports, imports, trade balance performance, and trade sustainability in the context of a developing economy highly dependent on climate-sensitive production systems. From a policy perspective, the findings highlight the importance of integrating climate adaptation and environmental sustainability into fisheries and trade policies. Strengthening climate-resilient aquaculture systems, improving water-use efficiency, reducing environmentally harmful emissions, and enhancing trade infrastructure and export competitiveness are essential for improving the long-run sustainability and resilience of Egypt’s fisheries sector under changing climatic conditions.

Despite its contributions, the study is subject to several limitations. The analysis relied on annual time-series data, which may limit the ability to capture short-term climatic fluctuations and seasonal variations. In addition, the estimated models focused primarily on climate indicators and did not incorporate some institutional, technological, and international market variables that may also influence fish trade performance.

Future research may extend the current analysis by employing panel data for multiple countries, incorporating institutional and governance variables, and applying alternative econometric approaches capable of capturing nonlinear and asymmetric climate effects on fisheries trade. Further studies may also examine the implications of climate change for specific fisheries subsectors and regional export markets.

10. Recommendations

Achieving sustainable fish foreign trade in Egypt requires a structural transition toward climate-resilient and resource-efficient systems supported by coherent policies that integrate environmental, economic, and trade dimensions. Based on the empirical findings and analytical results of this study, the following recommendations are proposed:

• The significant negative long-run effect of nitrous oxide emissions on fish exports indicates the need to reduce environmentally harmful emissions within fisheries and aquaculture activities through cleaner production technologies, improved waste management systems, and environmentally sustainable aquaculture practices in order to protect export competitiveness and long-run trade sustainability.
• Since rainfall was identified as the most influential short-run climatic variable affecting fish exports, imports, and total trade performance, fisheries policies should strengthen climate adaptation measures related to water management, early warning systems, and climate-resilient aquaculture systems to reduce the vulnerability of fish production and trade to short-term climatic fluctuations.
• The absence of long-run cointegration relationships for fish imports, total trade volume, and fish trade balance suggests that these indicators are highly sensitive to temporary market and climatic shocks. Accordingly, improving domestic production efficiency, storage systems, transportation infrastructure, and cold-chain logistics is essential to enhance market stability and reduce short-run trade disruptions.
• The positive trends observed in fish exports and export coverage ratios during the study period highlight the adaptive potential of Egypt’s fisheries sector. Therefore, policymakers should support export-oriented aquaculture development, improve compliance with international quality standards, and promote sustainable fisheries investments to strengthen long-run trade resilience and sustainability under changing climatic conditions.