Farm mechanization is necessary to increase agricultural productivity and ensure food security for the global population (Emami et al., 2018; Liu et al., 2025). Farmers in developing countries depend heavily on manual labor to operate within the agricultural sector, including harvesting activities (Fathallah, 2010). Farmers develop musculoskeletal conditions primarily because their work requires extended positions that demand awkward body postures and force them to lift heavy weight (Benos et al., 2020). Agricultural mechanization decreases dependence on human labor, while improving agricultural food quality, operation speed, and productivity of land resources (Abate and Kuang, 2021). Agricultural crop production benefits from mechanization because it boosts labor capabilities to sustain food supplies and decreases poverty in developing markets (Sims et al., 2016). Farm mechanization represents agricultural growth through the use of modern equipment and machinery (Amponsah et al., 2012). Rising labor wages combined with government subsidies have triggered a substantial increase in mechanization levels in agriculture (Yang et al., 2013).

Farm mechanization creates stronger resistance for farming households against different types of hazards and improves agricultural productivity, resulting in enhanced food security in nations. It also decreases farmers’ health risks by reducing physical farming activities (Olasehinde-Williams et al., 2020). It eliminates dangerous situations that occur during manual agriculture operations while creating opportunities for additional employment outside farming (Richards and Ramezani, 1990). Fram mechanization affects all four aspects of food security (food availability, economic and physical access to food, food utilization, and stability over time) (FAO, 2021a, 2021b). Through mechanization, there is enhanced protection against food loss during harvest periods with improved food safety standards (Daum, 2023). Sustainable agricultural mechanization leads to food independence along with economic growth, which includes all segments of society and helps advance sustainable agricultural primary objectives (Mrema, 2015). The wages farmers receive from mechanization improve access to better living conditions, while the reduced costs of production benefit families who purchase food in rural and urban areas. In such situations, food costs claim between 50 and 70% of household income (Liu et al., 2025). They are unable to obtain nutritious diets because financial limitations restrict their means. Fram mechanization eases the physical labor associated with manual agriculture, thus reducing the amount of energy required to limit caloric deficiencies and improve food security (Liu et al., 2025).

Interest in agricultural land is increasing because of the growing population, inflated food prices, and food security concerns (Molotoks et al., 2021). Agricultural food supplies constitute the largest food source with essential ecosystem services. Several countries suffer from food insecurity owing to insufficient agricultural output caused by declining agricultural land per capita (Alami et al., 2021). Food security depends heavily on farm structures because of the accelerating climate change and increasing human population. Farming structures that utilize traditional farming approaches and rain-fed systems encounter higher exposure to climatic problems, leading to dangerous consequences for food production (FAO, 2021a, 2021b). Sustainable farming practices that unite conservation agriculture with agroecological systems have successfully protected smallholder farms through improved yield production and reduced environmental damage (Pretty et al., 2018; Tittonell and Giller, 2013). The adoption of sustainable farm practices is affected by multiple barriers including farm structure issues (small landholding, labor force deficiencies, financial constraints, etc.) (Barrett et al., 2017). Smallholder farmers can significantly boost food security through sustainable farming systems (Reardon et al., 2019).

Labor dynamics functions as a fundamental farm structure element that shapes the levels of food security in farms. Labor shortages emerge in regions where rural populations age together with high migration to urban areas, thus creating food security risks and decreasing farming productivity (Christiaensen and Todo, 2014). The implementation of farm technologies is hindered for smallholder farmers because they face financial challenges and lack necessary expertise (Fuglie et al., 2019). Cooperative farming systems and land reorganization programs show major potential in handling agricultural productivity and food security issues by improving access to resources and operational efficiency (Li et al., 2020). Farm structures play a pivotal role in agricultural sustainability and in ensuring global food security. Multiple crops under smallholder farms create diverse ecosystems that improve soil quality and protect the environment and food security yet need proper support from public institutions (Pretty et al., 2018). Farm structure has emerged as a major determining factor for food security because it directly affects productivity levels and sustainability alongside resilience factors. Public policies are required to ensure equal land access and technology for smallholder farmers to address different structural and sustainability issues (Touch et al., 2024). The world faces enormous challenges, such as climate change and increasing human population. Thus, farm structures and adopting sustainable farm practices are key to guaranteeing food security (Rehman et al., 2022).

