A Multi-Method Analysis of Risk Mitigation Strategies for the Livestock Supply Chain

Abstract

The livestock sector is a significant contributor to the economy and rural livelihoods, but it is exposed to high risk across the supply chain, which is detrimental and needs to be addressed for sustainable development. Therefore, this study aimed to identify the major risk mitigation strategies (RMSs) and associated factors that affect their adoption. This study conducted a comprehensive literature review to identify the eight major RMSs and prioritized them through an analytical hierarchical process (AHP). Thereafter, a multivariate probit (MVP) model was developed to identify the factors affecting the adoption of major RMSs. The primary RMSs are livestock insurance, vaccination of livestock, and advisory/extension services. Further, the multivariate probit regression analysis shows that ‘age’, ‘social category’, ‘economic status’, ‘educational level’, ‘income level’, ‘the total number of animals’, and ‘perceived risk of foot and mouth disease’ are significant factors that influence the adoption of RMSs. This study’s findings will be useful for livestock supply chain partners to mitigate the risks along the livestock supply chain. This research will also help policymakers to develop policies/plans for incorporating these RMSs by considering the influencing associated factors.

1. Introduction

Livestock plays a multifaceted role in rural farm households and broader economies, serving as a critical source of food, a means of generating income, and a mode of transportation [1]. The growth of the livestock sector appears to contribute to a greater extent to poverty reduction in comparison to crop production [2,3]. Livestock serves as an asset that can be liquidated during times of income shocks caused by unforeseen events such as accidents, droughts, or floods, thereby providing a critical financial safety net [4]. This sector ensures nutritional and food security globally by fulfilling the energy requirements of the world population [5,6,7].

In India, dairy represents the largest agricultural commodity, contributing approximately 5% to the national economy and providing direct employment to over 80 million farmers. India’s leadership in the dairy sector is evident both nationally and globally, as it ranks first in milk production, accounting for 25% of the world’s total milk output (https://pib.gov.in/PressReleasePage.aspx?PRID=2086052#:~:text=2023%20(est.)-,(Food%20Outlook%20November’2024).,meat%20production%20in%20the%20world (accessed on 3 January 2025)). Further, India ranks second in egg production and fifth in meat production in the world. However, due to several risks across the supply chain, various losses have been recorded. These risks include biotic and abiotic stresses, lack of living infrastructure (which includes shelter, air, and lighting, as well as protection from noise), climate change, a high cost and poor quality of feed and fodder, lack of market information, and foot and mouth disease. These risks are not only affecting the productivity of livestock but also all stakeholders in the supply chain in terms of a reduction in income level [8,9].

These risks need to be mitigated through various risk management strategies (RMSs) that make the livestock supply chain efficient. Livestock farming practitioners can adopt a wide range of RMSs, such as vaccination, livelihood diversification, cooperative membership, saving, credit from formal and informal sources, pest management, organized input supply, and established shelters [10,11]. Farmers’ education, extension advisory services, and availability of credit stimulate them to adopt these RMSs [12]. A plethora of studies are available that examine RMSs across the food supply chain, considering various aspects such as perishable products, vegetable supply chains, and fresh food supply chains [13,14]. However, the understanding of RMSs for the livestock supply chain remains in its nascent stages, necessitating comprehensive research, particularly in the context of a developing country like India. It is evident from the literature review that there is very limited research examining the factors that influence the adoption of management strategies in agriculture in general and livestock [15]. It is imperative to analyze RMSs and the factors influencing their adoption within the livestock supply chain to enhance resilience and sustainability. Policy planners and managers are interested in understanding which factors are affecting the adoption of RMSs. Therefore, this paper aims to examine the RMSs relevant to producers in India. It is necessary to identify these strategies when exploring effective risk management in the livestock supply chain. Furthermore, identification itself is not enough because livestock farmers and other supply chain partners are not able to deploy all RMSs at once [16]. Therefore, it is necessary to prioritize these RMSs for better deployment using the optimal resources. Doing so means livestock supply chain partners can strategize their action plans for risk reduction along the supply chain. The present study is intended to assist policymakers in designing appropriate products and coming up with sensible policy designs. Therefore, the primary objectives of this study are as follows:

• Identify the significant RMSs for the livestock supply chain.
• Prioritize these RMSs for better adoption.
• Investigate the factors that contribute towards the adoption of these RMSs.

This study is considered to make significant contributions to the existent theory and practices in the area of risk management for the livestock supply chain. The major contributions of this study are the following:

• The proposed study will help managers and policy planners to identify and evaluate RMSs for the livestock supply chain. It is not only limited to the identification and prioritization of RMSs but also considers the underlying factors that are responsible for their adoption.
• This study lists eight RMSs that will improve risk management practices, livestock supply chain performance, economic viability, and sustainability.
• The identification and analysis of the RMS adoption factors will help managers, consultants, and professionals adopt these strategies efficiently.
• The proposed framework uses a combination of a decision model and an empirical model that is logically sound and reliable.

A three-phase framework is adopted to identify the risk in the livestock supply chain and determinants of risk management strategies’ adoption. The first two phases are based on qualitative analysis while the third phase analyzes the difference in the socio-demographic of adopters and non-adopters of RMSs using the Chi-square test, and factors affecting the adoption of RMSs are identified using the multivariate probit (MVP) model. Three hypotheses are formulated:

Hypothesis 1: There is no difference in adoption of risk management strategies across the socio-demographic profile of farmers.

Hypothesis 2: The socio-demographic profile of farmers does not affect the adoption of risk management strategies.

Hypothesis 3: Farm factors do not affect the adoption of risk management strategies.

On the basis of the AHP, this study concludes that livestock insurance, vaccination of livestock, and advisory/extension services are the major risk management strategies (RMSs). Furthermore, from the primary survey, a difference between the socio-demographic profiles of adopters and non-adopters of RMSs is found. Adoption of RMSs (livestock insurance, vaccination, and advisory/extension services) is high among educated and high-income farmers with large farms. Further, age, economic status, and the perceived risk of FMD have significant implications for the adoption of RMSs.

