The study was conducted in the Eastern Cape Province (ECP) in South Africa. The ECP is one of the largest provinces (first being Northern Cape Province, then ECP) and is located in the south-central part of South Africa. ECP has 168,966 km², which makes the province the second largest after the Northern Cape Province of South Africa. The province combined areas from the former Xhosa homelands of the Transkei and Ciskei, together with what was previously part of the Cape Province. The province is dominated by Xhosa tribes (which include AmaMpondo, AbaThembu, AmaMpondomise, AmaHlubi, AmaBhaca, AmaXesibe, AmaBomvana). The population of Eastern Cape Province is 7,230,000, of whom 86.3% are Black, 8.3% are Coloreds, 4.7% are White, and 0.4% are Indian/Asian.

The Eastern Cape becomes gradually wetter from west to east. The west of ECP is mostly the semi-arid Karoo, except in the far south, which is temperate rainforest in the Tsitsikamma region. The coast is generally rocky with scattered beaches. Most of the province is hilly to very mountainous, while the east from East London and Komani toward the KwaZulu-Natal border (a region known previously as Transkei) is luxurious grassland on rolling hills, scattered by deep valleys with alternating forest. The climate of the province is highly varied. The west is dry with thin rain during winter or summer, with frosty winters and hot summers. The area from Tsitsikamma to Makhanda receives more rainfall, which is also relatively evenly dispersed, and temperatures are mild. Further east, rainfall becomes more abundant and humidity increases, becoming more subtropical along the coast with summer rainfall. The interior can become very cold in winter, with heavy snowfalls occasionally occurring in the mountains. This makes the landscape of the province diverse and much productive land in the Eastern Cape, and agriculture remains important as people generate livelihoods from as the province is very poor and has a high unemployment rate (60%) and high food insecurity (78%) (Mdoda and Obi, 2019; Mdoda et al., 2024). The province practices citrus, crop, and vegetable farming, and livestock farming, and subsistence agriculture is the most dominant in the province. When it comes to livestock production, the province is mainly involved in the production of cattle, goats, and sheep, which forms the backbone of the rural livelihoods. These animals provide meat, milk, income, and social value, simultaneously serving as a form of financial security for smallholder farmers. The socio-economic importance of livestock in the ECP cannot be overstated as directly contributes toward livelihood. Hence the study investigates the effects of risk behavior in livestock farmers situated within the ECP. This area is suitable due to its large smallholder livestock production, climatic vulnerability and relevance to SA's rural development policies aimed at improving resilience and food security. Figure 2 illustrates the study sites.

The study was conducted in the Eastern Cape Province, adopting a descriptive research approach and cross-sectional research design. The study made use of a multi-stratified sampling procedure with specifically 3 stages. The unit of analysis for this study was livestock farmers. The first stage was selecting district municipalities in the province that have the potential for livestock farming and contribute to the livelihoods of households in the province. These districts were OR Tambo, Chris Hani, Amatole, and Alfred Ndzo District Municipalities. The second stage was selecting two local municipalities and wards where the study was conducted within the district. The stratified sampling was also implemented in this stage, where farmers were stratified into the type of farming they practice, which includes crops and livestock farming, with the livestock being the focal point of this study. In the last stage, random sampling was used to select the desired sample size of livestock farmers and to reduce bias.

In this case, smallholder farmers were selected based on the following criteria: (i) they owned or managed livestock as a primary or secondary livelihood activity, (ii) they operated on small plots of land (typically less than 10 hectares) using family labor and traditional production methods, and (iii) they derived part of their household income or food security directly from livestock production. Importantly, respondents were not chosen because they were already facing financial crises or risk mitigation problems; rather, they were purposively sampled within selected districts known for active livestock farming. Within these districts, random sampling ensured representation of households across different socio-economic and demographic categories, without bias toward only those experiencing problems. This approach allowed the study to capture a realistic picture of smallholder livestock practices, the risks they face, and the mitigation strategies they employ under varying conditions. Also, this ensured that the data collected reflected the experiences, perceptions and management practices of active livestock farmers in the study area.