Recent studies have demonstrated that agricultural value addition is a crucial element in boosting food security because of economic instability and changing climate patterns (Dos-Santos, 2020; Wijerathna-Yapa and Pathirana, 2022). The process of converting raw agricultural commodities into valuable products demonstrates substantial effectiveness in food availability through a reduction in postharvest losses. Paradigmatic techniques such as solar drying and cold storage systems improve the storage duration of fruits and vegetables. Post-harvest losses in the agricultural sector require immediate attention because they represent a major threat to food sustainability (Sheahan and Barrett, 2017; Kumar and Kalita, 2017). Agricultural value addition transforms crops into refined products, including flour, oils, and dried goods, which improves food choices. In addition, it decreases monoculture dependence, thus extending food availability throughout different seasons (Dahal et al., 2024).

Value addition to agricultural products positively affects accessibility, because it enables farmers to reach profitable high-value markets. Smallholder farmers convert milk into yogurt and cheese, thus providing sustainable income and better food availability for their households (Touch et al., 2024). The development of food value addition enables the establishment of important facilities for processing and storage, alongside strengthened market links. This will enable faster and fresh healthy delivery of food to consumers (Fagbemi and Oluwajuyitan, 2020). Agricultural value addition is fundamental for increasing food utilization because it enhances the dietary nutritional content of food products (Dos-Santos, 2020). Staple foods fortified with iron and biofortified with vitamin A are successfully reduce micronutrient deficiencies in developing nations (Hombali et al., 2019). Traditional food processing techniques, such as fermentation and germination, eliminate anti-nutritional components that make food more digestible and nutritious (Nkhata et al., 2018). Recent advancements have played a critical role in addressing malnutrition and food insecurity in many areas. Agricultural value addition establishes food systems that can survive unexpected incidents, including climate change and economic downturns, thus ensuring food stability. Regions exhibiting frequent climate disruptions require agricultural value addition to be used as food shortage shields against such events.

The gender dynamics in farming influence how successfully agricultural resources are used, and how food systems maintain their stability (Bryan et al., 2023). In addition, they affect resource sharing along with agricultural productivity levels. The worldwide female share of agricultural workers reaches 43%, while Africa and Asia have population percentages above 50% (FAO, 2011). Female farmers make important agricultural contributions; however, they face numerous systemic barriers. They lack access to land ownership together with basic farm resources, including credit and educational services, technology, and appropriate extension services. This inequality impairs production capability and at the same time strengthens the persistence of poverty and food insecurity. When female farmers obtain equal access to agricultural resources like those given to male farmers, agricultural output would rise by 20–30% thereby lifting millions out of food insecurity and malnutrition (FAO, 2011).

Agricultural food security initiatives face additional challenges because women perform different workload tasks from men in farming. Women handle core farming tasks, food processing work, and nutritional duties, while basic leadership functions and cash crop management responsibilities mainly fall under male control (Visser and Wangu, 2021). The segregated labor system results in an unequal distribution of resources and priorities, leading to undervalued food-producing activities that cannot sustain household nutrition (Quisumbing et al., 2014). The heavy unpaid care burden women carry diminishes their ability to join productive agricultural work, which ultimately leads to increased food insecurity at home. The difficulties faced by agricultural female farmers become substantially more severe because women tend to be more exposed to the disastrous impacts of climate change (Dibakoane et al., 2022). This is primarily because women have limited access to adaptive technologies and essential resources. Areas affected by prolonged droughts and erratic rainfall patterns make female farmers vulnerable to crop failures because they depend on traditional farming practices with rain-fed agriculture (UN Women, 2022). Establishing food systems resistant to environmental disturbances requires specific attention to gender-based risks. Women’s support in agricultural leadership creates widespread social and economic advantages that are selected simply by obtaining safe food supplies (Bryan et al., 2023). Women who achieve financial independence through resource control tend to dedicate their earned money to enhancing their families’ health levels and nutritional requirements. The positive effects generated from this scenario strengthen the overall food security of communities (World Bank, 2014). Achieving sustainable food security depends heavily on gender equality-focused policies and interventions that include land distribution reforms and improved access to credit and gender-focused extension services (Quisumbing and Doss, 2021). Tackling gender differences in agricultural production leads to enhanced food security. Raising women’s agricultural participation rates will enable the world to exploit female farmers’ complete potential, boost agricultural yields, and establish stronger food systems (Bryan et al., 2023).

