INTRODUCTION

Research and Markets [1] highlighted that as of 2025, the international livestock and meat market is estimated to be worth USD 573 billion. Analysts predict the market will increase significantly, reaching USD 637 billion by 2029. Due to changing diets, population growth, and rising wages, the livestock industry is the fastest expanding agricultural subsector in middle- and low-income nations. The entire livestock value chain offers significant potential for farmers, agribusiness, and employment providers [2]. The livestock sector’s significance should be emphasised as a consumer or user of natural resources and a source of livelihood support, which is essential in providing people and communities with sustainable options that, in turn, stimulate economic growth [3]. As the single biggest user of land systems on earth, the livestock industry occupies extensive areas of rangelands and utilises significant resources. It consumes a third of the world’s freshwater sources and a similar proportion of cropland for feed production. Despite these demands, Herrero et al. (2013) [3] view the livestock industry as a cornerstone of the global economy, accounting for 40% of agricultural Gross National Product (GNP) and covering 30% of the planet’s ice-free surface.

Drought significantly impacts commercial livestock farmers, affecting all regions of the world, including sub-Saharan Africa [4]. The reduced availability of water and quality pastures leads to nutritional deficiencies and decreased livestock productivity, increasing feed and water costs and placing financial strain on farmers [5]. Globally, these factors contribute to market fluctuations as supply decreases, impacting prices and overall stability in the livestock sector [6]. Furthermore, the increasing frequency of droughts linked to climate variability indicates long-term threats to food security and agriculture worldwide. Thus, influencing SDG2 and the African Union CAADP Strategy and Action Plan: 2026–2035, which aims to end hunger and poverty by intensifying sustainable food production, agro-industrialisation, and trade.

Farmers in South Africa face challenges that hinder their ability to achieve sufficient production and profitability. Key issues include increased demand for agricultural land and the rising costs of farming inputs, while output prices do not necessarily increase at the same rate [7]. These factors, among others, have created a financial squeeze for farmers. Additionally, farmers have to contend with erratic rainfall and other weather-related issues due to climate variability. The effects of climate change impact developing nations across all areas of development, particularly in agriculture [8]. As a result, the National Disaster Management Centre declared the persistent drought conditions throughout the Northern Cape Province of South Africa a disaster area [9]. This declaration highlighted the ongoing and varying intensity of drought in the province, which persisted over time. The Northern Cape Province has the lowest rainfall and is the most arid of all provinces in South Africa; due to the climate conditions, livestock production is the only viable agricultural activity.

Given the anticipated increase in drought frequency and intensity due to climate change, the adverse effects on people and the economy will likely worsen. Economic vulnerability is a significant adverse outcome of drought in agriculture. Effective policies and resilience measures are imperative to minimise drought risk and support affected communities, especially livestock producers who are disproportionately impacted. Enhancing resilience involves adapting to gradual and rapid changes and creating robust systems, which can withstand and recover from these challenges [10–12].

Lottering et al. (2020) [13] used a systematic approach to examine the impact of droughts on smallholder agriculture in sub-Saharan Africa and found that droughts had varying effects on the environment, social structure, and economic development. The authors suggest that critical follow up actions should be taken to enhance preparedness and mitigation measures to minimise the impact of drought on smallholder farmers and rural populations. Holman et al. (2021) [14] used a Driver-Pressure-State-Impact-Response (DPSIR) approach, an analysis of weekly agricultural trade publications and conducted semi-structured interviews to explore drought impacts and responses in the United Kingdom. The authors found that on farm, a diverse range of measures were implemented across institutional scales and the supply chain, reflecting complex interactions within the food system. Drought responses were dominated by reactive and crisis-driven actions to cope with or enhance recovery from, drought, which contributed little to increased resilience towards future droughts. Mdungela et al. (2017) [15] conducted an exploratory study to identify and quantify the indicators of economic drought vulnerability among community farmers in South Africa. The authors found that farm debt and financial safety nets were highest in the vulnerability index, indicating a strong influence on financial vulnerability. Mare et al. (2018) [7] assessed the impact of drought on commercial cattle farmers in South Africa during the 2015–2016 drought. The descriptive study was based on primary data collected from 350 commercial livestock producers in seven provinces. The findings revealed that the effect of drought had a significant impact on average herd/flock size, and livestock feed.