The study obtained ethical clearance from the institution before the commencement of data collection in the selected study areas. It adopted primary and secondary data as tools for data collection. The primary data was collected using a survey and face-to-face interviews administered by the researchers and enumerators in all district municipalities selected for this study. The survey made use of structured questionnaires for the purpose of data collection in the study. A pre-test of 10% of the sample size was carried out to standardize the survey instruments, check the relevance of questions, and time taken to respond to a platform of training enumerators, as well as the study. Enumerators asked the questions to the farmers in IsiXhosa, the home language of the province. Questionnaires were translated into IsiXhosa and then retranslated back to English to review and verify the meaning. Within districts and villages, enumerators utilized their contacts and used agricultural extensions to organize local farmers for the visits. Farmers were interviewed on a voluntary basis and screened on the condition that they practice livestock farming. While secondary data was collected from published peer-reviewed articles, books, and Departmental gazettes. The collected data were farmers' profiles, farming activities practices, perception and effect of risk in their farming, challenges they faced, and strategies adopted to minimize risk in their farming practices. Before asking questions, participants were provided an informed consent form to sign in agreement with the research conduct. Questionnaire-based face-to-face interviews were agreed upon with the heads of the farmers, but in cases where they were absent at the time of the interview, data were obtained from other senior farm members (managers or acting head of Departments).

Data quality was evaluated in terms of accuracy, completeness, and missing information. The data collected from livestock farmers in the study area were coded and entered into Excel for the purpose of analysis. The data was transferred from Excel to STATA 15 for analysis. The study adopted and used three types of analytical tools for analysis: Descriptive, Binary logistic, and Propensity score. Specifically, the binary logistic regression was used to estimate factors influencing the risk behavior of livestock farmers, while Propensity score matching estimated the effects of risk behavior on livestock farming.

Respondents in the Eastern Cape had different socio-economic characteristics, including gender, age, marital status, level of education, employment status, availability of infrastructure, distance to market, and land ownership. These characteristics illustrate the conditions people in the chosen area live under. Findings from the study regarding the socio-economic conditions is represented in Table 2.

Gender plays a vital role in shaping individuals' participation and potential in agriculture, often influenced by societal expectations. This study found a balanced representation of both male and female respondents, highlighting the importance of both genders in agricultural practices. Research by Krishnan et al. (2023) supports the need for balanced gender involvement to foster sustainable development. Including both genders ensures diverse perspectives in agricultural decision-making (Karat and Shinoji, 2022). However, Noge (2025) notes that women still face unique challenges like unequal pay and higher injury risks. Thus, while equal participation reflects shared capabilities, it also uncovers areas where deeper gender-based inequalities remain.

Age is often linked to experience and knowledge, which are valuable in farming. This study found that the majority of participants (55%) were between 40 and 59 years old, followed by younger farmers aged 20 to 39 (32.5%), and older farmers over 60, making up 12.5%. These figures show that farming attracts a wide age range, with middle-aged individuals being the most active. Similarly, Singh and Pandey (2018) observed a strong presence of middle-aged and older farmers in Muzapir. Research also shows that middle-aged farmers tend to be the most productive (Song et al., 2024), while youth participation reflects growing interest in agribusiness (Simbeko et al., 2023). Older farmers, meanwhile, bring valuable experience and passion to the sector (Conway et al., 2022).

Marital status states the formal relationship of a person regarding the familial bond. These statuses are classified into single, married, divorced, or widowed. This study indicated that 47.5% of the population is formally single, allowing these households to maintain farming activities independently, which is crucial for their livelihoods. These findings are in line with a study conducted in Vhembe district by Mudzielwana et al. (2022). The married group, comprising 35%, benefits from shared responsibilities in farming, such as livestock management, which enhances productivity and ensures effective farming practices. The stability and support provided by marriage may encourage greater involvement in labor-intensive activities like farming. Moreover, the presence of a spouse can provide additional labor and resources, which are crucial for sustaining production (Lai-Solarin et al., 2024). The remaining 17.5% of the population includes widowed and divorced individuals, underscoring the integral role of farming in sustaining households regardless of marital status. These findings further indicate that marital stability may play a role in sustained agricultural engagement (Nontu et al., 2024).