The relationship between resource management and food security exists because the sustainable control of essential natural resources, including water, soil, and energy, directly affects operational food production stability (Pinstrup-Andersen and Pandya-Lorch, 1998). Inadequate resource management creates greater food insecurity, but sustainable techniques lead to substantial improvements in food security (Mekonnen et al., 2024). Water is the principal factor in agriculture because it accounts for 70% of the total global freshwater usage. Water scarcity develops when improper extraction methods and inefficient irrigation techniques are combined to deplete available water reserves, thereby reducing both agricultural harvest output and national food yields (Ingrao et al., 2023).

Agricultural production receives vital support from sustainable water management practices including drip irrigation and rainwater techniques, particularly in water-scarce areas (Chartzoulakis and Bertaki, 2015; Penuelas et al., 2023). The implemented methods lead to better agricultural sector productivity, while strengthening food security (Barrett, 2021). Food security is in danger due to soil degradation resulting from overfertilization practices and failing land management and deforestation operations, which decreases cropland fertility (Pozza and Field, 2020). Soil health recovery through sustainable practice, including crop rotation with conservation tillage techniques, improves food production. Soil conservation practices integrated into farming systems offer great potential for building climate change resilience, which helps sustain better food security levels in developing nations (Şimşek, 2025; Sher et al., 2024).

The entire food system requires substantial energy consumption to maintain food security because agricultural production to food distribution phases is highly energy-demanding (Corigliano and Algieri, 2024; Menconi et al., 2022). Renewable and efficient energy strategies can reduce the environmental effects of food systems and maintain reliable food availability (Corigliano and Algieri, 2024). Climate-resilient practices represent an invaluable method for enhancing food security throughout regions where climate change has the greatest impact (FAO, 2020). Sustainable food production depends heavily on biodiversity because natural processes such as pollination, pest management and nutrient cycling help farmers secure food availability. Excessive use of natural resources, particularly deforestation and destructive fishing, increases the risk of essential food security services (Shahbaz et al., 2022a, 2022b). The key to maintaining sustainable food security throughout the years depends on effective measures for biodiversity preservation and ecological system management (Varzakas and Smaoui, 2024). The adoption of resource-friendly operations improves food safety outcomes (Vågsholm et al., 2020; Shahbaz et al., 2022a, 2022b).

Yearly data were used in this study covering the period from 1991 to 2019. The data on variables were collected from two sources, as our world in data were accessed to obtain data on the countries’ calorie supply, mechanization, farm structure, resource optimization, and male and female participation in agriculture, which is sourced by the Food and Agriculture Organization (FAO). The data regarding agricultural value addition (percent of GDP) and population density (people per square kilometer of land area) were obtained from World Development Indicators (WDI). The indicators to measure farm structure include three indicators: labor force share in agriculture (%), agricultural land per capita, and arable land needed for producing crops (index). Total factor productivity (TFP) is used as a proxy for resource optimization and the data were analyzed using STATA 21. Table 1 describes the study variables.

By including the above variables, the following functional form was specified, which includes calorie supply over the years as the dependent variable, and the remaining seven variables were considered as independent variables.

where Cal, FS, ROP, VAA, PD, MEC, MA, and FA indicate calorie supply, farm structure, resource optimization, agricultural value addition, population density, mechanization, males in agriculture, and females in agriculture, respectively. β shows the unknown coefficients to be measured, and ε depicts the error term.

The autoregressive distributed lag (ARDL) cointegration approach does not require unit root pre-testing because it can be applied to a set of variables of order I(0), I(1), or mixed. The ARDL bounds test requires all variables to be integrated of order 0 or 1, and no variable is integrated of order I(2) (Pesaran and Shin, 1999; Pesaran et al., 2001). Therefore, to justify the integrated order of all variables, two different unit root tests were applied: augmented Dickey-Fuller (ADF) and Phillips-Perron (PP).

The ARDL bound test-based F-statistics were used to determine the existence of long-run relationships among the variables. The ARDL approach has many advantages over the traditional cointegration approach, as the ARDL approach is flexible and allows analysis with an integrated order of 0 or 1. Its application and interpretation are very simple owing to its single equation set as well as different lag lengths for different variables, and it is very suitable for small sample sizes. It also provides unbiased estimates for long-run relationships and adequately addresses autocorrelation and endogeneity (Jalil and Ma, 2008; Harris and Sollis, 2005).