Bahta and Myeki (2022) [16] conducted a study in the Northern Cape Province of South Africa to assess the impacts of agricultural drought on smallholder farming households. Results revealed that nearly 80% of smallholder livestock producers were not resilient to agricultural drought, which underscored the need for more robust resilience strategies. Statistics South Africa [17] investigated the impact of agricultural drought resilience on the welfare of smallholder livestock farming households in the Northern Cape Province. The findings revealed that smallholder farmers achieved improved outcomes when supported during periods of drought. The study also found that economic, social, human, and natural capital influenced the welfare of smallholder farmers.

Many studies focused on smallholder livestock farmers, with Mare et al. (2018) [7] concentrating on commercial farmers. However, their study did not address economic vulnerability, but examined the adaptive measures implemented by farmers and the impacts on drought on herd/flock size and livestock feed. As a result, there is a gap in the current understanding of economic vulnerability indicators for agricultural drought, particularly in the context of commercial farmers, their revenue, and adaptive drought strategies. Therefore, this study identified and evaluated indicators of economic vulnerability to agricultural drought among commercial farmers in the Northern Cape Province of South Africa. The aim is to contribute to the dialogue around drought risk and mitigation policies. The findings will be in accord with Sustainable Development Goal 2 and the African Union’s Comprehensive Africa Agriculture Development Programme (CAADP) Strategy and Action Plan: 2026–2035, which focuses on eradicating hunger and poverty while enhancing sustainable food production, agro-industrialization, and trade. Although this research is conducted in South Africa, the conceptual framework could be applied to other regions.

MATERIALS AND METHODS

The Northern Cape Province is situated in the northwest of South Africa, with Kimberley as its capital. With an estimated population of 1.2 million, the Northern Cape has the smallest population among South Africa’s nine provinces. However, it has a surface area of 372,889 km², or 30.5% of the country’s total land area, making it the largest province [18]. Five district municipalities are present in the province: Namakwa (126,900 km²), Pixley Ka Seme (103,500 km²), ZF Mgcawu (102,500 km²), John Taolo Gaetsewe (27,300 km²), and Frances Baard (12,800 km²) (Northern Cape Municipalities, 2012).

According to Richaud (2019) [19], the Vegetation Condition Index (VCI) for 2019, compiled using 23 years of data, indicated that the drought was the most severe in the central, southern, and western parts of the Northern Cape Province. This area is west of the N10 road, north of the Western Cape border, east of the Atlantic Ocean, and south of the Namibian border. Regarding vegetation, three broad areas fall within this region: Namakwa, the Great Karoo, and the Green Kalahari. The Namakwa region comprises the Namakwa District Municipality, the Green Kalahari Region comprises part of the ZF Mgcawu District Municipality, and the Great Karoo Region comprises part of the Pixley ka Seme District Municipality. This area, as identified by Richaud (2019) [19], was designated as the study area and is shown in Figure 1. The local municipalities are listed in Table 1.

FIGURE 1

Figure 1. Study area. Source: Authors.

The livestock industry is a crucial part of the province’s economy, making it the ideal setting for assessing the economic vulnerability of commercial livestock farmers. However, there are other drivers, such as irrigation schemes to produce crops that require large amounts of water. Table 1, based on Stats SA [20], provides the number of commercial (Value Added Tax (VAT) registered) farmers in the different areas.

TABLE 1

Table 1. Total number of Value Added Tax (VAT) registered commercial farmers and those engaged in livestock farming in the study area.