2. Literature Review

2.1. Livestock Supply Chain Risks

In India, small farm-holders constitute a large portion of livestock farming but they are experiencing the problem of poor infrastructure and cannot overcome it due to a lack of investment [17]. These livestock farmers deal with a long supply chain, from producers to consumers, and are exposed to a variety of risks such as financial risks, production and operational risks, and market and price risks [1,17,18,19]. In addition, market risks also include those related to the product price, availability, accessibility, and production quality [20,21,22]. Thornton [7] suggested that prices of livestock products are generally highly volatile due to the perishable nature of the products. Additionally, an irregular supply and cost of feed and fodder are significant risks in this regard.

Livestock productivity shows seasonal variation, and its supply chain is characterized by an imbalance of demand and supply [23]. The primary market for livestock output typically consists of urban communities, which are often located far from the points of production. Therefore, most small farmers encounter the risk of market access and bargaining power. In addition, this spatial separation presents logistical challenges in ensuring the efficient transportation, storage, and distribution of perishable livestock products, such as meat, dairy, and eggs, while maintaining their quality and safety. Lack of contracts with urban wholesalers and retailers is yet another big risk for the farmers [24,25].

Financial risk encompasses risks associated with securing adequate funding to meet various operational and strategic needs within the livestock sector. These include ensuring the availability of working capital, procuring necessary inputs, obtaining livestock insurance, financing the marketing of livestock products, and managing household expenditures during lean seasons [26]. In addition, financial risk also includes the high interest rate and collateral requirement for credit [27]. Nath et al. [28] found that inadequate finance is a major constraint in the expansion of poultry farming. In the livestock sector, credit is crucial for the purchase of modern technologies, which helps to enhance productivity, but financial institutions are reluctant to sanction loans to livestock farmers due to the high risk of price volatility involved [29]. Dido [30] also found that business finance and lack of cash are major constraints for livestock farmers. As a result, small farmers experience the risk of poor animal health and production inefficiency [31].

Operational and production risks are one of the greatest challenges to livestock farmers. These include risks related to transportation and logistics, poor enterprise selection, an inadequate quality control system, errors in forecasting and planning, mechanical failure of equipment, use of outdated techniques, and operational disturbance caused by climate change, diseases like foot and mouth disease (FMD), and unavailability of feed and fodder [32,33,34].

2.2. RMSs and Their Adoption Factors

Some RMSs in the livestock supply chain are reported in the literature, such as insurance, vaccination, and advisory services [12,35,36]. For instance, studies suggest that insurance is one of the main RMSs in the livestock supply chain that has a strong ability to support farmers in coping with risk [37,38]. Livestock insurance is considered a risk management tool that is designed to stabilize farm income [39]. Insurance helps farmers to safeguard from economic loss in case of the death of livestock [40].

Several factors affect decisions on whether to adopt RMSs in the livestock supply chain. Rijpkema et al. [41] suggested three sets of characteristics that determine the adoption of livestock insurance: external factors, availability of resources, and socio-economic characteristics of farmers. Farmers with skills, knowledge, and a higher level of education have a better understanding and tend to purchase livestock insurance products [42,43]. Khan et al. [44] found that awareness is a significant driver for farmers’ participation in the livestock insurance market. Income is an important factor in the adoption of cattle insurance, as it provides sufficient purchasing power to afford the premium costs, and prosperous farmers are more likely to be able to afford and adopt such insurance [45].

Farmers of greater ages and with limited understanding of insurance tend to be unwilling to pay for livestock insurance [46]. The education level of the head of the household positively impacts the insurance of livestock. Respondents with higher education levels are more interested in participating in the livestock market [42,45,47]. Knowledge about insurance among farmers is determined by membership of self-help groups, associations like farmer’s producer organizations, and cooperative societies [48]. Weak participation in the livestock insurance market is due to the unavailability of information about government schemes, lack of education and infrastructure, income levels, low risk aversion, and diversification of farm enterprises [38]. Formal and informal education with extension information positively influences the adoption of agricultural insurance [49]. Social grouping enriches trust, ideas, and information exchange [50]. Castellani and Vigano [46] reported in their study that off-farm income and the number of livestock negatively impact the adoption of formal insurance because both have inherent risk mitigation mechanisms.

Castellani and Vigano [46] point out that investing in livestock is always risky because of frequent epidemics and drought situations. To overcome these problems, Noordhuizen [51] recommended that vaccination is the key to high livestock production. Vaccines have been used as an RMS to control animal diseases, and they have been shown to be cost-effective [35,52]. They have significant value as a strategy against FMD. Various countries have attained an FMD-free status due to the adoption of corresponding vaccination for livestock [53]. To avoid the issues of parasites in livestock, vaccines offer a promising solution [54]. According to Layton et al. [55], livestock vaccination practices are critical for both animal and human health, ensuring both economic value and lowering the possibility of spillover diseases in humans. Large FMD vaccination programs were well-received by farmers and led to sufficient disease suppression, with income increasing from large ruminant sales as a result [56].

Extension and advisory services help strengthen the knowledge about inputs used, reduce information asymmetry, improve access to tangible and intangible resources, and facilitate efforts to solve problems in livestock farming [57]. They help livestock farmers to be resilient in the event of an external shock or uncertainty. McCole et al. [58] pointed out that the diffusion of knowledge is imperative for the capacity building of farmers. Lack of access to extension and advisory services led to difficulty in obtaining technical information [31]. In the context of India, Thakur and Chander [36] found that the delivery of livestock extension services is limited [36]. The implementation of proficient and productive extension services enhances the practice of livestock farming in India [59].

