The Interconnectedness of Land–Crops–Livestock and Environmental Quality in Emerging Asian Economies: Challenges of Agriculturalization and Carbonization

Abstract

The release of greenhouse gases (GHGs) is a major contributor to global warming, endangering both human and nonhuman well-being, environmental integrity, economic development, and the planet’s long-term survival. This study delves into the interplay between crop production, livestock production, fertilizer utilization, and agricultural land usage on CO₂ emissions in four Asian economies: China, India, Pakistan, and Bangladesh. Employing panel data analysis techniques, the research uncovers the significant impacts of various agricultural activities on environmental degradation. The findings derived from the panel autoregressive distributed lag (PARDL) estimation reveal that crop production in these emerging economies contributes to CO₂ emissions, as evidenced by the positive coefficients and statistically significant results. Similarly, livestock production and agricultural land used for crop production exhibit a substantial impact on CO₂ emissions, further highlighting their role in environmental degradation. While fertilizer usage also displays a positive coefficient, its impact on CO₂ emissions is not statistically significant. The results of our study highlight the critical importance of addressing the environmental impacts of agricultural practices, particularly in emerging economies. Crop and livestock production, along with the expansion of agricultural land, significantly contribute to CO₂ emissions, which underscores the urgent need for sustainable agricultural practices. These findings suggest that policymakers should prioritize the development and implementation of strategies that mitigate the environmental impacts of agriculture. This could include promoting sustainable land management practices, investing in technology that reduces emissions from crop and livestock production, and encouraging the adoption of eco-friendly fertilizers.

1. Introduction

In the past century, global climate change has emerged as a paramount concern, primarily driven by the escalating levels of greenhouse gases, notably CO₂ emissions. These emissions are mostly caused by human activity like burning fossil fuels and cutting down trees, and have cast a profound shadow on diverse aspects of society. The persistent surge in CO₂ emissions carries profound repercussions across all domains of human civilization. Consequently, mitigating the adverse impacts of global warming and fostering sustainable development have evolved into pressing global imperatives [1,2]. Nonetheless, the agricultural sector, accounting for approximately 14–30 percent of global greenhouse gas (GHG) emissions, has emerged as a pivotal contributor due to its substantial reliance on nonrenewable energy sources [3]. As such, curbing CO₂ emissions while enhancing environmental quality remains a critical nexus in the pursuit of global sustainability. Atmospheric temperature is influenced by greenhouse gases such as CO₂ emission, nitrous oxide, and methane, with their movement controlled by biochemical processes stemming from organic origins and alterations in the physical landscape. The measurement of greenhouse gas (GHG) emissions serves as an indicator of the efficiency of soil carbon cycling and nitrogen utilization. Thus, achieving climate-conscious and productive farming systems hinges on a comprehensive understanding of how land management decisions intricately govern emissions of these gases [4,5]. Within the complex array of issues facing the worldwide food and agricultural sector, there arises an urgent need to increase both the quantity and quality of food production in order to meet the rapidly growing global population. In parallel, the general public, governments, and organizations have focused their efforts on assessing the present natural circumstances that support agricultural operations and livestock production [6]. However, a more comprehensive understanding of the impact of climate change on agricultural yields and livestock production in different geographical areas and management strategies is necessary by formulating broad assessments regarding the impact on food production at a global or regional level but is premature [7].

Amid escalating environmental challenges, a global movement has gained momentum, focusing on the advancement of energy-efficient production, ecologically conscious farming practices, the proliferation of renewable energy sources, and the exploration of alternative economic paradigms. Notably, the agricultural sector finds itself particularly susceptible to the influence of global warming and amplified weather volatility. Unsustainable activities amplify the degradation of land quality, water shortages, species extinction, and the destruction of natural ecosystems. Simultaneously, as the world rallies to mitigate the effects of climate change, the twin goals of cultivating resilient farmland and alleviating poverty by 2030 stand as central pillars of international initiatives [8,9]. In the pursuit of optimizing farmland resources and facilitating the swift expansion of agriculture, the adoption of a low-carbon approach to their utilization has emerged as a pivotal strategy. This strategy stands as a crucial means to not only safeguard the environment but also to uphold food security and maintain the integrity of agricultural produce. As a result, it is now essential to promote environmentally friendly practices while using resources from cultivated land. Therefore, it is essential to accurately determine the carbon footprint produced by their use [10,11]. In the realm of agricultural economics, land stands as an invaluable asset, carrying paramount importance. However, it also holds a substantial role in the generation of greenhouse gas (GHG) emissions. CO₂ emission is intricately linked to a range of practices, including the prolonged persistence of agricultural waste, elevated consumption of resources and energy, intensive cultivation of crops, and the incineration of organic matter [12,13]. However, in this era of urbanization and industrialization, it is difficult to attain food security via effective utilization of inputs due to the effects of global warming and the discharge of significant quantities of harmful substances. A great deal of investigation is now being conducted on how to increase agricultural and environmental efficiency [14,15,16]. When examining the holistic interplay between agricultural productivity and environmental sustainability, a formidable challenge arises from the escalating waste generated throughout the production process. Excessive waste not only contaminates the environment but also worsens the complexities of attaining sustainability. Significantly, agriculture has an essential role in meeting the numerous needs of modern society; it accommodates different food preferences, promotes urban development, adapts to changing lifestyle choices, and ensures worldwide population increase. However, this multifarious role has brought to light stark environmental concerns arising from the intensive utilization of agricultural resources and the disproportionate emission of greenhouse gases (GHG). These issues pose a grave threat to human civilization. It is noteworthy that agriculture emerges as the foremost contributor to human-induced emissions of non-carbon dioxide gases, rendering it a substantial contributor to overall greenhouse gas (GHG) emissions [17,18]. Energy plays a pivotal role in the functioning of economies across various industries, including agriculture, essential for attaining the maximum economic growth. Consequently, the emission of greenhouse gases has become a global phenomenon, leading to atmospheric warming and disruption. This surge in greenhouse gas (GHG) emissions, occurring at an unprecedented pace, has contributed to escalating sea levels, severe climatic events, ozone depletion, and chlorofluorocarbons. This cascade of events has propelled scientific exploration into the myriad of factors impacting natural well-being [19].

Beyond the realm of technological complexities, the endeavor to reduce agricultural CO₂ emissions is entwined with intricate social dimensions. This is due to the fact that farming operations are the primary source of CO₂ emissions. Distinct variations in land utilization patterns and levels of mechanization hold the potential to exert a profound influence over energy consumption in agriculture and the subsequent CO₂ emission [20].