The null hypothesis (H₀ = no cointegration) would be rejected if the F-statistic value exceeds the upper bound value, and accepted if the F-statistic value remains lower than the lower bound value, while if the F-statistic value falls between the lower bound and upper bound, then it is unclear whether cointegration exists (Pesaran et al., 2001). Therefore, the following cointegration model is specified by exploring the long-run relationships among the variables.

Where, △ indicates the change operation, and is the optimal lag number based on the lag length criterion depicted by t−1, and ∂ 0 to ∂ 7 and β 1 to β 7 are the elements to be examined. From the above equation, the following hypothesis was analyzed to determine the cointegration among the variables.

If the f-statistics value exceeds the upper bound, then H0 is rejected, and cointegration among variables exists; then, the short-and long-run ARDL models are examined.

where ECT is the error correction term that indicates the speed of adjustment from the short-run shock to the long-run equilibrium. It must be negative and significant, with a value of less than 1.

The most important assumption of the ARDL bound test is that the error term is serially independent and normally distributed. For this purpose, several tests were employed to analyze autocorrelation, heteroskedasticity, and normality. For autocorrelation, the Breusch–Godfrey LM test and Durbin–Watson d-statistic for heteroskedasticity, White’s test, and Breusch-Pagan-Godfrey test were used. For normality, the Jarque-Bera test was used.

Thus, it is necessary to confirm the stability of a model with an autoregressive structure. For this purpose, the cumulative sum of recursive residuals (CUSUM) and cumulative sum of recursive residuals of squares (CUSUMQ) (Brown et al., 1975) tests were used. A similar method was proposed by Pesaran and Pesaran (1997) to determine the parameter stability. Figure 1 shows the methodology used in this paper.

Table 2 presents the outcomes of two different unit root tests: ADF and PP. The first stationarity level of variables was tested by considering only constant, and then the trend option was added with a constant term to check the robust outcomes regarding stationarity of variables. The findings revealed that both ADF and Phillips Peron confirmed that all the variables were stationary at the first difference; hence, the order of integration for all variables is I(1). For the ARDL bound testing proposed by Pesaran and Shin (1999) and Pesaran et al. (2001), both ADF and PP unit root tests also confirmed that there is no variable of order I(2).

Determining the optimal lag length is an empirical issue, and there is no rule of thumb for determining the optimal lags for the variables. The selection of too few lags can cause a specification error, whereas too many lags cause the problem of multicollinearity. Moreover, the sample size and selection criteria significantly affect the selection of optimal lags. For annual data, usually one or two lags, for quarterly data, a maximum of eight lags for monthly data up to 24 lags is suitable (Godil et al., 2022). There are five different types of selection criteria to determine the optimal lags for each variable: likelihood ratio, Akaike information criteria, final prediction error, Schwartz information criteria, and Hannah-Quinn information criteria. Based on these two conditions of sample size and selection criteria, we first focus on the selection criteria. Based on the Akaike Information Criterion, we found the optimal lags for each variable except for males in agriculture, which had an optimal lag of 4. Then, considering the sample size, we assumed the optimal lag for this variable to be equal to 2, and ARDL (2,2,2,2,2,2,2,2) was specified.

The above specified model has passed all the diagnostic tests as the R square is 0.97, and the adjusted R square is 0.94, which implies that the variables explain 97% of the variation in the Cals, which indicates the high explanatory power of the model. Autocorrelation tests, such as the Durbin-Watson and Breusch-Godfrey LM tests, confirm that there is no autocorrelation, and the model is not spurious. Similarly, the F-statistics (=252.22 at p-value = 0.00) also indicate that the null hypothesis is rejected, which shows that the variables have non zero coefficients. Table 3 indicates that the model has also passed all other diagnostic tests regarding heteroskedasticity and Jarque-Berra normality.

ARDL is a model specifically used for testing long-run relationships among variables. The ARDL bound test developed by Pesaran et al. (2001) identifies the existence of cointegration or long-run relationships among the variables. Table 4 presents the ARDL bounds test’ outcomes the F-value greater than the upper bound indicates the long-run association do exist between the variables.