3. Research Methodology

A three-phase framework is proposed to identify the risk in the livestock supply chain, and determinants of the adoption of RMSs are illustrated in Figure 1. In the first phase, qualitative techniques are used for the identification of the risks and RMSs related to the livestock supply chain. The qualitative technique includes a two-stage process, a (1) literature review and (2) expert feedback, to finalize the risks and associated RMSs in the livestock supply chain. In the second phase, evaluation of the RMSs is performed using an analytical hierarchical process (AHP), which is a well-known multi-criteria decision-making (MCDM) method. As per the steps of the AHP, experts’ inputs are provided in the form of a pairwise comparison matrix for risks as well as RMSs. This matrix is used to prioritize the RMSs. Based on the results of the AHP, the top three RMSs are recognized for use in the third phase of analysis. In the third phase, the difference in the socio-demographic profiles of adopters and non-adopters of RMSs is explored by applying the Chi-square test. Furthermore, major factors affecting the adoption of RMSs are also identified using the multivariate probit (MVP) model. A detailed description of the applied method is provided in the specific section.

4. Data Analysis

This section presents the three-phase methodology for the identification and analysis of risks and RMSs. The first phase deals with the identification of the risks and RMSs for the livestock supply chain, and the second phase deals with the prioritization of the identified risks and RMSs. Further, the third phase analyzes the factors that are responsible for the adoption of RMSs.

4.1. Phase I: Risks and RMSs for Livestock Supply Chain

In the first phase, the two-step process of the literature review and expert feedback was performed for the identification of the risks and RMSs related to the livestock supply chain. The Scopus database was used for the identification of articles to conduct the literature review. The authors used the search keywords “risk”, “livestock supply chain”, “RMS”, and “livestock sector”. These keywords were searched using Boolean operators (OR and AND) in the Scopus database. A total of 188 articles were found in the first stage of the keyword search. The titles and abstracts of the articles were reviewed by two of the authors to assess their relevance to the research objectives. A total of 152 articles were found to be misaligned with the research objectives and were therefore excluded from further analysis. Consequently, 36 articles were finally selected, because they broadly discuss various risks and their associated mitigation strategies. After finalizing the relevant articles, the authors conducted a full-text review for the identification of the risks and RMSs related to the livestock supply chain. These identified risks and RMSs were refined based on expert feedback. These risks and RMSs were finalized through discussion with the expert panel. In this study, the expert panel was formulated with seven experts working in livestock firms at the managerial level, two agribusiness consultants, and three academics, and their details are provided in Appendix A. These experts have significant knowledge and practical experience about livestock supply chain risks and associated issues. In this manner, the authors finalized the risks and RMSs related to the livestock supply chain.

4.2. Phase II: Evaluation of the Risks and RMSs

Analytical Hierarchical Process (AHP)

The analytic hierarchical process (AHP) is among the prominent methods used to solve MCDM problems [60]. The AHP is widely applied in different fields such as supply chain management, sustainability, food supply chain management, and the livestock supply chain [61]. MCDM is a systematic approach often used in complex decision-making scenarios, particularly where multiple conflicting attributes or criteria exist [61,62]. The AHP method is built on three primary stages: (i) hierarchical structural development; (ii) a pairwise comparison of the factors; and (iii) a synthesis of priorities [63]. The steps of the AHP are as follows:

Step 1: Creation of the hierarchal structure

The hierarchical structure is built to organize the factors (i.e., risks) and alternatives (i.e., RMSs). After developing the hierarchal structure of risks and RMSs, experts were invited to make pairwise comparisons among them employing a 9-point scale. Pairwise matrices were developed by the experts.

Step 2: Determination of pairwise comparison decision matrix

In this step, the pairwise comparison of risks and RMSs served to determine the relative weights of risks and RMSs. Specifically, each risk was compared by the expert panel against every other risk to determine its relative weight. Similarly, each RMS was compared with the others under the same risk category to assess its relative importance. This comparison was carried out using a nine-point scale (as shown in

Table 1. Standardized comparison scale of nine levels.

Definition	Value
Equally important	1
Moderately important	3
Strongly important	5
Very strongly important	7
Extremely important	9
Intermediate values	2, 4, 6, 8

), which is used in the AHP. The scale ranges from 1 (equal importance) to 9 (extremely important), allowing evaluators to indicate the degree to which one element is preferred over another.

Step 3: Calculation of criteria weight

The risk weight and RMS weight are calculated using the following Equation (1) [60]:

AW = λ_maxW

(1)

where A is the priority matrix, W is the importance weight of a risk or RMS, and λ_max is the maximum eigenvalue of matrix A.

Step 4: Calculation of the consistency

Consistency is required for a reliable ranking of the risks and RMSs. To measure the consistency of the matrix [A], a consistency ratio is used. The consistency ratio (CR) is defined as follows [60]:

$$\mathit{CR}={\displaystyle \frac{\mathit{C}\mathit{I}}{\mathit{R}\mathit{I}}}$$

(2)

where CI is the consistency index and RI is a random index. The CI is calculated using Equation (3) [60]:

$$CI={\displaystyle \frac{{\lambda }_{max} - n}{n - 1}}$$

(3)

Furthermore, the value of the RI depends on the different counts of factors, and the value of the RI is shown in

Table 2. The relationship between RI value and count of criterion.

N	1	2	3	4	5	6	7	8	9
RI	0	0	0.58	0.90	1.12	1.24	1.32	1.41	1.45

.

If the CR is less than 0.10, the result is acceptable, and experts’ input is largely consistent. Otherwise, we repeat the process from the beginning.