Level of education denotes the amount of formal education the participant has obtained. Farming can be conducted with either the availability or the unavailability of formal education. In this study, the level of education was classified into no education, primary, secondary, and tertiary education. From the results, half of the study population had secondary education, followed by tertiary education holders at 27.5%, primary education at 20% and the remaining 2.5% having no formal education. These results demonstrate the profound capability of farming across all factorial levels of life, including access to formal education. These results are in line with findings by Mulugeta et al. (2024), who observed that education significantly affects knowledge of occupational hazards, with higher education correlating with better awareness and practices. Another study demonstrated that educated farmers are more likely to be early adopters of agricultural innovations (Rizzo et al., 2024). Pavlov and Pavlova (2024) further argued that the level of education is a key factor in forming human capital in agriculture, which is essential for ensuring food security. This indicates that higher education levels contribute to better agricultural practices and outcomes.

Employment status denotes the respondent's involvement in employment opportunities other than farming for themselves. This shows whether the household has another income source available apart from selling farm produce, grants, or remittances. This study discovered that 85% of the study population is unemployed and the remaining 15% is employed in various areas of specialization. This shows that most households that participate in livestock farming are unemployed, depending on the livestock they produce and other sources such as grants and remittances. Unemployment in agriculture, especially among youth and graduates, remains alarmingly high due to skill mismatches, poor sector perceptions, low profitability, and structural barriers (Rani and Sharma, 2024).

Availability of infrastructure refers to the availability of tangible assets that contribute toward the overall success or development of a society. These include roads, schools, hospitals, storage facilities, and many more. From this study, the available infrastructure includes schools being the one many (45%) have access to, roads (20%), other infrastructure (17.5%), hospitals (15%), and storage facilities (2.5%). This distribution highlights a significant disparity in infrastructure access. Schools are the most commonly available, suggesting a prioritization of basic education infrastructure (Opabola and Galasso, 2024). However, the limited access to roads and hospitals indicates challenges in transportation and healthcare, which can hinder economic development and wellbeing (Lawrence, 2024). The extremely low access to storage facilities (2.5%) is particularly concerning for agriculture, as it restricts the ability to store and preserve produce, leading to post-harvest losses and reduced income for farmers. The limited availability of storage facilities exacerbates the negative effects of long market distances by increasing post-harvest losses (Kaur and Watson, 2024).

Distance to market reveals the availability and efficiency of the product to be sold or the accessibility of certain requirements. Half of the study sample goes over 15 km to reach markets, followed by those having a distance below 9 km, accounting for 35% and the other 15% with the distance of between 10 km and 15 km. This shows that most farmers live in areas that are far from the markets. Traveling more than 15 km to reach markets presents significant challenges in market accessibility for a large proportion of smallholder farmers. Such long distances increase transportation costs, reduce the frequency of market visits, and limit farmers' ability to sell fresh produce promptly (Sennuga et al., 2024). The lack of storage facilities (available to only 2.5% of respondents) further compounds these challenges, leading to higher post-harvest losses. Farmers reported that perishable products such as milk, eggs, and chickens often spoiled or deteriorated in quality due to delays in reaching markets, with some produce lost entirely before being sold. For larger livestock such as cattle, sheep, and goats, the losses were indirect: animals lost body weight during long treks to markets or while waiting for buyers in informal trading points, reducing their sale value. These combined effects lower household incomes, limit reinvestment in critical inputs such as feed and veterinary care and discourage active market participation. Ultimately, the interaction between long market distances and weak infrastructure negatively impacts both farm profitability and rural livelihoods. The 35% traveling less than 9 km is likely to have better access to markets, which can facilitate more frequent sales and better integration into the supply chain. On the other hand, the 15% traveling between 10 km and 15 km are in an intermediate position, facing some challenges but not as severe as those traveling over 15 km. Market accessibility can therefore be noted as a critical factor in agricultural productivity and income. Farmers located farther from markets often face reduced profitability due to higher transport costs and limited access to market information. In regions with poor road infrastructure and long distances to markets, agricultural productivity and rural development tend to be delayed (Ezeudu and Fadeyi, 2024).