The substantial contribution of CO₂ emissions to worsening global warming highlights a threat to the gradual and sustainable development of mankind. Anthropogenic activities are the main sources of the significant amount of CO₂ emissions released globally. The intricate tapestry of modernization, encompassing processes like industrial advancement, technological strides, and enhancements in both climate and living conditions, embodies the inclination of human society to evolve. Consequently, delving into the intricate interplay between modernization and CO₂ emissions holds a dynamic research frontier. Conversely, the shifting climate landscape imperils the rich tapestry of temperature zones that define the Asian economies. Particularly in the context of fluxes in water and land resources, human needs exert a profound influence over both food provisioning and ecological equilibrium. Despite housing some of the planet’s most ecologically diverse ecosystems, South Asia faces the relentless encroachment of rapid destruction and development, risking the loss of its invaluable biodiversity.

Thus, this present study makes a distinctive and noteworthy contribution to the prevailing literature by addressing the intricate interplay of crop production, livestock production, fertilizer utilization, and agricultural land allocation for crop production across four South Asian economies, specifically, China, India, Pakistan, and Bangladesh. This contribution is forged through the innovative utilization of panel data series, affording a comprehensive exploration of the nexus between these variables. The research methodology embraces the panel autoregressive distributed lag model and panel dynamic least squares, strategically employed to untangle the intricate relationships among these factors. Furthermore, the study employs the Dumitrescu Hurlin panel causality technique to meticulously scrutinize the bidirectional causality linkages that intertwine these variables.

The sections of this investigation are arranged as follows: Section 2 encapsulates a synthesis of the pertinent literature previously explored in connection to the subject matter. The presentation of research data is encapsulated within Section 3, complemented by econometric methodologies. Section 4 assumes the role of presenting the empirical findings and ensuing discussion. Ultimately, Section 5 the study culminates with concluding remarks while also delineating potential avenues for future research endeavors.

2. Related Literature Review

Derived from the expansive geography of land utilization, urban configuration encompasses facets such as city area, population density, and interconnected elements. Yet, present urban design approaches fall short of understanding the complex interaction between various land-use categories and energy-linked CO₂ emissions’ dispersion. In light of local realities encompassing high population densities, inadequate infrastructure, and urban–rural divides, an imperative arises for an exhaustive exploration at the regional scale. This exploration is geared toward unraveling the nuanced connections binding the spatial distribution patterns of distinct land utilization types and the consequential energy-linked CO₂ emissions [21,22]. Consequently, the key contributors in the escalation of greenhouse gases are the emissions of nitrous oxide, carbon dioxide, and methane. These greenhouse gases not only contribute to environmental pollution but also play a pivotal role in catalyzing climate change. The longevity of agricultural productivity and production is significantly compromised by the adverse effects of carbonization and associated undesirable byproducts. When it comes to curbing human-generated greenhouse gas (GHG) emissions in the realm of agriculture, factors such as fertilizers, augmented irrigation practices, seedbed preparation techniques, judicious soil nitrogen management, and effective insect control exert a modest yet consequential influence [23,24,25].

Furthermore, owing to its vast array of ecosystems and diverse natural resources, South Asia assumes a pivotal position in the realm of global economics and ecology. Across the spectrum, recreation, agriculture, transportation, and notably industry have reaped the advantages of harnessing nonrenewable energy sources, particularly fossil fuels, in both burgeoning and established economies over the last decade. This surge in utilization has led to a stark escalation in global CO₂ emissions, underscoring the urgency of the matter [26]. In recent times, global warming has emerged as a prominent issue of worldwide concern, capturing the attention of nations, non-governmental entities, and international organizations alike. While the consequences of climate change have yet to resonate through developed economies, the effects are already acutely felt in emerging economies, especially those with a heavy reliance on agriculture. This variance in impact could be attributed to the nimble responsiveness of advanced economies in tackling climate issues and mitigating their consequences. Conversely, owing to their heightened dependence on agriculture, developing nations confront an elevated vulnerability to the challenges posed by climate change [27,28]. The repercussions of climate change are poised to manifest in the form of severe food scarcity, exacerbated by the dual forces of rapid population growth and sweeping alterations within global food distribution networks. Among the pivotal climatic factors, temperature and precipitation exercise paramount influence on crop production. Both advanced and burgeoning economies harbor profound apprehensions regarding the prevailing climate predicament. In economies undergoing development, agriculture assumes a pivotal role, constituting a cornerstone of the economic framework. In the context of underprivileged nations, where arable land stands as a critical catalyst for improving living standards and ensuring food security, the battle against poverty faces a perilous threat due to the disruptions wrought by global warming [29,30].

The degradation of the environment, propelled by human-induced climate change stemming from operations, such as using fossil fuels and uncontrolled forest fires, can significantly hinder the trajectory toward achieving sustainable development. The release of CO₂ emissions into the atmosphere triggers climate change, culminating in a global warming phenomenon. The response of governments and legislators to the looming threat posed by global warming to natural resources might be characterized by uncertainty. The formulation of policies aimed at curtailing CO₂ emissions, albeit critical, presents an additional conundrum: the prospect for slowing economic growth. This stems from the intrinsic link between reducing CO₂ emissions and concomitant reductions in energy consumption [31,32]. Greenhouse gases, CO₂ emission in particular, stand as prominent causes behind global warming, primarily stemming from human activities like forest combustion and the burning of fossil fuels. The repercussions of the relentless increase in CO₂ emissions are poised to wreak havoc across all sectors of society. Reducing CO₂ emissions and improving air quality have become urgent worldwide priorities in an effort to secure strong development and lessen the negative effects of environmental degradation [33,34]. Moreover, modern farming is primarily reliant on the extraction of natural resources. The triumvirate of water, soil, and energy, being indispensable to the growth and sustenance of crops and agricultural operations, forms an inseparable triad within production systems. The carbon dioxide emanating from the combustion of fossil fuels in agricultural contexts interplays with both water and soil components. As such, the pivotal factors influencing CO₂ emissions in farming hinge on the synergy and efficacy of agricultural land and water energy systems. Comprehending the intricate interrelationship among water, land, and energy in farming, alongside its consequential impact on the environment, is progressively gaining significance. This increased significance arises from the growing requirements imposed on these resources as a consequence of rapidly growing global populations [35,36].

The significance of upholding robust environmental well-being hand in hand with a flourishing economy has been consistently underscored. Diminishing CO₂ emissions have been consistently highlighted in multiple studies as indicative of improved climatic circumstances. The escalation in greenhouse gas (GHG) emissions, particularly CO₂ emissions, stands as a clear indicator of the environmental toll concomitant with economic growth endeavors [37]. Further, the achievements in the realm of agriculture are intricately entwined with the welfare of individuals and the enduring prosperity of the enterprise. In contrast to other sectors, the utilization of energy and the generation of CO₂ emissions within the agricultural domain exhibit distinctive attributes. Although the cultivation of crops, livestock rearing, and seafood harvesting do lead to CO₂ emissions, these operations markedly consume less energy and yield substantially fewer CO₂ emissions compared to the transportation sectors conventionally perceived as being energy-intensive and high-emission industries [38,39]. Prominent drivers of biodiversity decline encompass climate change and shifts in land utilization. Anticipating the repercussions of global warming and land-use alterations on biodiversity assumes paramount importance, ensuring its preservation and the associated advantages bestowed upon humanity. While extensive scrutiny has been devoted to these two factors, often in isolation, their influences on organisms and ecosystems are reasonably well documented. Nonetheless, emerging evidence indicates that the impact of temperature fluctuations and changes in land use on biodiversity does not always increase, as climate change acts to moderate the effects of land-use changes [40,41,42].