4.3. Phase III: Analysis of Determinants for Adoption of RMSs

4.3.1. Data Collection and Survey Instruments

Phase 3 of this study was based on data collected through the primary survey. A sampling frame of livestock farmers was not available, which restricted us to using the adoption probability sampling method. In this case, non-probability was the best-suited sampling method. Therefore, non-probability sampling was used to collect data. Among the types of non-probability sampling, purposive sampling was applied in data collection. Data were collected from three main districts, namely Aligarh, Mathura, and Moradabad of Uttar Pradesh, India. Due to constraints in human resources, finances, and time, data were collected from only these three districts, which are nevertheless among the main agricultural regions. During the Green Revolution, significant emphasis was placed on Western Uttar Pradesh, and these three districts fall within that region. The population of these three districts combined is above 10 million. Therefore, these districts are very suitable to represent the entire population of the region. With limited financial and human support, data collection from specific livestock farmers was a challenging task. The sample data were collected from only 424 farmers, who practiced livestock farming as their main business and agreed to respond to a structured questionnaire. As the population size was above 10,000, power test analysis revealed that 384 was the minimum sample size; therefore, 424 was a suitable sample size to represent the selected population. To ensure a good participation of the respondents, face-to-face interviews were conducted at the convenience of the livestock farmers in terms of time. To avoid errors regarding language, people with a good understanding of the regional language were involved in the data collection process. The survey was conducted in the last quarter of 2024 through a structured questionnaire.

The structured questionnaire has two parts: the first section comprises information related to the socio-demographic profiles (age, operational holding, income, education level, social category, family type, and economic status) of the respondents, while the second part includes questions about the livestock, farm characteristics, and RMSs adopted by the farmers. Livestock farmers were asked to respond on the adoption of RMSs as to whether they adopt them in order to avoid risk in the livestock supply chain.

4.3.2. Analytical Approach

The collected data was digitized using SPSS 24 version spreadsheets and Stata 17.0. The present study explored the difference between adopters and non-adopters of RMSs. Therefore, the Chi-square test was used to examine whether there was a significant difference in the frequency distribution of various socio-demographic variables (i.e., age, operational holding, income, education level, social category, family type, and economic status) across adopters and non-adopters of RMSs (i.e., livestock insurance, vaccination, and advisory services).

MVP was used to determine the factors influencing the adoption of RMSs. The MPV is appropriate when the outcome variable has a categorical choice and independent variables are a mixture of continuous and discrete predictors. MVP can account for shared underlying interrelations among any number of choices and reveals whether any two choices are complementary or substitutionary in their effect. MVP is one of the main techniques that predict the probability of a dichotomous outcome with respect to the set of predictor variables that may affect the outcome. In the present study, the adoption of livestock insurance, vaccination, and advisory services were taken as dependent variables while the age of the respondents (AGE), family size (FAMSIZE), social category of respondents (SOC), economic status (ECON STATUS), type of house, level of education (EDU), income of the farmers (INC), operational holding (OH), perceived risk of foot and mouth disease (RISKFMD), and availability of organized input supply (OIS) were taken as independent variables. Adoption of livestock insurance, vaccination, and advisory services are recorded on a binary scale. If a farmer adopts an RMS to avoid risk, they are labeled 1, while non-adopters are given 0. The empirical form of the regression model is as follows:

$$\begin{gathered} y_i^{*}= \alpha +{\sum }_{i=1}^n{\beta }_i{X}_i+{\epsilon }_i \\ { y}_i=\left\{\begin{array}{c}0 if y_i^{*} \le 0\\ y_i^{*} if y_i^{*}>0\end{array}\right. \end{gathered}$$

(4)

where $y_i^{\mathrm{*}}$ is the unobserved response of RMSs (which cannot be observed directly, but observed as binary), Xi is the matrix of predictor variables (age, family size, social category, economic status, house type, level of education, income of farmers, operational holding, perceived risk of foot and mouth disease, and organized input supply), β_i is the matrix of parameter estimates of explanatory variables, and α and ${\epsilon }_i$ are the intercept and error term in the model.

5. Results and Discussion

5.1. Phase I: Identification of Risks and RMSs

The major risks of the livestock supply chain and their RMSs were identified through the literature review and validated through expert feedback. After discussion with the expert panel, three risk dimensions and eight RMSs were selected as most relevant for the sample population.

Table 3. Risks and RMSs for livestock supply chain.

	Risks/RMSs	Codes	References
Risks	Financial risks	FR	[64]
	Production and operation risks	PR	[65]
	Market and price risks	MR	[66]
RMSs	Loan from formal/informal sources	ST1	[67,68]
	Livestock insurance	ST2	[69]
	Maintenance of breeding bull	ST3	[70]
	Building of animal shed	ST4	[71]
	Diversification of products	ST5	[72]
	Vaccination of animal	ST6	[73]
	Organized input supply	ST7	[74]
	Advisory/extension services	ST8	[75]

shows the risk dimensions and associated RMSs.

5.2. Second Phase II: Prioritization of RMSs

The second objective of this study was to determine valuable strategies to mitigate the livestock supply chain risk. In order to do this, the identified three major risks and eight RMSs were used to develop a hierarchical structure for the livestock supply chain (Figure 2). The hierarchical structure contained three levels: the top level of the hierarchy indicates the evaluation of RMSs, followed by the livestock supply chain risks, and finally, the bottom level of the hierarchy signifies RMSs.

After the development of the hierarchical structure, an expert panel was formed for the prioritization of the risks and RMSs. Furthermore, the experts were requested to provide their responses in the form of pairwise comparison matrices for the risks and associated RMSs. The final pairwise comparison matrices of risks are provided in

Table 4. Pairwise comparison matrix for risks.

Risk	FR	PR	MR	Weight
FR	1	1	1	0.326
PR	1	1	2	0.416
MR	1	0.5	1	0.260
CR = 0.046

.

In addition, this pairwise comparison matrix was used to calculate the importance weight of three major risks, which are provided in Table 4. Similar to the pairwise comparison matrices of risks, experts also provided pairwise comparison matrices for the RMS for each risk, as provided in

Table 5. Pairwise comparison matrix of RMSs for financial risk.