A multi-stage sampling procedure was used. In the first stage, the Northern Cape Province was deliberately selected because it is South Africa’s primary livestock-producing province, and the National Drought Mitigation Center has declared the Northern Cape a disaster area 6 times since 2002, the latest in the 2020/2021 production season. The second stage involved random sampling of the study area by ballot from the VAT list of registered commercial livestock farmers. The ballot system involves randomly selecting a subset of cast ballots to ensure a representative, impartial sample. Three district municipalities were included because they were most affected during previous droughts. In the third stage, commercial livestock farmers were selected using simple random sampling and interviewed. Meanwhile, the sampling process had several stages; only people who met certain requirements were included in the sample. To be eligible for inclusion, participants had to be commercial livestock farmers, willing to take part, and members of the district municipalities.

As indicated in Table 1, 302 VAT-registered commercial livestock farmers were identified in the study area [20]. The Cochran (1997) [21] and Bartlett et al. (2001) [22] methods for simple random sampling were used to determine the sample size, with Bahta (2022) [23] also using this method to study smallholder farmers’ social vulnerability. The correct sample size was determined using Cochran’s (1977) [21] sample size formula (Equation (1)):

$\text{Sample size=}\frac{{\text{(q)}}^{\text{2}}\text{*(z)(r)}}{{\text{(w)}}^{\text{2}}}$ (1)

where “q” is the level of confidence/alpha level (value for the selected alpha level indicates the level of risk the researcher is willing to take so that the actual margin of error may exceed the acceptable margin of error); z and r are the estimates of the variance of the population; estimate of variance calculated as = 0.25 (maximum possible proportion (0.5) × 1 − maximum possible proportion (0.5) produces maximum possible sample size); and “w” is an acceptable margin of error for proportion being estimated = 0.05 (5%) (error researcher is willing to take). If this formula were applied to the study with an alpha level of 1.28 (0.20), estimated variance of 0.5 and an error level of 0.05, the formula would be as follows (Equation (2)):

$\text{Sample size=}\frac{{(1.28)}^2*(0.5)(0.5)}{{(0.05)}^2}$ (2)

Thus, resulting in a sample size of 164 participants. With 302 registered commercial livestock farmers and applying the formulas in Equations (3) and (4), a sample size of 164 would represent 54% of the total population, requiring the application of Cochran’s (1977) [21] formula.

${\text{N}}_{\text{1}}\text{=}\frac{\text{Sample size}}{{\text{1+(N}}_{\text{0}}\text{/population)}}$ (3)

${\text{N}}_{\text{1}}\text{=}\frac{\text{164}}{\text{1+(}\frac{164}{302}\text{)}}$
${\text{N}}_{\text{1}}\text{=106}$ (4)

Therefore, 106 commercial livestock farmers had to be interviewed. However, the study obtained complete questionnaires from 123 participants. For sample adequacy, a post hoc power analysis using G × Power indicates that with 123 observations, 10 predictors, and an alpha of 0.05, the study attains 80 per cent power to detect medium effect sizes (Cohen’s f² = 0.15).

Primary data were collected from the questionnaires, visits to farmers’ associations, and face-to-face interviews. The data collection period was from August 2023 to October 2024. Both qualitative and quantitative data were gathered via the questionnaires. The questionnaire covered a wide range of topics, including the effects of drought, socio-economic characteristics of farmers, perceptions of drought, changes in the availability of natural resources, livestock production, and other farming activities. It also addressed resource availability, household vulnerability, food security, and social safety nets. The questionnaire covered a broad range of topics in a larger study, with this article focusing on economic vulnerability. The study received the necessary ethical clearance to gather the data (Ethical clearance number: UFS-HSD2023/0373/3).

The survey data from the participants were processed and utilised to calculate the economic vulnerability index. Composite indices are widely used in economic vulnerability and resilience research because vulnerability is inherently multidimensional and cannot be captured using a single indicator. These indices are also widely employed by international organisations, including the United Nations, World Bank, and OECD, to support policy prioritisation, vulnerability screening, and resource allocation decisions. Meanwhile, these composite indicators inevitably involve methodological choices; their value rests on their ability to synthesise complex information into interpretable metrics that support comparative analysis and evidence-based decision making.