3. Methods and Data

3.1. Data Description and Sources

Utilizing panel data encompassing four burgeoning economies, China, India, Pakistan, and Bangladesh, this study was primarily designed to disentangle the influence of crops production, livestock production, fertilizer consumption, and agricultural land dedicated to crops’ productivity in relation to CO₂ emissions. Figure 1 illustrates the important characteristics affecting carbonization in the above countries.

The dataset for the study variables was sourced from two prominent repositories: World Development Indicators (2022) (https://data.worldbank.org/) (accessed on 20 June 2024) and Our World in Data (https://ourworldindata.org/) (accessed on 20 June 2024), spanning the time frame of 1980 to 2020.

Table 1. Description of study variables.

Study Variables	Acronym	Lag Form	Measurement Units
Annual CO2 emission	ACO2e	LACO2e	Million tons
Crops’ production	CP	LCP	Index of 2014–2016 = 100
Livestock production	LP	LLP	Index of 2014–2016 = 100
Fertilizer consumption	FC	LFC	Kilograms per hectare of arable land
Agricultural land used for the crops’ production	AL	LAL	Thousand hectares

provides a detailed interpretation of the variables. A visual representation of the study’s methodological trajectory is presented in Figure 1.

3.2. Model Specification

In the present study, the panel data encompassing the period from 1980 to 2020 were harnessed to scrutinize the interplay among variables, including CO₂ emissions, crop production, livestock production, fertilizer consumption, and agricultural land allocation for crop cultivation. This exploration was carried out within the purview of developing economies, namely China, India, Pakistan, and Bangladesh. To unveil this intricate relationship, the foundational functional structure of the model is articulated as follows:

$$\mathrm{A}\mathrm{C}\mathrm{O}2\mathrm{e}=\mathrm{f}\mathrm{}(\mathrm{C}\mathrm{P},\mathrm{}\mathrm{L}\mathrm{P},\mathrm{}\mathrm{F}\mathrm{C},\mathrm{}\mathrm{A}\mathrm{L})$$

(1)

Equation (1) can be expanded as follows:

$${\mathrm{A}\mathrm{C}\mathrm{O}2\mathrm{e}}_{\mathrm{i}\mathrm{t}}=\mathrm{}\mathrm{f}\mathrm{}({\mathrm{C}\mathrm{P}}_{\mathrm{i}\mathrm{t}}^{\mathsf{\theta }1},\mathrm{}{\mathrm{}\mathrm{L}\mathrm{P}}_{\mathrm{i}\mathrm{t}}^{\mathsf{\theta }2},\mathrm{}{\mathrm{}\mathrm{F}\mathrm{C}}_{\mathrm{i}\mathrm{t}}^{\mathsf{\theta }3},\mathrm{}{\mathrm{}\mathrm{A}\mathrm{L}}_{\mathrm{i}\mathrm{t}}^{\mathsf{\theta }4})$$

(2)

Equation (2)’s functional and logarithmic form can be expressed as:

$$\mathrm{L}{\mathrm{A}\mathrm{C}\mathrm{O}2\mathrm{e}}_{\mathrm{i}\mathrm{t}}=\mathrm{}{\mathsf{\theta }}_0+\mathrm{}{\mathsf{\theta }}_1{\mathrm{L}\mathrm{C}\mathrm{P}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\theta }}_2{\mathrm{L}\mathrm{L}\mathrm{P}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\theta }}_3{\mathrm{L}\mathrm{F}\mathrm{C}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\theta }}_4{\mathrm{L}\mathrm{A}\mathrm{L}}_{\mathrm{i}\mathrm{t}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(3)

In Equation (3), the variable LACO2e represents the natural logarithm of annual CO₂ emissions, LCP stands for the logarithm of crop production, LLP denotes the logarithm of livestock production within emerging economies, LFC uncovers the logarithm of fertilizer consumption, and LAL signifies the logarithm of agricultural land dedicated to crop production. In this case, “t” denotes the length of time component of the panel measurement and “I” stands for the cross-sections. Panel data provide a more comprehensive perspective by combining cross-sectional and time-series information, offering advantages such as better control for unobserved heterogeneity, enhanced statistical power, dynamic analysis, mitigation of endogeneity, and the ability to distinguish between temporary and permanent effects.

3.3. Slope Homogeneity and Cross-Sectional Dependence

Investigating whether or not the panel data are stable is a prerequisite to performing the initial exploration. However, when applied to panel data series, unit root approaches may not be as beneficial because of CSD and slope homogeneity issues. Reliable findings for wide panel data sequences, particularly those with numerous cross-sections, may be obtained using two CSD tests: the Breusch–Pagan LM test and the Pesaran scaled LM test. These two approaches, however, fail to offer credible statistical findings for lower panel samples with lower cross-sections. As a consequence, by following the Pesaran (2015) [43] study, which suggested the cross-sectional test as a solution to this dilemma, it accounts for the relatively small sample bias found that is characteristic of panel data, making it robust. The cross-sectional test can be stated as:

$$\mathrm{C}\mathrm{D}=\sqrt{{\displaystyle \frac{2\mathrm{T}}{\mathrm{N}\left(\mathrm{N}-1\right)}}}\mathrm{}\left(\sum _{\mathrm{m}=1}^{\mathrm{N}-1}\mathrm{}\sum _{\mathrm{k}=\mathrm{m}+1}^{\mathrm{N}}\mathrm{}{\mathsf{\mu }}_{\mathrm{m}\mathrm{k}}\right)$$

(4)

In the aforementioned Equation (4), “N” reveals the integer of observations; “T” depicts the time series and ${\mathsf{\mu }}_{\mathrm{m}\mathrm{k}}$ exposes the country-based residuals. Likewise conventional econometric analyses and methods are also likely to fail due to the variability present between CD and panel data series. The slope homogeneity test is suggested as a solution to the issue. To determine if this issue actually appears in panel data, authors Pesaran and Yamagata (2008) [44] suggested the slope homogeneity test as a measure. The corresponding equation for the evaluation is:

$$\check{∆}=\sqrt{\mathrm{N}}\mathrm{}\left({\mathrm{N}}^{-1}\mathrm{S}\tilde{\mathrm{W}}-{\displaystyle \frac{\mathrm{n}}{2\mathrm{n}}}\right)~\mathrm{}{\mathrm{X}}^2$$

(5)

Furthermore,

$$\check{{∆}_{\mathrm{a}\mathrm{d}}}=\sqrt{\mathrm{N}}\mathrm{}({\mathrm{N}}^{-1}\mathrm{S}\tilde{\mathrm{W}}-{\displaystyle \frac{\mathrm{n}}{\mathrm{v}\left(\mathrm{T},\mathrm{n}\right)}}))~\mathrm{}\mathrm{N}(\mathrm{0,1})$$

(6)

In Equation (6), “N” demonstrates the number of countries utilized for the analysis. The symbol “n” represents an autonomous variable, whereas “v(T, n)” is a complete expression, and “SW” denotes Swami’s statistics.