Strategies	ST1	ST2	ST3	ST4	ST5	ST6	ST7	ST8	Weight
ST1	1	0.5	5	3	1/3	1/2	1	1/4	0.09
ST2	2	1	3	4	3	1/3	3	1	0.177
ST3	1/5	1/3	1	1/3	1/2	1/4	1/5	1/5	0.033
ST4	1/3	1/4	3	1	2	1/4	1/2	1/4	0.062
ST5	3	1/3	2	2	1	1/2	1/2	1/4	0.073
ST6	2	3	4	4	2	1	2	1/2	0.200
ST7	1	1/3	2	5	1	½	1	1/2	0.111
ST8	4	1	4	4	4	2	2	1	0.254
CR = 0.093

,

Table 6. Pairwise comparison matrix of RMSs for production risk.

RMS	ST1	ST2	ST3	ST4	ST5	ST6	ST7	ST8	Weight
ST1	1	3	1/4	1/5	1/4	1/6	1/2	1/5	0.04
ST2	1/3	1	1/2	1/2	1/3	1/2	1/2	1/6	0.044
ST3	4	2	1	1	2	1/3	3	1/4	0.121
ST4	5	2	1	1	1/2	1/2	1	1/3	0.1
ST5	4	3	1/2	2	1	1	2	1/2	0.139
ST6	6	2	3	2	1	1	5	1	0.213
ST7	2	2	1/3	1	½	1/5	1	1/2	0.067
ST8	5	6	4	3	2	1	2	1	0.277
CR = 0.075

and

Table 7. Pairwise comparison matrix of RMSs for market and price risk.

RMS	ST1	ST2	ST3	ST4	ST5	ST6	ST7	ST8	Weight
ST1	1	1/2	2	1/2	1/2	1	1/2	1/6	0.065
ST2	2	1	2	2	3	3/2	1	1/5	0.135
ST3	1/2	1/2	1	1	1	1/2	2	1/2	0.081
ST4	2	1/2	1	1	1	1/3	1/4	1/4	0.065
ST5	2	1/3	1	1	1	1/2	1/3	1/6	0.064
ST6	1	2/3	2	3	2	1	3	1/2	0.144
ST7	2	1	1/2	4	3	1/3	1	1/4	0.106
ST8	6	5	2	4	6	2	4	1	0.340
CR = 0.096

.

The importance weight of each RMS for financial risk, production risk, and market and price risks was calculated using Equations (1)–(3). The consistency ratio (CR) was calculated and fell within the acceptable level (i.e., < 0.10), which allowed us to infer that the results were reliable. After calculating the importance weights of risks and RMSs, the final weights and ranks of the RMSs were determined and are shown in

Table 8. Final importance weight and rank of each RMS for the livestock supply chain.

RMS	FR (0.326)	O&P (0.416)	MR (0.26)	Final Weight	Final Rank
ST1	0.09	0.04	0.065	0.06288	8
ST2	0.177	0.044	0.135	0.111106	3
ST3	0.033	0.121	0.081	0.082154	6
ST4	0.062	0.1	0.065	0.078712	7
ST5	0.073	0.139	0.064	0.098262	4
ST6	0.2	0.213	0.144	0.191248	2
ST7	0.111	0.067	0.106	0.091618	5
ST8	0.254	0.277	0.34	0.286436	1

.

The results of the AHP show that ‘extension/advisory services’, ‘vaccine’, and ‘livestock insurance’ are the top three RMSs for the livestock supply chain. The results indicate that these RMSs are the most likely to be effective in mitigating the impact of adverse events. The adoption of these RMSs is dependent on certain factors, which were examined in the next phase of this study.

5.3. Phase III: Analyzing the Adoption Factors of RMSs

5.3.1. Socio-Economic Profile Analysis of Adopters and Non-Adopters of RMSs

The socio-economic profiles of adopters and non-adopters of RMSs are presented in

Table 9. Difference in profiles of adopters and non-adopters of RMSs.

		Livestock Insurance			Vaccine			Extension Services
Socio-Demographic Variables	Descriptions of Variables	Adopters	Non-Adopters	Chi-Square	Adopters	Non-Adopters	Chi-Square	Adopters	Non-Adopters	Chi-Square
Age	Age of the Farmers in Years
Less than 30 Years		16.3	15.6	χ2 = 0.823	17.4	12.8	χ2 = 2.058	17	15	χ2 = 0.599
30 to 50 Years		61.8	66	p = 0.663	63.2	63.2	p = 0.357	61.5	65	p = 0.741
Over 50 Years		21.9	18.4		19.4	24		21.6	19.9
Operation Holding	Ownership of landholding by farmers (in hectares)
Marginal Farmers (Less than 1 Hectare)		54.1	64.5	χ2 = 9.198	51.5	72	χ2 = 15.455	50	65.5	χ2 = 10.474
Small Farmers (1–2 Hectares)		31.8	18.4	p = 0.027	31.1	18.4	p = 0.001	32.1	22.3	p = 0.015
Medium Farmers (2–4 Hectares)		11.3	12.1		13	8		13.8	9.2
Large Farmers (Over 4 Hectares)		2.8	5		4.3	1.6		4.1	2.9
Income in INR (Yearly)	Yearly income of livestock farmers in rupees
Less than 50,000		22.3	31.9	χ2 = 12.631	22.7	32	χ2 = 13.179	31.2	19.4	χ2 = 10.299
50,000 to 100,000		39.6	46.1	p = 0.006	39.5	47.2	p = 0.004	35.3	48.5	p = 0.016
100,000 to 200,000		26.1	17		25.8	16.8		23.4	22.8
Over 200,000		12	5		12.0	4.0		10.1	9.2
Education Level	Highest level of education obtained by the farmers
Illiterate		9.9	19.9	χ2 = 18.971	10.0	20.8	χ2 = 15.976	11.5	15	χ2 = 3.288
Primary and upper primary		30.4	41.8	p = 0.001	32.1	39.2	p = 0.003	33.9	34.5	p = 0.511
High school		21.9	14.9		22.4	12.8		20.2	18.9
Intermediate		22.6	14.2		20.4	18.4		18.8	20.9
Graduate and above		15.2	9.2		15.1	8.8		15.6	10.7
Social Category	Which social category do farmers belong to?
General		22.3	12.1	χ2 = 6.626	17.8	21.6	χ2 = 2.892	14.7	23.3	χ2 = 5.872
OBC		69.1	79.4	p = 0.036	74.8	67.2	p = 0.235	75.1	69.9	p = 0.053
SC		8.5	8.5		7.4	11.2		10.1	6.8
Economic Status	What is the economic status of the farmers in term of poverty level?
BPL		30.4	57.4	χ2 = 28.861	37.1	44.8	χ2 = 2.176	32.1	47.1	χ2 = 9.952
APL		69.6	42.6	p = 0.000	62.9	55.2	p = 0.14	67.9	52.9	p = 0.002