Table 5 presents both long-and short-run relationships between the dependent and independent variables. The findings indicate that the coefficients of FS, PD, VAA, MA, and FA are statistically significant. This implies that these variables strongly affect Cals in the long run, leading to higher food security in the country. These variables are crucial for improving food security by increasing Cals in the long run and improving farm productivity, distribution, and accessibility to food items. The significantly negative value of the lagged error correction term (ECT) highlights how quickly the system returns to equilibrium after a shock in the short run. The negative coefficient of ECT −0.503 indicates that the long-run relationship among the variables is stable, and 50.3% of the deviation from the previous period’s equilibrium in the short run is correctly backed to its equilibrium in the current period.

FS often indicates the size of the farm, which plays a crucial role in determining the farm scale and farm productivity. As the findings indicates 1 unit increase in FS, caloric availability in the country will increase by almost 673 kcal per day per capita. As the FS improves, the adoption of sustainable and efficient farm practices also increases. The adoption of new farm technologies makes farms more productive and profitable through stable food production and supply, which ensures a stable caloric supply for the growing population. Therefore, changes in FS led to the integration of innovations in cropping systems, which greatly affects farmers’ incomes and nutritional outcomes. For example, an efficient cropping system can increase soil fertility and improve the Cals for farm families (Prusty et al., 2022). The author described that adjustments in existing crop systems at farms and the adoption of a new efficient FS can improve farm productivity, performance, and sustainability. On the other hand, FS also positively and significantly influences Cals in the short term. The low and significant positive coefficient of FS (first difference = D1) indicates that FS can enhance calorie supply through immediate changes in farm size, land use, and cropping pattern. Similarly, the significant and large positive coefficient of FS (lagged difference = LD) signifies that past adjustments in the FS of country have a more substantial impact on Cals, which indicates that adjustments in FS need some time or lag before materializing the impact of FS on food production and Cals. Therefore, in the short run, adjustments in FS have beneficial outcomes in the form of a high-Cals, but the most favorable gains are realized as past adjustments and improvements in FS. This requires continuous adjustments in the FS to obtain optimal effects on Cals to improve the food security of the country. Hongqing et al. (2021) also describe that the optimizing the land use or improving the FS substantially contributes to the farm benefits especially the food security of the farm families. Similarly, Vatta (2024) determined the favorable and diversified impacts of optimizing FS on farm income, food security, and nutritional outcomes. They reported that a properly adjusted FS substantially contributes to accessibility to food, diet diversity, and nutritional outcomes.

PD strongly affects Cals in both the long and short run. In the long run, a one-unit increase in PD reduces calorie supply by 433.3 kcal per day per capita, while the negative impact of PD indicates that it exerts great pressure on calorie supply because of limited food resources, land, and infrastructure. The negative and significant coefficient of PD at the first difference signifies that the rapid increase in PD in the country lowers calorie supply per capita in the short run. This may be due to the gap between food production and the increase in PD, which disrupts the food supply on an immediate basis. Similarly, high PD also causes farmland fragmentation, which reduces farm size, productivity, and food supply. Popp et al. (2013) highlighted that an increase in PD increases the demand for food by 70%, causes changes in food consumption patterns, and lowers the availability of land for food production, which results in a low supply of calories per capita. Widhiyastuti et al. (2023) determined that in 2019–19 the PD increase by 22.88 people per kilometer square, and over the same time period harvested are and production of rice decreased. Therefore, although food production is increasing globally, the average food self-sufficiency at the country level has been decreasing over the past half-century, and PD is one of the main reasons; by 2010, economies with high PD and urbanization had large deficits in food production (Schramski et al., 2019).

The significant and positive coefficient of VAA indicates that VAA strongly influences calorie supply over time. This implies that a 1% increase in VAA increases calorie supply by 653.3 kcal per day per capita. This indicates that VAA over time substantially enhances calorie supply by improving food availability and lowering food losses. Moreover, the lower coefficient of first difference (=94.58) in the short run as compared to the lagged difference (=194.38) indicates that immediate VAA also improves the calorie supply in the country, while their full impact takes time to materialize. VAA enhances production efficiency and food supply. The adoption of high-yielding crop varieties and value-added practices at farms optimizes the use of land and increases crop yield (Vatta, 2024), which increases food availability (Malec et al., 2023; Sarwar, 2025). VAA to farm products enhances the shelf life of farm products and lowers the spoilage risk of food, which leads to high food availability throughout the year by ensuring food availability during food scarcity (Ceesay et al., 2021; Ogutu et al., 2020). Moreover, VAA also improves food accessibility through wide distribution across different regions of the country, which leads to high market access owing to long shelf life and easy transportation, thereby enhancing food availability in diverse regions (Hazrana and Mishra, 2025: Breisinger et al., 2019).