3.4. Unit Root Testing by Employing Second-Generation Strategies

For the panel data series, we cannot employ the first-generation unit root testing due to the existence of cross-sectional dependency (CSD) and variability. Consequently, for improved results, the second-generation unit root tests were used [43,45]. These use a different methodology to generate more trustworthy findings. As a result, the CIPS unit root test is commonly used in this investigation, and the corresponding test equation can be written as:

$${∆\mathrm{X}}_{\mathrm{i}\mathrm{t}}={\mathsf{\tau }}_{\mathrm{i}\mathrm{t}}+{\mathsf{\tau }}_{\mathrm{i}}{\mathrm{Y}}_{\mathrm{i}\mathrm{t}-1}+{\mathsf{\tau }}_{\mathrm{i}}\overline{{\mathrm{X}}_{\mathrm{t}-1}}+\sum _{\mathrm{m}=0}^{\mathrm{a}}{\mathsf{\tau }}_{\mathrm{i}\mathrm{m}}\overline{{\mathrm{X}}_{\mathrm{t}-1}}\mathrm{}+\sum _{\mathrm{m}=0}^{\mathrm{a}}{\mathsf{\tau }}_{\mathrm{i}\mathrm{m}}{\mathrm{Y}}_{\mathrm{i}\mathrm{t}-1}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(7)

In the above Equation (7), $\overline{{\mathrm{X}}_{\mathrm{t}-1}}$ demonstrates the cross-sectional averages, and furthermore, the CIPS can be obtained by the following equations as:

$$\widehat{\mathrm{C}\mathrm{I}\mathrm{P}\mathrm{S}}={\displaystyle \frac{\mathrm{I}}{2}}\mathrm{}\sum _{\mathrm{i}=1}^{\mathrm{n}}\mathrm{}{\mathrm{C}\mathrm{A}\mathrm{D}\mathrm{F}}_{\mathrm{i}}$$

(8)

In Equation (8), CADF demonstrates the cross-sectional Augmented Dickey–Fuller.

3.5. Panel Cointegration Technique

Panel cointegration tests based on error correction to cointegration techniques will be used in the following investigation to uncover the data series’ long-term dynamics. The long-term associations amid the chosen series were verified with the help of second-generation cointegration investigations. By following the authors Persyn and Westerlund (2008) [46], even without cross-sectional dependence, the results are still reliable. The null assumption was examined in relation to the factor that is dependent of concern. The demonstration of the panel cointegration technique can be stated by the following equation:

$${∆\mathrm{Y}}_{\mathrm{i}\mathrm{t}}={\mathsf{\lambda }}_{\mathrm{i}}{\mathsf{\beta }}_{\mathrm{t}}+{\mathsf{\tau }}_{\mathrm{i}}{\mathrm{Y}}_{\mathrm{i}\mathrm{t}-1}+{\mathsf{\lambda }}_{\mathrm{i}}{\mathrm{X}}_{\mathrm{i}\mathrm{t}-1}+\sum _{\mathrm{m}=0}^{\mathrm{a}\mathrm{i}}{\mathsf{\tau }}_{\mathrm{i}\mathrm{m}}{∆\mathrm{Y}}_{\mathrm{i}\mathrm{t}-\mathrm{m}}+\sum _{\mathrm{m}=-\mathrm{r}\mathrm{i}}^{\mathrm{a}}{\mathsf{\tau }}_{\mathrm{i}\mathrm{m}}{\mathrm{X}}_{\mathrm{i}\mathrm{t}-1}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(9)

The denial of the null hypothesis indicates the occurrence of cointegration in the data series.

3.6. Panel ARDL Technique

This research used panel autoregressive distributed lag with long- and short-run prediction to find the link between the factors. The autoregressive distributed lag method specifications tend to be as follows:

$${\mathrm{Y}}_{\mathrm{i}\mathrm{t}}=\sum _{\mathrm{q}=1}^{\mathrm{b}}{\mathsf{\eta }}_{\mathrm{i}\mathrm{q}}{{(\mathrm{Y}}_{\mathrm{i}})}_{\mathrm{t}-\mathrm{q}}+\sum _{\mathrm{q}=0}^{\mathrm{m}}{\mathsf{\lambda }}_{\mathrm{i}\mathrm{q}}{{(\mathrm{X}}_{\mathrm{i}})}_{\mathrm{t}-\mathrm{q}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(10)

In Equation (10), ${\mathrm{Y}}_{\mathrm{i}\mathrm{t}}$ shows the annual CO₂ emissions; a series of variables that are autonomous is denoted by ${\mathrm{X}}_{\mathrm{i}}$, including crop production, livestock production, fertilizer consumption, and agricultural land used for the crops’ productivity. The short-run and long-run links among the variables, as seen through the lens of the unrestricted error correction framework, can be stated as follows:

$${∆\mathrm{Y}}_{\mathrm{i}\mathrm{t}}={\mathsf{\theta }}_{\mathrm{i}}({\mathrm{Y}}_{\mathrm{i},\mathrm{t}-1}-{\mathsf{\xi }}_{\mathrm{i}}{\mathrm{X}}_{\mathrm{i},\mathrm{}\mathrm{t}-1})\sum _{\mathrm{q}=1}^{\mathrm{b}-1}{\mathsf{\eta }}_{\mathrm{i}\mathrm{q}}{{∆(\mathrm{Y}}_{\mathrm{i}})}_{\mathrm{t}-\mathrm{q}}+\sum _{\mathrm{q}=0}^{\mathrm{m}-1}{\mathsf{\lambda }}_{\mathrm{i}\mathrm{q}}{{∆(\mathrm{X}}_{\mathrm{i}})}_{\mathrm{t}-\mathrm{q}}+{\mathsf{\epsilon }}_{\mathrm{i}\mathrm{t}}$$

(11)

where ${\mathrm{X}}_{\mathrm{i}}$ presents the error correction term in the equation and ${\mathrm{Y}}_{\mathrm{i}}$ indicates the long-run coefficients.