. These were examined across three RMSs, i.e., livestock insurance, vaccination, and extension advisory services. The difference in profile was examined for age, operation holding, income, education level, type of family, social category, and economic status. The response showed that a significant proportion of farmers (66%) had adopted livestock insurance as part of their RMSs in livestock farming. The results showed that most of the respondents belonged to the age group of 30–50 years. Interestingly, no significant difference (χ² = 0.823, p = 0.663) was found in the ages of adopters and non-adopters. In the case of the operational holding, the chi-square test revealed that the landholding sizes of insurance adopters and non-adopters were significantly different (χ² = 9.198, p = 0.027). Marginal farmers were more highly represented among non-adopters of livestock insurance as compared to adopters. In the economic dimension, the results indicated that the majority of the farmers were from the income group of 50,000 to 100,000 INR per annum (~USD 600 to 1200), and a significant difference was found (χ² = 12.631, p = 0.006) between the insurance adopters and non-adopters. Regarding education, livestock insurance adopters had a better education level compared to non-adopters (χ² = 18.971, p = 0.001). Further, a statistically significant difference was found among social categories (χ²= 6.626, p = 0.036), where, in the general category, livestock insurance adopters were more highly represented than non-adopters. As far as economic status is concerned, the Chi-square test revealed that respondents from above the poverty line were more highly represented among adopters of livestock insurance as compared to people from below the poverty line (χ² = 28.861, p = 0.000).

Similarly, respondents’ profile summaries of vaccine adopters and non-adopters are also illustrated in Table 9. Chi-square test results revealed a significant difference in operational holding, income, and educational status across the adopters and non-adopters of vaccines as an RMS in the livestock supply chain. The chi-square statistic (χ² = 15.455, p = 0.001) showed a significant difference in the operational holdings of adopters and non-adopters of vaccines. The percentage of non-adopters was higher in the marginal landholding class. Across the insurance adopters and non-adopters of vaccines, a statistically significant difference was found (χ² = 13.179, p = 0.004) in income level. Vaccine adopters belonged to relatively more upper-income groups than non-adopters of vaccines. Regarding education, the chi-square test revealed that the education levels of vaccine adopters and non-adopters were significantly different (χ² = 15.976, p = 0.004). Farmers with higher education levels more often used vaccination as an RMS in the livestock supply chain as compared to less educated farmers, and we postulate that the reason may be that education improves awareness and decision-making ability.

The difference in the profiles of adopters and non-adopters of agricultural extension advisory services was examined. Of the total 424 farmers, 218 were adopters of extension/advisory services. χ² analysis showed no difference in age pattern distribution across the different age categories for adopters and non-adopters (χ² = 0.599, p = 0.741). As far as the operational holding is concerned, the majority of the farmers were marginal and small. The chi-square statistic (χ² = 10.474, p = 0.015) indicated a significant difference in the operational holdings of adopters and non-adopters of extension advisory services. The percentage of farmers of medium and large farms was higher for adopters as compared to non-adopters. Regarding income, the results of the χ² test indicated a significant difference in the income levels of adopters and non-adopters (χ² = 010.299, p = 0.016). Regarding economic status, the analysis indicated that adopters of advisory services were significantly more highly represented in the Above Poverty Line (APL) category as compared to the Below Poverty Line (BPL) category. Surprisingly, no difference was found in education level across the adopters and non-adopters of advisory services. From the above analysis, it was found that hypothesis H1, which assumed no difference in the adoption of risk management strategies across the socio-demographic profiles of farmers, could be partially rejected.

5.3.2. Determinants of RMSs

Some studies have examined the factors affecting the adoption of RMSs in the livestock supply chain for different countries [12,38]. In the present study, the three top RMSs were determined based on the AHP—livestock insurance, vaccination, and extension/advisory services. Parameter estimates of the regression model are presented in

Table 10. Determinants for the adoption of RMSs.