The significant positive impact of MA over the years indicates that the long-run impact of male participation in agricultural activities in the country is stronger than the short-term impact on caloric supply. Males perform various farming activities such as mechanization and agro-processing, which enhance farm production and efficiency. Male farmers are commonly busy using farm machinery and irrigation, which enhances the production per unit of land. Similarly, male farmers are more capable of cultivating larger areas of land through learning and adopting modern farming techniques. Moreover, humans invest in efficient farm technologies, such as irrigation, which improves food production. Over time, male farmers can accumulate capital and reinvest in their farms, thereby expanding food production and the supply of calories across the country. Therefore, male farmers are more likely to commercialize their farms, which improves their household nutrition and leads to a high-calorie supply (Mgomezulu et al., 2024). They described that in the long run, male farmers are more likely to engage in commercial farming, enhancing their productivity and farm production, leading to high food availability and calorie supply. Moreover, male farmers’ involvement in food sales enhances their family income, which increases their diet diversity, leading to high calorie content, and they are more likely to engage in farming endeavors that directly improve their diet diversity (Yusriadi et al., 2024).

The coefficient of FA indicates that female participation in agriculture strongly enhances caloric supply in the long run, whereas it has a significant negative impact on caloric supply in the short run. This negative impact may be explained by structural barriers, initial productivity disruption, and an inefficient labor market. Historically, low female participation in agriculture is highly linked with cultural and social barriers (Ragasa, 2012), and a sudden increase in female participation in agriculture may require skills, disrupt labor efficiency, and face challenges in male-dominated farming activities. Moreover, the involvement of females in agriculture may not have easy access to the necessary farm resources, which lowers their contribution to calorie supply in the short run. Abdelali-Martini (2011) highlights that the females lack the necessary resources like land, credit and also have limited decision-making power which limited their role in enhancing farm productivity in Middle East and North Africa countries. In the long run, female participation in agriculture gains experience and becomes more productive, contributing to high farm production over time. Das et al. (2023) described the importance of gender equity for development of sustainable and healthy diets, as they highlight that the female role in food systems is very crucial in low-and middle-income economies agriculture as they are structurally disadvantages.

The findings of the ARDL model indicate that ROP and MEC have a strong positive impact in the short run. ROP has a delayed impact on the supply of calories, whereas MEC has an immediate impact on calorie supply. The immediate impact of MEC on farms is primarily positive, and it improves farm efficiency and productivity, which ultimately enhances food availability, resulting in a high-calorie supply. MEC reduces the impact of labor shortages on farms and enhances farm operations. It also increases farm production through appropriate land preparation. Yasar et al. (2024) determined that MEC enhances the food availability by almost 125% in developing economy which results in high food security. Moreover, MEC increased net farm income by 31%, family income by 19%, and food consumption by 5%, which justifies its immediate positive impact on calorie supply (Jena and Tanti, 2023). However, the delayed impact of ROP indicates that resource efficiency needs to adopt efficient technologies that require time to obtain the materialized impact of ROP on Cals.

For robustness, we applied the structural stability test, which confirms the stability of the long-run outcomes. For this purpose, we used the CUSUM and CUSUMQ. Figure 5 presents CUSUM and Figure 6 shows CUSUMQ, and both figures indicate that CUSUM and CUSUMQ remain within the critical bounds of 5%. This signifies the constancy of the parameters and stability of our study model, as both CUSUM and CUSUMQ do not cross the straight lines of the critical bounds.

After determining the long-run relationships among the variables, we used the Granger causality test to analyze the causality between the variables. Owing to cointegration, unidirectional or bidirectional causality between the variables is expected. The findings of Granger causality highlight that FS has a strong causal relationship with Cals at 5%, while reverse causality is not significant, which implies that improving FS can significantly affect the Cals in the country. Similarly, ROP also causes unidirectional Granger causes Cals, and implies that ROP can strongly influence Cals. The outcomes regarding VAA, PD, MEC, and MA indicate that these variables have a bidirectional causality relationship with coal. Table 6 presents the findings of Granger causality for all variables and Figure 7 shows a graphical representation of causality among the variables.