3.7. Panel Causality (Dumitrescu Hurlin) Technique

In the analysis, next, we employed the DH panel causality technique, which is introduced by the authors Dumitrescu and Hurlin [47] in a directive to check the causative linkage amid the CO₂ emissions, crop production, livestock production, fertilizer consumption, and agricultural land utilization for the crops’ productivity. The D-H causality test implies that all values vary across the cross-section, which can help policymakers more effectively comprehend the causative connections among the highlighted variables and adopt suitable policies to enhance sustainable crop production. The testing paradigm can be expressed as an equation as follows:

$${\mathrm{Y}}_{\mathrm{i}\mathrm{t}}={\mathsf{\beta }}_{\mathrm{i}}+\sum _{\mathrm{q}=1}^{\mathrm{Z}}{\mathsf{\eta }}_{\mathrm{i}}^{\left(\mathrm{q}\right)}{\mathrm{Y}}_{\mathrm{i},\mathrm{t}-\mathrm{q}}+\sum _{\mathrm{q}=1}^{\mathrm{Z}}\mathrm{}{\mathsf{\phi }}_{\mathrm{i}}^{\left(\mathrm{q}\right)}{\mathrm{X}}_{\mathrm{i},\mathrm{}\mathrm{t}-\mathrm{q}}+{\mathsf{\epsilon }}_{\mathrm{i},\mathrm{t}}$$

(12)

In Equation (12), “z” indicates the lag length for the variables, ${\mathsf{\eta }}_{\mathrm{i}}^{\left(\mathrm{q}\right)}$ shows the autoregressive constraint, and ${\mathsf{\phi }}_{\mathrm{i}}^{\left(\mathrm{q}\right)}$ stands for the slope coefficient that varies between different cross-sections.

4. Empirical Findings and Discussion

4.1. Descriptive and Covariance Analysis

An examination of the descriptive statistics for the following variables: CO₂ emissions, agricultural acreage used for crop production, livestock production, fertilizer usage, and crop production in four developing economies, are stated in

Table 2. Descriptive analysis.

	LACO2e	LCP	LLP	LFC	LAL
Mean	19.591	4.170	4.018	5.011	18.134
Median	19.341	4.177	4.082	4.994	18.236
Maximum	23.117	4.747	4.837	6.141	20.085
Minimum	15.843	3.325	2.679	3.515	16.023
Std. Dev.	2.080	0.357	0.515	0.600	1.523
Skewness	0.025	−0.202	−0.480	−0.187	−0.105
Kurtosis	1.789	1.904	2.393	2.831	1.529
Jarque–Bera	10.033	9.320	8.820	1.155	15.082
Probability	0.006	0.009	0.012	0.561	0.000

. Outcomes display that CO₂ emission has the highest maximum value (23.117). Further, the agricultural land for the crops’ production has a huge minimum value (16.023). The Jarque–Bera statistics for all variables ACO2e, CP, LP, FC, and AL are (10.033), (9.320), (8.820), (1.155), and (15.082), respectively. Similarly, the Kurtosis statistics of all tested variables are (1.789), (1.904), (2.393), (2.831), and (1.529). On the other hand, the test statistics of the covariance analysis for CO₂ emission and all other independent variables are explored in

Table 3. Covariance analysis.

	LACO2e	LCP	LLP	LFC	LAL
LACO2e	1.000
	-----
	-----
LCP	0.246	1.000
	3.239	-----
	(0.001)	-----
LLP	0.244	0.928	1.000
	3.215	31.726	-----
	(0.001)	(0.000)	-----
LFC	0.670	0.452	0.588	1.000
	11.515	6.464	9.259	-----
	(0.000)	(0.000)	(0.000)	-----
LAL	0.943	−0.044	−0.059	0.461	1.000
	36.347	−0.569	−0.756	6.624	-----
	(0.000)	(0.570)	(0.450)	(0.000)	-----

.

4.2. Cross-Sectional Dependence Outcomes

We first present the CSD test for the research model, after which we show how to implement the Breusch–Pagan LM, Pesaran scaled LM, Bias-corrected scaled LM, and Pesaran CD methods for identifying the panel unit root. Based on the data presented in

Table 4. CSD outcomes.

Tests	LACO2e [Stat. (p-Values)]	LCP [Stat. (p-Values)]	LLP [Stat. (p-Values)]	LFC [Stat. (p-Values)]	LAL [Stat. (p-Values)]
Breusch–Pagan LM	238.935 *** (0.000)	230.563 *** (0.000)	232.726 *** (0.000)	122.650 *** (0.000)	53.993 *** (0.000)
Pesaran scaled LM	67.242 *** (0.000)	64.825 *** (0.000)	65.450 *** (0.000)	33.674 *** (0.000)	13.854 *** (0.000)
Bias-corrected scaled LM	67.192 *** (0.000)	64.775 *** (0.000)	65.400 *** (0.000)	33.624 *** (0.000)	13.804 *** (0.000)
Pesaran CD	15.457 *** (0.000)	15.182 *** (0.000)	15.251 *** (0.000)	9.577 *** (0.000)	−1.953 ** (0.050)

Note: ** signifies p < 0.05, and *** signifies p < 0.01.

, we assume that our panel data set is subject to cross-sectional interdependence. Due to the interconnected nature of the economies in all countries, it is essential to use advanced, second-generation techniques for evaluating the CSD technique. Based on the findings presented in Table 4, study variables have spatial dependence cross-sectionally. CSD in panel data can indicate the presence of spatial spillover effects. Spatial spillovers occur when an event or condition in one location influences the outcomes or behaviors in nearby locations. For example, an increase in economic activity in one region may positively impact the economic performance of neighboring regions due to trade and investment linkages. Commenting on the cross-sectional dependence result for the sample of four Asian developing economies (China, India, Pakistan, and Bangladesh) is essential for understanding the interrelationships between the variables and their impact on ecological degradation. In this context, spatial dependence among these countries may yield valuable insights into how their agricultural activities collectively contribute to ecological degradation in the region. The presence of cross-sectional dependence suggests that the observations in the panel dataset are not independent across the four countries. This could be due to shared characteristics, geographical proximity, common environmental challenges, or similar policy frameworks influencing agricultural practices and ecological outcomes.

When examining the spatial dependence among these countries, several key observations may emerge: The ecological degradation in one country could potentially spill over to neighboring countries, especially if environmental issues, such as air and water pollution, do not respect national borders. Understanding spatial spillover effects can highlight the need for regional collaboration and coordinated efforts to address common environmental challenges. Additionally, the four countries face similar environmental challenges, leading to the implementation of shared policies and regulations to mitigate ecological degradation. Analyzing these shared policies can provide insights into the effectiveness of regional approaches to environmental management. Economic interdependencies and trade relationships among the countries may influence their agricultural practices and ecological impacts. For instance, if a country is highly dependent on agricultural exports to other economies, this might impact the use of fertilizers and agricultural land in countries that import and export the goods.

4.3. First- and Second-Generation Stationary Testing Outcomes

Table 5. Unit root testing (first-generation) results.