	Livestock Insurance				Vaccine				Extension Services
Variables	Coef.	Std. Err.	Z-Score	p > |z|	Coef.	Std. Err.	Z-Score	p > |z|	Coef.	Std. Err.	Z-Score	p > |z|
AGE	−0.416	0.140	−2.97	0.003	−0.280	0.138	−2.02	0.043	−0.070	0.131	−0.54	0.591
FAMSIZE	−0.024	0.019	−1.25	0.21	−0.006	0.020	−0.31	0.760	−0.067	0.018	−3.68	0.000
SOC (OBC = 1, Otherwise = 0)	−0.309	0.164	−1.89	0.059	0.249	0.153	1.62	0.105	0.089	0.150	0.59	0.552
ECON STATUS (APL = 1, BPL = 0)	0.553	0.147	3.75	0.000	−0.048	0.151	−0.32	0.750	0.316	0.141	2.24	0.025
House Type (Furnished = 1, Otherwise = 0)	0.048	0.147	0.33	0.743	−0.041	0.151	−0.27	0.785	−0.075	0.140	−0.54	0.592
EDU (< High School = 0, > High School = 1)	0.275	0.147	1.87	0.061	0.292	0.147	1.99	0.046	0.140	0.139	1.01	0.314
INC	0.413	0.161	2.57	0.010	0.358	0.158	2.26	0.024	−0.474	0.159	−2.97	0.003
OH (Above 1 ha = 1, Otherwise = 0)	0.068	0.149	0.46	0.647	0.513	0.150	3.43	0.001	0.460	0.138	3.32	0.001
RISKFMD (Yes = 1, No = 0)	0.399	0.148	2.69	0.007	0.249	0.148	1.68	0.093	0.331	0.145	2.29	0.022
OIS (Yes = 1, No = 0)	−0.067	0.140	−0.48	0.633	0.042	0.139	0.3	0.763	0.042	0.131	0.32	0.749
Constant	0.046	0.278	0.17	0.869	−0.182	0.277	−0.66	0.511	0.193	0.265	0.73	0.466
Wald ꭓ2(30) =	141.77;
Log likelihood = −733.41783;
Prob > chi2 = 0.0000

. First, factors affecting the adoption of livestock insurance will be discussed. Age, social category, economic status, educational level, income level, and risk of FMD significantly affect the adoption of livestock insurance as an RMS in the livestock sector in India. The negative coefficient of age indicated that young farmers were more likely to adopt livestock insurance, which was significant at 5 percent level (β = −0.416, p = 0.003). Our findings align with the results reported in [12]. The significant negative coefficient of the social category reflected that general-class people were more likely to adopt an RMS (β = −0.309, p = 0.059). Households above the poverty line were the main adopters of insurance, as revealed by the MVP analysis (β = 0.553, p = 0.000). The estimated coefficient of multivariate probit for the educational level was positive and significant (β = 0.275, p = 0.061), implying that farmers with an educational level above high school are more likely to adopt insurance to avoid risk. Income is always a determining factor in the purchase of any risk mitigation tools, as observed in the case of livestock insurance, as per [37]. Probit regression estimates revealed that the adoption of livestock insurance is significantly determined by the farmers’ income (β = 0.413, p = 0.010), in that those with an income above INR 50,000 per annum (~600 USD) are more likely to purchase insurance to mitigate risk in the livestock sector. The results of the present study regarding income are in line with the findings in [38], which concluded that the higher-income group is more likely to adopt livestock insurance. MVP analysis revealed that livestock insurance as an RMS is also significantly determined by the perceived risk of foot and mouth disease (β = 0.399, p = 0.007). From the above analysis, it can be concluded that younger farmers belonging to the general category and higher income group with an education level above high school are more likely to adopt livestock insurance as a risk management strategy; further, the adoption of livestock insurance is relatively high among those farmers who perceive FMD as a risk.

The results of the adoption of vaccines for livestock as an RMS are also presented in Table 10. Factors affecting the adoption of vaccines are the age of the farmers, level of education, income, operational holding, and perceived risk of FMD. The regression coefficient of the MVP analysis is negative and significant for age (β =−0.280, p = 0.043), revealing that vaccination as a risk mitigation strategy is more likely to be adopted by lower-age farmers. Vaccine adoption for livestock is significantly influenced by the education level of farmers (β =0.292, p = 0.046). This implies that farmers with high school education get their livestock vaccinated. The parameter estimates of the regression show a positive and significant coefficient for income (β = 0.358, p = 0.024), suggesting that vaccination of livestock increases with income. The regression results on operational holdings reveal that farmers of large farms are more likely to adopt vaccination for their livestock as compared to those of small and medium farms (β = 0.513, p = 0.001). The perceived risk of FMD significantly influences the adoption of vaccines. MVP analysis exhibits that farmers who perceive a high risk of foot and mouth disease are more concerned about the vaccination of animals to mitigate risk. The estimated value of the MVP revealed that younger farmers with an education level above high school, higher income groups with operational holdings over one hectare, and those who perceive the risk of FMD are more likely to adopt vaccines as an RMS.

Furthermore, parameter estimates show that the family size, economic status, income, operational holding, and perceived risk of foot and mouth disease have statistically significant influences on the use of advisory/extension services as an RMS. The MVP regression coefficient of family size is negative and significant (β = −0.067, p = 0.000), showing that small families are more likely to use advisory/extension services to avoid risk in the livestock sector. The results of multivariate probit regression analysis revealed that farm households above the poverty line often use advisory/extension services as a risk mitigation strategy (β = 0.316, p = 0.025). The regression coefficient for income is negative and significant, which implies that higher-income farmers are less likely to adopt extension services for risk mitigation (β= −0.474, p = 0.003). Farmers with large landholdings are likely to use advisory/extension services. The parameter estimates of MVP regression for the perceived risk of foot and mouth disease showed that advisory/extension services were used by the farmers who perceived FMD as a risk (β = 0.331, p = 0.022). This regression model also had a reasonable fit. The model summary of MVP showed that the likelihood ratio test was significant at 5 percent, implying that the null hypotheses that all the rho values (ρ12 = ρ 13 = ρ 23 = 0) were jointly equal to zero or all three strategies were independently determined could be rejected. This revealed that the decision to adopt all three RMSs was interdependent. The results of the model showed that the correlations among the RMSs of vaccination, livestock insurance, and extension and advisory services were positive and significant, implying that if the farmers adopt one strategy, it is more likely that they might adopt another strategy. This can be attributed to the observation of the benefits derived from one strategy, which motivates the farmer’s willingness to adopt an additional RMS. It can be concluded from the above regression analysis that farmers who perceive foot and mouth disease as a risk, belong to smaller family sizes and lower income groups, and have operational holdings above one hectare are more likely to adopt extension services as risk management strategies. From the estimated regression values, it is revealed that hypothesis H2, which assumes that the socio-demographic profiles of the farmers do not affect the adoption of RMSs, can be rejected for most socio-demographic indicators. Similarly, H3, which assumes that farm factors do not affect the adoption of risk management strategies, can be partially rejected.