Land ownership is a crucial factor affecting the process of farming. This is because land is one of the most important factors of production in a business that deals with goods such as crops or livestock (Saqib et al., 2025). Findings illustrate that 80% of the farming population does not own the land they practice farming on. They either rent the land they are using or jointly own the land. The remaining 20% owns land which may have been purchased or left as an inheritance. Not owning the farmland underscores a widespread issue of land tenure insecurity that affects agricultural productivity, investment, and rural livelihoods. This is supported by global trends showing increasing land ownership concentration and a large share of farmers operating on leased or rented land (Georgieva, 2025). This may be due to the fact that historical land distribution patterns, land reforms, and policies have resulted in large landholdings concentrated in a few hands, forcing many farmers to lease or sharecrop. High land prices and limited access to credit prevent many farmers from purchasing land (Pun et al., 2024). Increasing land acquisition by investors and corporations (including foreign investors) can reduce land availability for smallholder ownership.

This study looked at the patterns of livestock ownership, use, and losses among farming households, focusing on five major livestock types: cattle, sheep, goats, pigs, and chickens. The findings reveal a clear trend in the multifunctional role of livestock, with most animals being retained rather than sold, consumed, donated, or lost. These findings are outlined in Table 2.

Cattle were found to be the most kept livestock, with 1,095 animals retained by households. A total of 138 cattle were sold, while 33 were consumed, 15 donated, and 32 were reported lost. This highlights the significance of cattle as long-term economic assets and symbols of wealth in rural communities (Julius and Odemero, 2024). The relatively high number of cattle kept suggests their integral role in agricultural activities such as plowing and as a form of security. The sales figures, while considerable, indicate that cattle are not typically sold frequently but are instead used as a source of income in times of financial need (Riesman, 2024). The reported losses reflect environmental challenges such as drought, disease, or theft.

Sheep followed cattle in overall importance, with 589 kept. A notable number of sheep were consumed (53) and sold (59), suggesting that sheep are more commonly used for both income generation and household nutrition. Only six sheep were donated, and one was lost. This dual function indicates a balance between the economic and subsistence roles of sheep within households (Chappa et al., 2024).

Goats also featured prominently, with 256 animals kept, 32 sold, and 18 consumed. Donations and losses were minimal. The data suggest that goats are moderately utilized for consumption and income but are primarily retained, potentially due to their resilience in harsh environmental conditions and relatively low maintenance requirements (Cooke et al., 2024).

Pigs demonstrated the highest proportion of animals sold relative to the number kept, with 38 out of 151 pigs sold. Consumption (5) and donation (0) were low, which may point to pigs being viewed more as commercial assets than sources of food or social capital. The absence of reported losses suggests pigs may be better confined to or less susceptible to loss compared to larger animals (Zeltner and Pedersen, 2024).

Chickens were also widely kept (327), with 98 sold and 35 consumed. Like pigs, no chickens were donated or reported lost. The high number sold indicates chickens are an accessible source of short-term income, while their consumption underscores their role in improving household food security (Bist et al., 2024). Their absence in the donation records may reflect cultural preferences or their lower perceived social value compared to larger livestock (Table 3).

Results reveal that livestock ownership plays a crucial role in rural livelihoods, serving various functions including income generation, food provision, social obligations, and asset accumulation. Cattle appear to be central to economic and cultural life, while species such as chickens and pigs serve more immediate financial and nutritional needs. Sheep and goats offer a hybrid model, contributing to both subsistence and market functions.

Smallholder farmers use different feeds to cater to their livestock. Figure 3 illustrates the feeding practices among farmers, showing a prevalent reliance on free-range grazing (85%), with minimal adoption of cultivated pasture (5%), hay (5%), forage (3%), and concentrated feed (2%). This distribution emphasizes a dependence on extensive grazing systems, a characteristic feature of smallholder livestock farming in Southern Africa. The preference for free-range systems is often attributed to their low input requirements and alignment with traditional farming practices (Bist et al., 2024). However, such systems are increasingly vulnerable to climatic variabilities, including prolonged droughts and erratic rainfall patterns, which adversely affect pasture availability and quality. The limited use of cultivated pastures and hay suggests constraints such as inadequate access to land, resources, and knowledge necessary for establishing and managing these feed resources (Alimi et al., 2024). Moreover, the minimal adoption of concentrated feeds may be due to their cost implications and limited availability in rural areas.