Methods	LACO2e	LCP	LLP	LFC	LAL
	[t-Stat (p-Values)]	[t-Stat (p-Values)]	[t-Stat (p-Values)]	[t-Stat (p-Values)]	[t-Stat (p-Values)]
Im, Pesaran and Shin (Level)	−0.144 (0.442)	0.822 (0.794)	−0.847 (0.198)	−0.966 (0.166)	−2.979 (0.001)
Im, Pesaran and Shin (First Difference)	−3.562 *** (0.000)	−8.120 *** (0.000)	−4.784 *** (0.000)	−10.463 *** (0.000)	−5.817 *** (0.000)
ADF—Fisher (Level)	−0.131 (0.447)	0.777 (0.781)	−0.795 (0.213)	−1.007 (0.157)	−2.148 *** (0.015)
ADF—Fisher (First Difference)	−3.458 *** (0.000)	−7.279 *** (0.000)	−4.518 *** (0.000)	−8.648 *** (0.000)	−5.341 *** (0.000)
PP—Fisher (Level)	−0.663 (0.253)	−0.225 (0.410)	−0.795 (0.213)	−2.573 *** (0.005)	−3.822 *** (0.000)
PP—Fisher (First Difference)	−7.352 *** (0.000)	−9.903 *** (0.000)	−6.525 *** (0.000)	−8.474 *** (0.000)	−7.516 *** (0.000)

Note: *** signifies p < 0.01.

and

Table 6. CADF unit root test outcomes.

Variables	At Level	p-Value	1st-Difference	p-Value
LACO2e	−1.717	0.554	−3.104 ***	0.002
LCP	−1.386	0.799	−3.174 ***	0.002
LLP	−0.578	0.995	−2.841 ***	0.012
LFC	−2.217	0.176	−2.764 ***	0.018
LAL	−2.285	0.141	−3.917 ***	0.000

Note: *** signifies p < 0.01.

display the outcomes of the first- and second-generation unit root tests’ results for the variables; this reveals that the variables are stable at the initial difference I(1). Next, we need to think about how strongly the factors are linked through cointegration. Therefore, cointegration is found to be prevalent among variables if they are incoherently unstable but maintain the same pattern and move in the same direction or together.

A unit root is a feature of a non-stationary time series, indicating the presence of a stochastic trend and its statistical characteristics, the variance and mean, for example, vary throughout time. The presence of unit roots in panel data can have important implications for econometric modelling and inference. If the panel data contain non-stationary series (unit roots), they may require first-differencing or other transformations to achieve stationarity before conducting further analyses, such as panel cointegration or panel regression. In our analysis, we proceeded with cointegration analysis to avoid spurious regression in panels: When non-stationary variables are regressed against each other in a panel dataset, spurious regression can occur. This means that apparent significant relationships may arise purely due to the presence of common time trends or deterministic components across the entities, rather than genuine economic relationships. Cointegration analysis helps distinguish between spurious and meaningful relationships by identifying the long-run equilibrium relationships.

By identifying cointegrating relationships and distinguishing them from spurious correlations, researchers can gain a better understanding of the underlying economic dynamics and obtain more reliable and meaningful results in panel data analysis. Table 6 shows panel cointegration results, which imply the long-run interactions and co-movements between non-stationary variables at the cross-sectional level.

The cointegration among the factors under contemplation via long-run estimation is supported by the results of the Johansen and Fisher (J.F) group test shown in

Table 7. JF (Johansen Fisher) panel cointegration test.

H-No. of CE(s)	F-Stat. (from Trace Test)	Prob.	F-Stat. (from Max-Eigen Test)	Prob.
(None)	70.60	(0.000)	41.77	(0.000)
(max. at (1))	43.44	(0.000)	24.56	(0.001)
(max. at (2))	24.60	(0.001)	13.26	(0.103)
(max. at (3))	17.84	(0.022)	11.13	(0.194)
(max. at (4))	22.35	(0.004)	22.35	(0.004)

and the individual cross-section tests shown in

Table 8. Johansen F panel cointegration test (individual cross-section results).

Individual Cross-Section Results
Cross-Section	Trace Test Statistics	Prob.	Max-Eigen Test Statistics	Prob.
(Hypothesis of no cointegration)
China	155.864	(0.000)	64.575	(0.000)
India	71.576	(0.036)	31.801	(0.086)
Pakistan	61.417	(0.194)	32.249	(0.077)
Bangladesh	86.235	(0.001)	32.353	(0.075)
(Hypothesis of at most 1 cointegration relationship)
China	91.288	(0.000)	43.894	(0.000)
India	39.774	(0.230)	17.105	(0.570)
Pakistan	29.168	(0.760)	14.102	(0.815)
Bangladesh	53.881	(0.012)	27.326	(0.053)
(Hypothesis of at most 2 cointegration relationship)
China	47.394	(0.000)	27.116	(0.006)
India	22.669	(0.262)	11.706	(0.576)
Pakistan	15.065	(0.775)	9.207	(0.815)
Bangladesh	26.555	(0.113)	13.120	(0.441)
(Hypothesis of at most 3 cointegration relationship)
China	20.277	(0.008)	14.708	(0.042)
India	10.962	(0.213)	8.366	(0.342)
Pakistan	5.858	(0.712)	4.713	(0.777)
Bangladesh	13.435	(0.099)	8.423	(0.337)
(Hypothesis of at most 4 cointegration relationship)
China	5.569	(0.018)	5.569	(0.018)
India	2.596	(0.107)	2.596	(0.107)
Pakistan	1.144	(0.284)	1.144	(0.284)
Bangladesh	5.011	(0.025)	5.011	(0.025)

.

4.4. Long- and Short-Run Estimations

The investigation also analyzed the long-term and short-term tendencies of CO₂ emission, crop production, livestock production, fertilizer usage, and agricultural land used for crop production in the chosen Asian countries by using the panel ARDL technique. The outcomes of long-run analysis presented in

Table 9. Short- and long-run analysis.

Long-Run Dynamics
Variables	Coefficients	Std Error	t-Statistic	Prob. *
LCP	0.343 **	0.156	2.201	0.029
LLP	0.695 ***	0.228	3.043	0.002
LFC	0.090	0.101	0.888	0.376
LAL	0.222 *	0.382	0.581	0.062
Short-run analysis
CointEq(-1)	−0.279 *	0.164	−1.700	0.091
D(LCP)	−0.074	0.074	−1.004	0.316
D(LLP)	0.127	0.266	0.477	0.633
D(LFC)	0.035	0.034	1.010	0.314
D(LAL)	1.092	1.530	0.713	0.476
C	2.596 **	1.290	2.012	0.046
Trend	0.005	0.004	1.197	0.233
[M-dep. var]	(0.053)	[SD-dep. Var]		(0.040)
[SE Regression]	(0.035)	[AIC]		(−3.567)
[SS Resid]	(0.170)	[Schwarz Criterion (SC)]		(−2.962)
[Log-likelihood]	(324.529)	[H-Quinn Criterion]		(−3.321)

Note: * signifies p < 0.1, ** signifies p < 0.05, *** signifies p < 0.01.

show that crop production in the selected Asian economies positively influenced CO₂ emission with a coefficient (0.343) and probability value (0.029). The findings also show that livestock production and agricultural land used for crop production also positively impact CO₂ emissions with coefficients (0.695), (0.222) and probability values (0.002) and (0.062), respectively. These findings are consistent with the broader understanding that intensive agricultural practices, particularly in crop and livestock production, are major sources of greenhouse gases (GHGs).