From both conceptual and empirical perspectives, economic vulnerability is well documented [24]. Briguglio et al. (2009) [24] define economic vulnerability as an economy’s exposure to exogenous shocks arising from economic openness. The initial step in identifying the economic vulnerability indicators for drought in the Northern Cape used the BBC (Bogardi, Birkman, Cardona) framework, presented in Figure 2. According to Cardona (1999 &2001) [25,26], the BBC conceptual framework underscores the intricate nature of vulnerability and resilience in the face of external shocks, whether human- or environment-related. This framework was chosen for its comprehensive approach in addressing social, environmental, and economic vulnerabilities, which are fundamental to sustainable development. It emphasises that vulnerability assessment goes beyond merely evaluating the impact of disasters; it focuses on coping capabilities and intervention methods to mitigate vulnerabilities. Moreover, the BBC model is favoured for its emphasis on proactive risk reduction, which highlights the importance of identifying economic vulnerabilities before a disaster strikes to understand the degree of vulnerability [27].

FIGURE 2

Figure 2. Conceptual Framework for Vulnerability. Source: Bogardi and Birkmann (2004) [27]; Cardona (1999 &2001) [25,26].

The selected economic indicators are presented in Table A1. Indicators were selected from literature, expert opinions (extension officers), observations, ease of measurement, feedback from farmers, relevance, availability, and the importance of the indicator to the Northern Cape Province of South Africa. A lack of access to essential resources such as water and quality feed can heighten vulnerability. Additionally, the sensitivity of livestock product prices to market changes significantly affects farmers’ income stability. On-farm diversification allows farmers to mitigate risks by producing a variety of livestock or crops, providing a buffer against adverse conditions. The level of debt is also crucial; those with higher debt may struggle to recover from losses. Market access is vital in ensuring that farmers can sell their products effectively. Financial safety nets and alternative income sources offer essential support during adverse times, while off-farm diversification can secure additional revenue when farm income declines.

A decrease in farm income (revenue) was added, calculated by subtracting the farm income of drought years from farm income during normal years (income earned without drought) and dividing it by the normal year’s income (income earned without drought) to use as a base. A decline in farm income and subsequent additional costs during a drought can worsen vulnerability. Farming practices instituted to minimise potential drought effects were added to the indicators; these practices, such as conservation agriculture, can make farmers less vulnerable to drought. Examining the interconnected elements enabled a comprehensive understanding of how agricultural drought influenced commercial livestock farming. A Likert scale was used to categorise the indicators by assigning the following scores: extremely high (5), high (4), moderate (3), low (2), and extremely low vulnerability (1). Empirically, the economic vulnerability index is calculated as (Equations (5) and (6)):

$V^{eco}={\sum }_{i=1}^{20}{w}_i^{eco}{v}_i^{eco}$ (5)

$V^{eco}f(w_1^{eco}{v}_1^{eco},w_2^{eco}{v}_2^{eco},\dots \dots ,w_9^{eco}{v}_9^{eco})$ (6)

where: $v_1^{eco}$ = Lack of access to resources; $v_2^{eco}$ = Price sensitivity of products; $v_3^{eco}$ = On-farm diversification; $v_4^{eco}$ = Level of debt; $v_5^{eco}$ = Market access; $v_6^{eco}$ = Financial safety nets/ alternative sources of income; $v_7^{eco}$ = Off-farm diversification; $v_8^{eco}$ = Decrease in farm income (farm income in normal years (without drought) (Definitions for “normal years” and “drought years” are generally determined by comparing precipitation or water supply levels against long-term averages (climatology) for a specific region, often using standardised indices. Drought year is a prolonged period of below-normal precipitation resulting in water shortages, environmental damage, or, in socioeconomic terms, demand exceeding supply. Normal year comprises years with precipitation, streamflow, or soil moisture levels close to the long-term average.)—farm income during drought years)/farm income during normal years (without drought) × 100); $v_9^{eco}$ = Farming practices applied; and $w_1^{eco}$ − $w_9^{eco}$ are weights created using principal components analysis.