Extension visits and vaccination are the important aspects that each farmer needs to have access to. These elements assist in the upholding of the overall production. Extension visits ensure training and accumulation of more information whereas vaccination ensures that the livestock can withstand diseases and other effects associated with seasonal variations. Figure 4 illustrates the rate of these factors in terms of how much time smallholder farmers receive the visits and the frequency of vaccination for livestock.

The findings indicate that smallholder farmers in the Eastern Cape continue to rely heavily on indigenous, home-based livestock management practices to sustain their herds. Night kraaling is the most dominant practice, with 80% of respondents enclosing animals overnight as a measure to prevent theft and predation. Family labor and herding remain central to livestock management, with 70% of farmers depending on household members rather than hired labor, which reflects both cost-saving strategies and cultural traditions. The use of crop residues and kitchen waste was reported by 65% of farmers, highlighting the integration of livestock and crop production systems to maximize resource use. Over half of the farmers (55%) rely on traditional medicine and ethnoveterinary practices for disease prevention and treatment, largely due to limited access to formal veterinary services. Meanwhile, 40% practice rotational grazing, an indigenous method that helps preserve pasture and reduce overgrazing. Collectively, these practices demonstrate the resilience and adaptive strategies of smallholder farmers, but they also underscore the need for complementary modern interventions to enhance productivity and reduce risks.

The Figure 5 displays results generated from a fieldwork survey. This field survey was conducted to find out the risks that the respondents perceive as being the riskiest.

Figure 6 presents insights into the key factors perceived by respondents as making their farming businesses risky. Particularly, pests and diseases emerged as the most significant threat, mentioned by 33% of the surveyed farmers. This aligns with broader findings in the literature, which identify livestock diseases as a major constraint to productivity and profitability, particularly among smallholder farmers in Sub-Saharan Africa (Kimeli et al., 2025). Disease outbreaks, such as foot-and-mouth disease or brucellosis, often lead to severe economic losses, reduced animal productivity, and restricted market access.

The second most noted risk was lack of knowledge, reported by 17% of respondents. This highlights the persistent information gap that limits technical capacity among small-scale farmers, which constrains informed decision-making in areas such as animal nutrition, breeding, and health management. Recent studies emphasize that improved access to farmer training and livestock extension services can substantially mitigate this risk (Kalogiannidis and Syndoukas, 2024; Prabex et al., 2024; Prajapati et al., 2025).

Access to credit was identified as a major risk factor by 15% of respondents. Financial constraints often limit farmers' ability to purchase inputs such as feed, vaccines, or improved livestock breeds. Moreover, the absence of tailored credit products for livestock farming hampers investment and innovation in rural settings. Jatto and Gambo (2025) indicate that credit access, particularly through cooperative models or government-backed schemes, is essential for enabling smallholder resilience and scale.

Animal health practices were stated by 10% of the respondents, suggesting challenges in preventive health management, including vaccination and regular veterinary services. These findings echo those discussed earlier, where limited extension visits and poor vaccination uptake were evident. Without regular veterinary input, even basic preventive care becomes difficult, compounding risks related to disease burden (Owusu and Sanad, 2025).

Also, animal feed was seen as a relatively minor risk, with only 2% of farmers identifying it as a key concern. This could be due to the high reliance on free-range systems (as seen earlier), where farmers do not incur substantial feed costs. However, this perception may overlook the long-term risks posed by climate-induced changes in natural pasture availability.

Finally, the remaining 3% of respondents selected “other” as a risk factor, indicating that while the major challenges are well-defined, there is the presence of localized or context-specific issues such as theft, water shortages, or land tenure insecurity that require further exploration. These challenges play a major role in the farming process as they determine the total production a farmer can have.

Overall, these findings underscore the multifaceted nature of risk in smallholder livestock farming. Addressing these risks requires a holistic strategy involving improved veterinary services, knowledge dissemination, financial inclusion, and institutional support to enhance farm resilience and sustainability.