In contrast, results from the short-run are not statistically significant except the constant term of the panel regression analysis. Considering the sign of the estimated parameters, livestock production, fertilizer consumption, and agricultural land utilization have a positive impact on environmental degradation. The lack of statistically significant results in the short-run, except for the constant term, suggests that the immediate impact of agricultural practices on CO₂ emissions may be less pronounced or more variable across the selected countries. This could be due to seasonal variations, differences in agricultural practices, or short-term mitigation efforts. However, the significant long-term impacts underscore the cumulative effect of continuous agricultural expansion and intensification on the environment.

Global warming has had a disastrous influence on the economic well-being of many nations, particularly in the agricultural sector. Agricultural endeavors are found in practically every nation and provide producers with marketable resources, enabling them to develop a wide range of goods. Agriculture, more than any other sector of the economy, is experiencing the consequences of changing climates and other severe weather events, including unusual rainfall and temperature swings [48,49]. CO₂ emissions are the primary cause of global warming, which in turn challenges human civilization’s ability to progress sustainably. Human actions account for the vast majority of worldwide CO₂ emissions. Industrial upgrading, technological progress, and improvements to natural and living conditions are all phenomena included under the concept of advancement, which refers to the inherent tendency of society to progress. As a result, it is a moving dilemma to investigate how modern technology influences CO₂ emissions. Several distinct weather patterns coexist in South Asia, making the region both adaptable and susceptible to global warming. Especially in the face of shifts in water and land resources, human demands have a significant impact on food supply and ecological stability. South Asia has some of the most exceptional ecosystems globally, but they are at risk as a consequence of rapid deterioration and intensified industrialization [50,51,52].

The decrease in agricultural land is a result of increasing development, construction, temperature changes, and environmental issues such as energy use, mining, and urbanization, which provide challenges for rapid agricultural output growth. Greenhouse gas pollution-induced climate change impacts temperature, precipitation, and ultraviolet ray exposure, ultimately affecting agricultural production. The agricultural production patterns are being affected by these developments, which are substantially to blame for the rising food insecurity in emerging economies. Climate change will reduce agricultural production, especially in the world’s poorest countries [53,54]. The health and welfare of people, as well as the availability of water and food, are all put at risk by the effects of severe temperature and weather events, such as contaminants in the air. Variations in the increase in Earth’s normal temperature as a consequence of the greenhouse effect are the root cause of global warming, which in turn effects variations in precipitation, sunshine, weather patterns, and heat waves [55]. Reduced agricultural output and rising global temperatures are both a consequence of CO₂ emissions and consequent climate change. Although the significance of low-carbon consumption, manufacturing, and a green economy is widely recognized, the transition from traditional energy sources like petroleum and natural gas to renewable energy while simultaneously lowering CO₂ emissions is challenging [56,57,58].

Changes in the climate and environmental degradation are two of the most urgent global issues today. Reducing total CO₂ emissions is important to the climate change battle. Key home economic areas linked to people’s day-to-day existence and steady economic growth includes food production, energy usage, and CO₂ emissions. Agriculture and livestock are sources of CO₂ that release less than traditionally deemed high-emitting and resource-intensive firms, but they share many of the same traits with those sectors [59]. An increase in atmospheric carbon dioxide exacerbates other natural pressures. Several factors contribute to rising CO₂ emissions, including modernization, urbanization, unequal economic development, population increase, rapid growth changes, and resource utilization. In the context of rising populations and expanding economies, increasing agricultural demand is crucial for ensuring food supply and promoting a fairer distribution of commodities. The rise in CO₂ emissions has been much greater than was predicted, and it can be traced directly to improper actions [60,61]. The long-term interconnection of crop production, livestock production, fertilizer usage, and agricultural land use for crop production to CO₂ emission in selected emerging economies is illustrated in Figure 2.

4.5. Panel Dynamic OLS (Robustness Check)

In the investigation, we also employed the panel dynamic least squares (DOLS) technique to uncover the nexus among the variables for the economies of China, India, Pakistan and Bangladesh. The findings shown in

Table 10. Robustness test (panel dynamic least squares (DOLS)).

Variables	Coefficients	Std. Error	t-Statistic	Prob.
LCP	0.650 ***	0.202	3.217	0.002
LLP	0.562 *	0.303	1.855	0.067
LFC	0.147	0.127	1.159	0.250
LAL	0.255	0.544	0.469	0.639
[R Sq.]	0.999	[M-dep. var]		19.600
[Adj-R Sq.]	0.999	[SD-dep.var]		2.063
[SER]	0.056	[Sum-Sq resid]		0.219
[Long-run var]	0.002	-

Note: * signifies p < 0.1, *** signifies p < 0.01.

demonstrate that crop production in emerging economies positively impacted environmental degradation with coefficient (0.650) and having the probability value (0.002). This suggests that agricultural practices in these countries, particularly in crop production, contribute significantly to environmental degradation. The intensive use of resources such as water, energy, and fertilizers in crop production likely drives this impact, emphasizing the need for sustainable agricultural practices. Further, livestock production shows a positive coefficient (0.562), which means it also positively influenced CO₂ emission in sample countries. Livestock farming is a known source of methane, a potent greenhouse gas, as well as CO₂ emissions related to feed production, manure management, and other associated activities. The significant impact of livestock production on environmental degradation in these emerging economies highlights the need for improved livestock management practices that reduce greenhouse gas emissions.

Moving to fertilizer consumption and agricultural land utilized for the productivity of the crop demonstrated a constructive linkage with coefficients (0.147) and (0.255), which means these two variables also positively impacted CO₂ emission in selected economies, although they are not statistically significant. This might suggest that while these factors do contribute to environmental degradation, their effects are either less pronounced or more variable in the selected period. Nonetheless, the positive signs indicate that these practices still contribute to CO₂ emissions, reinforcing the need for careful management of fertilizer application and land use to mitigate environmental impacts.

The results suggest that emerging economies need to adopt more sustainable agricultural practices to mitigate the environmental impact of crop and livestock production. Policymakers should focus on promoting sustainable land management, efficient fertilizer use, and improved livestock practices to reduce CO₂ emissions. Additionally, investments in technology and infrastructure that support sustainable agriculture could help balance the need for agricultural productivity with environmental sustainability.

4.6. DH Panel Causality Results

Table 11. DH panel causality technique.