From the above results, it is revealed that younger farmers are generally more inclined to adopt innovative products; therefore, they are more willing to engage with new-generation RMSs like livestock insurance, vaccination, and extension/advisory services. Young farmers have a greater risk-taking propensity; therefore, they generally participate in the latest management strategies. Education empowers the individual in terms of decision-making and awareness, and they can easily understand the benefits of insurance. These results may be attributed to the fact that higher-income farmers have more purchasing power, so they can easily participate in the insurance market. Animal disease is a severe problem for farmers; on many occasions, it leads to the death of animals. Therefore, the farmers perceive the risk of diseases like FMD and purchase insurance to avoid financial loss. Educated farmers tend to have greater awareness of vaccination practices and are less likely to believe the negative myths regarding vaccination, such as the misconception that vaccination decreases productivity—a belief more commonly observed among less educated or uneducated farmers. Vaccination is widely regarded as the most effective method to prevent diseases. Farmers concerned about potential animal losses due to diseases prioritize vaccinating their livestock to safeguard their health and ensure productivity. Vaccination is a costly strategy; only higher-income farmers can afford it. Prosperous farmers use extension and advisory services as they follow the scientific method of farming, and they acquire timely information and updated techniques for production. Therefore, they use advisory services. Extension/advisory services give farmers the confidence to make decisions by providing a package of suggestions regarding vaccination, price information, and production techniques. Farmers facing FMD issues among their animals generally opt for advisory services to avoid a health risk.

6. Implications and Recommendations

This study suggests that practitioners in livestock farming prioritize three key areas: livestock insurance, animal vaccinations, and seeking advice from extension agencies. The proposed risk mitigation framework aids the farm manager in developing and assessing the RMSs they may employ to effectively manage and mitigate risks. The significance of vaccination needs to be communicated to low-income, illiterate farmers and those of small and marginal farms, and the government should arrange vaccination at a subsidized rate. The findings indicate that families who are above poverty line are adopting extension services as a risk mitigation strategy in livestock farming. India possesses a robust extension services infrastructure, which should be directed towards economically disadvantaged households (BPL families) to encourage their adoption of these services, hence mitigating the risks associated with livestock farming. Insurance companies should design low-cost insurance products so they can target low-income livestock farmers.

The findings of the present study have some practical implications for insurance service providers on how to design appropriate products and marketing strategies. Moreover, this research may assist policymakers in formulating policies, plans, and strategies for vaccination programs and the dissemination of advisory services. Older farmers should obtain sufficient knowledge of RMSs to mitigate potential risks in the livestock sector. Advisory services should be readily accessible and affordable so that even farmers of marginal and small farms can avail themselves of them. In addition, uneducated livestock farmers should be targeted to encourage the adoption of RMSs. Finally, this study demonstrates that there is the scope for further research on the topic to better generalize the results. Future research may include some psychological factors as explanatory variables to improve the explanatory power of the regression model. An empirical model could be based on the theory of planned behavior (TBP), allowing researchers to understand the intentions of farmers in adopting RMSs in the livestock supply chain and thus to reach better conclusions. The results of this study could be validated using multiple case studies in future research. Further, the proposed model could also be used to evaluate the RMSs for other types of supply chains, such as the meat supply chain.

7. Conclusions, Limitations, and Future Scope

Farmers, policymakers, and other stakeholders are interested in identifying the risks in the livestock supply chain, the RMSs, and the factors affecting their adoption. Therefore, this study examined the major RMSs and factors affecting the adoption of these strategies across the livestock supply chain. From the literature survey, the eight most relevant RMSs were identified, namely (1) s loan from formal/informal sources, (2) livestock insurance, (3) maintenance of breeding bulls, (4) building of animal sheds, (5) diversification of products, (6) vaccination of animals, (7) organized input supply, and (8) advisory/extension services. The results of the AHP suggest that livestock insurance, vaccination of livestock, and advisory/extension services are the major RMSs. Furthermore, from the primary survey, a difference between the socio-demographic profiles of adopters and non-adopters of these three RMSs—livestock insurance, vaccination, and advisory/extension services—was found. Adoption of RMSs is high among educated and high-income farmers of large farms. Further, parameter estimates of the multivariate probit model (MVP) reveal that the age, social category, economic status, education level, income of the farmers, and perceived risk of foot and mouth disease are the major factors affecting the adoption of livestock insurance. As far as the factors affecting the adoption of vaccination of animals are concerned, empirical estimates show that the age, economic status, levels of education and income, and perceived risk of FMD have significant implications for the adoption of vaccination as an RMS. Similarly, the family size, economic status, operational holding, income of the farmers, and perceived risk of FMD significantly increase the likelihood of farmers deciding to adopt advisory/extension services. The present study has managerial, academic, and policy implications for livestock farming, the government and policymakers, bank and insurance companies, and other extension service providers.

There are some limitations of this study. The present study used a literature review and expert inputs for the finalization of the risks and their RMSs, which may have led us to overlook some significant RMSs. The potential for this can be reduced through an extensive literature survey including other citation databases such as Web of Science and PubMed. Additionally, the experts may have been biased towards some RMSs, and care should be taken in this regard while collecting responses in the future. This limitation can be further mitigated by integrating fuzzy or grey theories with the AHP. The AHP assumes that risks and RMSs are independent, whereas, in reality, risks and RMSs are interconnected. Another limitation of this study is the collection of data from only three districts due to financial constraints. In the future, research could be carried out with an increasing coverage in terms of area and sample. Moreover, researchers may explore the determinants of the remaining five risk management strategies. In addition, a power test should be used to determine the appropriate sample size in the future.