The Table 6 shows field results for sources of risk. The sources of risk were ranked on a Likert scale, where the respondent was able to choose the sources of risk and further indicate the extent to which the impact is severe or slight to them.

Table 6 illustrates the perceived severity of various risk sources affecting smallholder livestock farming operations, based on a Likert scale assessment by respondents. These sources include production, market, financial, institutional, human/personal, and environmental risks. The results highlight the complex and multi-dimensional nature of risk in smallholder livestock systems, with several categories showing a high proportion of farmers experiencing moderate to severe impacts.

Production risk was identified as severe by 22.5% of respondents, while 42.5% rated it as slight. This category includes uncertainties related to animal health, feed availability, and reproductive performance. The relatively low proportion of severe ratings may reflect the adaptive capacity of farmers relying on indigenous breeds and traditional knowledge. However, it still suggests that a significant segment of farmers remains vulnerable to disruptions in inputs or biological performance. Pezo et al. (2024) emphasize that unmanaged production risks, particularly diseases, can rapidly escalate into economic shocks in smallholder systems.

Market-related risk was rated severe by 35% of respondents, with an additional 17.5% reporting mild risk. This indicates concerns over price volatility, lack of access to formal markets, and limited bargaining power. Market instability has been a persistent challenge in the Southern African livestock sector, where rural farmers often lack access to real-time pricing information or reliable buyers (Morepje et al., 2024). Unpredictable demand and seasonal fluctuations further exacerbate these risks.

Financial risk was perceived as severe by 35%, mild by 20%, and medium level by 35% of respondents. This reflects widespread challenges in accessing capital for farm operations, purchasing inputs, or scaling production. As highlighted by Mishra (2024), the absence of affordable and farmer-friendly credit facilities remains a barrier to resilience and growth among rural livestock farmers. Financial constraints also limit farmers' ability to respond effectively to other risks, such as disease outbreaks or climate-related shocks.

With 42% of respondents quoting severe institutional risk and 25% citing none, the data reveals a stark divide. Institutional risk includes limited access to extension services, ineffective local governance, and a lack of support from veterinary and marketing institutions. As seen in earlier findings, most farmers lack regular contact with extension agents, which contribute to both production and animal health challenges (Arowosegbe et al., 2024). The significant number mentioning no institutional risk may reflect reliance on informal knowledge networks rather than public services.

Human and personal risk, which includes illness, aging, family labor constraints, and household-level disruptions, was rated severe by 27.5% and medium by another 27.5%. Such risks are often underestimated in livestock development planning but are crucial in smallholder systems where operations are heavily dependent on family labor. Human risks have been shown to influence decisions around herd management, marketing, and investments (Melak et al., 2024).

Environmental risk stood out, with 40% of farmers rating it as severe, and another 40% as medium. These include risks related to drought, pasture degradation, floods, and water scarcity. The findings reinforce growing concerns about climate change impacts on livestock systems. Studies have consistently shown that increased climate variability, such as drought, threatens feed availability and water access, thereby exacerbating other production risks (Sintayehu et al., 2025).

Table 7 presents results showing how various risk management strategies and socio-demographic factors influence livestock production. The results are derived from regression coefficients (β) and significance levels (p-values), indicating the strength and statistical significance of these variables in predicting livestock production outcomes.

The results reveal that avoiding risk exhibits the strongest positive influence on livestock production, with a very high β value of 33.768 and a highly significant p-value (p < 0.001). This suggests that farmers who actively engage in risk avoidance strategies such as diversifying income sources, adopting conservative production practices, or reducing exposure to known hazards tend to achieve better livestock production outcomes. This finding is consistent with previous studies highlighting the importance of risk-averse behaviors in improving farm stability and productivity (Patil and Veettil, 2024; Spada et al., 2025).