(Null Hyp.)	(W-Stat.)	(Zbar-Stat.)	(Prob.)
LCP ≠→ LACO2	5.242	1.428	0.153
LACO2 ≠→ LCP	6.355 **	2.208	0.027
LLP ≠→ LACO2	5.330	1.490	0.136
LACO2 ≠→ LLP	2.949	−0.180	0.856
LFC ≠→ LACO2	6.303 **	2.172	0.029
LACO2 ≠→ LFC	5.207	1.403	0.160
LAL ≠→ LACO2	2.688	−0.363	0.716
LACO2 ≠→ LAL	3.923	0.502	0.615
LLP ≠→ LCP	5.741 *	1.778	0.075
LCP ≠→ LLP	2.913	−0.205	0.837
LFC ≠→ LCP	5.679 *	1.735	0.082
LCP ≠→ LFC	2.737	−0.328	0.742
LAL ≠→ LCP	7.782 ***	3.210	0.001
LCP ≠→ LAL	8.198 ***	3.502	0.000
LFC ≠→ LLP	2.788	−0.293	0.769
LLP ≠→ LFC	4.408	0.843	0.399
LAL ≠→ LLP	6.809 ***	2.527	0.011
LLP ≠→ LAL	6.916 ***	2.602	0.009
LAL ≠→ LFC	5.396	1.536	0.124
LFC ≠→ LAL	6.398 **	2.239	0.025

Note: * signifies p < 0.1, ** signifies p < 0.05, *** signifies p < 0.01. ≠→, indicates “does not homogeneously cause”.

presents a concise overview of the results obtained from the Dumitrescu Hurlin panel causality test. The analysis reveals several statistically significant causal relationships between the variables of CO₂ emissions, crop production, livestock production, fertilizer usage, and agricultural land used for crop production. Specifically: There is a significant causality from CO₂ emissions to crop production. CO₂ emissions’ causal link to crop production suggests that changes in environmental conditions, potentially due to increased emissions, could be influencing agricultural practices. This might reflect a need for adaptive measures in crop production as a response to the changing climate, leading to alterations in farming practices or shifts in the types of crops being grown.

Fertilizer consumption significantly causes CO₂ emissions. This underscores the environmental impact of intensive fertilizer use. Fertilizers, particularly nitrogen-based ones, can lead to increased emissions of nitrous oxide, a potent greenhouse gas, thereby contributing to overall CO₂ emissions.

Livestock production significantly causes crop production. The causality from livestock production to crop production indicates that livestock farming practices might be influencing the demand for crops, likely as feed. This relationship highlights the interconnected nature of agricultural sub-sectors and the cascading effects that changes in one area can have on others.

Fertilizer consumption significantly causes crop production. Fertilizers, particularly those containing essential nutrients like nitrogen, phosphorus, and potassium, play a critical role in enhancing soil fertility, which in turn boosts crop yields. This causality suggests that as farmers increase their use of fertilizers, they are likely to see a corresponding increase in crops productivity.

Agricultural land significantly causes livestock production and livestock production significantly causes agricultural land. This bidirectional causality suggests a cycle where increases in agricultural land lead to greater livestock production and the resulting growth in livestock farming further drives the expansion of agricultural land. While this cycle can promote agricultural growth, it also poses challenges for sustainable land management.

Fertilizer consumption significantly causes agricultural land. This causal relationship underscores the interconnected nature of agricultural inputs and land use. While fertilizers are a critical tool for increasing agricultural productivity, their use can also lead to greater pressure on land resources. This finding highlights the need for sustainable fertilizer management practices that optimize land use without leading to unnecessary expansion, which could have adverse environmental impacts, such as deforestation, soil degradation, and loss of biodiversity.

Finally, there is bidirectional causality between agricultural land and livestock production. This bidirectional relationship highlights the close interdependence between land use and livestock farming. To manage this relationship sustainably, policies should focus on optimizing the use of existing agricultural land for livestock production through improved management practices, rather than relying on continuous land expansion.

5. Conclusions and Policy Implications

Undoubtedly, GHG emissions are the predominant cause of global warming, exerting a profound impact on agricultural systems in both developed and developing economies worldwide. This study meticulously examines the effects of crop production, livestock production, fertilizer utilization, and agricultural land allocation for crop productivity within four developing economies: China, India, Pakistan, and Bangladesh. Utilizing a panel data series and applying a range of panel techniques, this investigation delves into the intricate relationships between these variables. The long-term outcomes from the panel autoregressive distributed lag (PARDL) approach reveal that crop production, livestock production, and agricultural land used for crop production all positively influence CO₂ emissions. Fertilizer usage, while also showing a positive coefficient, does not have a statistically significant impact on CO₂ emissions in the long term. The results from the panel dynamic least squares method further support these findings, demonstrating the significant positive influence of these agricultural activities on CO₂ emissions in these rapidly developing countries. Additionally, the Dumitrescu Hurlin panel causality tests establish several key causal links, including bidirectional causality between agricultural land and livestock production. These results underscore the interconnectedness of agricultural practices and their environmental impacts, highlighting the need for sustainable management practices in these regions.

5.1. Policy Implications

The findings highlighted in the study underscore the urgent need for policy interventions aimed at mitigating the environmental impacts of agricultural practices in rapidly developing Asian economies such as China, India, Pakistan, and Bangladesh. Given that crop production, livestock farming, and the expansion of agricultural land significantly contribute to CO₂ emissions, it is essential that these countries adopt more sustainable agricultural practices. Policymakers must prioritize strategies that not only enhance productivity but also minimize the environmental footprint of agricultural activities. This could involve the promotion of crop rotation, agroforestry, and organic farming techniques that maintain soil health while reducing the dependence on chemical inputs.

In addition to adopting sustainable practices, there is a critical need to invest in technologies that can lower emissions from agricultural activities. Precision farming technologies, which optimize the use of inputs such as water and fertilizers, offer a promising approach to reducing waste and emissions. Similarly, advancements in livestock management, including methane-reducing feed additives and improved manure management systems, can help significantly cut down on greenhouse gas emissions from livestock farming, which has been identified as a major contributor to CO₂ emissions in the study.

Moreover, the study points to the importance of integrated land-use planning as a key strategy to address the environmental impact of agricultural expansion. As agricultural land use emerges as a significant driver of CO₂ emissions, it is vital that land is managed in a way that balances agricultural productivity with environmental conservation. This could involve zoning regulations that protect ecologically sensitive areas from agricultural encroachment and initiatives to restore degraded lands. By optimizing the use of existing agricultural land and preventing unnecessary expansion, policymakers can help reduce the pressure on natural ecosystems and contribute to long-term environmental sustainability.

5.2. Future Research Directions and Limitations

This study delves into the ramifications of crop production, livestock productivity, fertilizer consumption, and agricultural land allocation for crop productivity within a specific cluster of Asian economies, namely China, India, Pakistan, and Bangladesh. Nevertheless, certain limitations inherent to this study can serve as signposts for future research endeavors. Researchers are urged to employ similar econometric methodologies across different sets of economies, spanning varied data periods for diverse parameters in both developed and developing contexts. Such efforts hold the promise of illuminating intricate environmental matters. To expand the current research horizon, incorporating supplementary variables such as financial expansion, agricultural commodities, agricultural credit, temperature, rainfall, urbanization, energy consumption, and environmental sustainability emerges as a critical step. By doing so, more robust and comprehensive outcomes can be gleaned, painting a more detailed and nuanced picture of the complex interplay between these factors.