Other significant risk management strategies include technology investment (β = 8.952, p = 0.003) and market strategies (β = 3.092, p = 0.009). Investment in improved technologies such as better breeds, feeding systems, or veterinary care has been shown to enhance production efficiency and disease resistance (Dayoub et al., 2024). Meanwhile, strategic engagement with markets through better pricing information, contract farming, or cooperative marketing helps farmers mitigate price risks and access premium markets (Saroj and Paltasingh, 2024). The variable production diversification also positively influences livestock production, albeit with a smaller coefficient (β = 0.003, p = 0.005). This aligns with the concept that diversified farming systems, incorporating multiple livestock species or crop-livestock integration, enhance resilience by spreading risk across various outputs (Bilotto et al., 2024).

Socio-demographic factors further contribute significantly to livestock production. Household size (β = 0.968, p = 0.005) and age (β = 1.648, p = 0.002) indicate that larger households and older farmers tend to have higher production, possibly due to greater labor availability and accumulated experience, respectively. Education, measured by number of years in school (β = 0.557, p = 0.007), also positively impacts production, underscoring the role of formal education in improving farmers' capacity to adopt innovations and effective management practices. Other socio-economic variables, such as savings and asset accumulation (β = 4.776, p = 0.029), gender (β = 0.114, p = 0.026), occupation (β = 0.002, p = 0.023), and sharecropping and other production relations (β = 1.014, p = 0.004), also significantly affect livestock productivity. Savings and assets provide critical buffers that allow farmers to invest in inputs and recover from shocks. Gender significance points to differential access to resources and decision-making power, with implications for production efficiency (Wann et al., 2024). Occupation and production relations influence how farmers allocate resources and labor, affecting overall output.

The model fit statistics demonstrate explanatory power, with an adjusted R² of 0.976 and R² of 0.965, indicating that over 96% of the variation in livestock production is explained by the variables included in the model. The significant chi-square statistics (50.446) further confirm the model's strength.

The gross margin analysis was undertaken to evaluate the profitability of livestock farming among smallholder farmers in the Eastern Cape. This approach allows for a clear understanding of the balance between income generated from livestock sales and the variable costs incurred in maintaining production. By focusing on the major livestock types (cattle, sheep, goats, pigs, and chickens), the analysis highlights which enterprises provide the greatest financial returns and how these align with household livelihood strategies. The results are particularly important given the study's finding that 85% of farmers are unemployed and depend heavily on livestock as their primary income source. Presenting gross margins both per farmer and per 100 animals sold provides a comprehensive picture of the economic contributions of different livestock species within smallholder systems. The Table 8 shows the gross margin results of smallholder livestock farming in the study area.

The gross margin analysis demonstrates that livestock farming among smallholder farmers in the Eastern Cape generates a positive return, though with varying contributions across species. Cattle dominate the revenue stream, contributing nearly ZAR966,000 from sales, with a gross margin of ZAR690,000, which is equivalent to about ZAR500,000 per 100 animals sold. This supports the study's observation that cattle are retained as long-term financial assets and symbols of wealth, only sold during times of financial pressure. Sheep and goats, with gross margins of ZAR150,000 and R130,000 per 100 animals respectively, provide a hybrid role by serving both income-generation and nutritional needs. Pigs, though fewer in number, yield strong margins (ZAR190,000 per 100 animals), highlighting their role as a more commercialized livestock option compared to ruminants. Chickens, on the other hand, provide the lowest gross margins (ZAR7,000 per 100 birds), but their short production cycle and accessibility make them essential for meeting immediate household needs and food security, consistent with the study's findings that pigs and poultry offer quick cash inflows.

At the household level, the average gross margin per farmer is estimated at ZAR5,620 annually. This aligns with the sociodemographic profile presented in the study, where 85% of respondents were unemployed and thus heavily reliant on livestock as a primary income source. The analysis confirms that while livestock income does cover basic operational costs (such as feed supplementation, veterinary care, and transport), it is also diverted to meet pressing household obligations like food, education, and social commitments. These results are consistent with the findings of Mdoda and Obi (2019), Manganyi et al. (2023), Mbelebele et al. (2024) and Mujuru et al. (2022), who emphasize that smallholder farming remains the most reliable and trusted source of income generation in the Eastern Cape Province, particularly in the context of high unemployment and poverty. Smallholder farming plays a critical role in sustaining livelihoods, as the income generated is not only essential for supporting household needs but also for maintaining and improving farming operations. This dual use of income underscores the vulnerability of smallholder farmers: although livestock farming is economically viable, reinvestment into herd improvement and risk mitigation remains limited (Mdoda et al., 2019; Nondzutha et al., 2020). The findings reinforce the study's argument that institutional gaps, such as inadequate extension services and poor vaccination uptake, constrain productivity, and profitability.

The propensity score matching results presented above indicate that risk behavior significantly influences farm returns among smallholder livestock farmers in the Eastern Cape Province. Results are shown in Table 9. Across all matching algorithms (nearest neighbor, radius, kernel, and stratification), the Average Treatment Effect on the Treated (ATT) is positive and statistically significant, ranging from R1,250 to R1,450. This suggests that risk-averse farmers, who actively engage in strategies such as diversification, precautionary savings, and adoption of preventive animal health measures, achieve higher farm returns compared to risk-preferring farmers. The highest ATT value observed under nearest neighbor matching (R1,450, p < 0.01) implies that risk-averse behavior yields approximately a 26% increase in net farm returns relative to the average gross margin per farmer (R5,620). These results affirm the hypothesis that risk-averse decision-making enhances financial resilience and income stability, consistent with findings by Mdoda et al. (2025) and Patil and Veettil (2024) that cautious and diversified management strategies improve profitability in uncertain agricultural contexts.

The results further emphasize the importance of behavioral and managerial choices in determining economic outcomes in smallholder systems. The significant ATT values across multiple matching methods confirm the robustness of the estimates, indicating that differences in farm returns are attributable to behavioral differences rather than sample selection bias. Risk-averse farmers tend to adopt a longer-term perspective, reinvesting profits into herd improvement, veterinary care, and market-oriented production, while risk-preferring farmers often exhibit reactive or speculative behaviors that expose them to greater financial volatility.

This study distinguishes itself through methodological innovation, conceptual depth, and contextual relevance to smallholder livestock systems in South Africa and beyond. It advances existing knowledge by integrating risk mitigation, livestock management, and economic performance within a unified analytical framework, a perspective seldom explored in African agricultural research. Through a case study strategy grounded in empirical fieldwork, the research captures the lived experiences of smallholder livestock farmers in the Eastern Cape Province, an area where evidence-based insights remain limited. The study employs a triangulated econometric approach combining descriptive statistics, binary logistic regression, and propensity score matching (PSM) to provide a rigorous and multidimensional analysis. This methodological fusion enhances internal validity and allows the research not only to identify determinants of risk behavior but also to quantify the economic effects of risk management strategies on livestock productivity. The adoption of a multi-stratified random sampling technique across diverse districts strengthens representativeness, ensuring that variations in socioeconomic and environmental contexts are captured. Additionally, the translation and back-translation of survey instruments into IsiXhosa, along with enumerator training and ethical compliance, reflect a deep commitment to research integrity, cultural sensitivity, and scientific robustness.

Beyond its analytical sophistication, the study contributes meaningfully to both theoretical and policy discourse by introducing an integrated conceptual framework that visually and empirically connects sources of risk, farmer perceptions, mitigation strategies, and production outcomes. This framework offers a transformative lens for understanding risk and resilience in smallholder systems, moving beyond deficit-based models toward an agency-centered perspective that acknowledges farmers as adaptive, strategic decision-makers. The study's methodological rigor (anchored in the triangulation of descriptive statistics, binary logistic regression, and PSM) systematically isolates and measures the influence of risk behavior on livestock-specific outcomes such as farm returns, herd productivity, and animal health management. Unlike general agricultural analyses, this research focuses explicitly on behavioral risk attitudes, institutional support systems, and indigenous management strategies as core determinants of livestock resilience. The PSM results provide robust comparative estimates of the Average Treatment Effect on the Treated (ATT), revealing that risk-averse farmers achieve higher economic returns than risk-preferring counterparts, directly linking behavioral risk management to economic stability. By bridging econometric precision with contextual empathy, the study not only deepens scholarly understanding of livestock risk management economics but also generates actionable policy insights on how targeted interventions in extension services, vaccination access, market integration, and financial inclusion can foster resilience and sustainable rural development across the Global South.