Principal Component Analysis (PCA) was used to construct the Economic Vulnerability Index from the nine economic indicators. Before analysis, all variables were standardised using z-score normalisation to eliminate scale differences. The suitability of PCA was confirmed using the Kaiser-Meyer-Olkin (KMO) test and Bartlett’s Test of Sphericity in Table 2. The first principal component was retained as the Economic Vulnerability Index based on eigenvalues greater than one and the proportion of variance explained. Factor scores generated using the regression method were used as the composite vulnerability index.

The KMO measure of sampling adequacy was 0.581, which reflects an acceptable suitability of the data for Principal Component Analysis. The Bartlett’s Test of Sphericity was also significant at one percent which confirms that the correlations among variables were sufficiently large for PCA.

TABLE 2

Table 2. KMO and Bartlett’s Test of Sphericity.

The results of in Table 3 below show the total variance explained where PCA extracted four components with eigenvalues greater than one, which explaining 61.7% of the total variance. The first component explained 21.98% of the variance, which suggests that economic vulnerability is multidimensional rather than driven by a single dominant factor, and the cumulative variance exceeds the recommended 50%.

TABLE 3

Table 3. Total Variance Explained.

Economic Vulnerability Index in the IPCC Framework

Due to the multidimensional nature of vulnerability as discussed by Ayanlade (2023) [28], this study adopts the Intergovernmental Panel on Climate Change (IPCC) vulnerability framework, which defines vulnerability as a multidimensional, complex phenomenon, a propensity or predisposition to be adversely affected by climate change. In its Sixth Assessment Report (AR6), the IPCC highlights that economic vulnerability is not just a measure of GDP, but a compound risk influenced by social, political, and environmental factors. The IPCC conceptualises vulnerability as a function of exposure, sensitivity, and adaptive capacity. Within this framework, economic vulnerability to agricultural drought is understood as the extent to which farming households are economically susceptible to drought shocks, their structural sensitivity to these shocks, and their ability to cope and adapt through available resources and strategies.

The Economic Vulnerability Index was constructed using a PCA-based weighting approach based on the IPCC vulnerability framework. In this process, indicators were first grouped into the exposure, sensitivity, and adaptive capacity dimensions according to economic vulnerability theory, as shown in Table 4. Principal Component Analysis was thereafter employed separately within each dimension to derive data-driven weights and construct sub-indices. The resulting factor scores were normalised using a min-max transformation, and the Economic Vulnerability Index was obtained as the sum of the weighted (using the PCA variance explained) exposure and sensitivity indices, plus the inverse of the adaptive capacity index, to ensure that higher values reflect greater economic vulnerability.

$EV{I}_i=[(w*Exposure)+(w*sensitivity)+w(1-Adaptive)]$ (7)

TABLE 4

Table 4. Indicators used in constructing the Economic Vulnerability Index-IPCC.

The principal component analysis (PCA) diagnostics in Table 5 confirm that the selected indicators are statistically appropriate for constructing the economic vulnerability index. The Kaiser-Meyer-Olkin (KMO) values for the exposure (0.50), sensitivity (0.55), and adaptive capacity (0.64) sub-indices all meet the minimum acceptable threshold of 0.50, indicating adequate sampling adequacy and sufficient intercorrelations among the variables. Moreover, the Bartlett’s Test of Sphericity was statistically significant across all three sub-indices, confirming that the correlation matrices are not identity matrices and that meaningful latent structures exist within the data. This provides strong statistical justification for applying PCA as a data reduction and weighting technique in the construction of the vulnerability index. The PCA results further show that one principal component was retained for each vulnerability dimension following the eigenvalue greater than one criterion.

The PCA factor loadings in Table 6 provide valuable insights into the relative significance of indicators contributing to economic vulnerability among commercial livestock farmers. In the exposure dimension, price sensitivity (0.759) and a decrease in farm income (0.759) show identical and strong factor loadings, indicating that market price fluctuations and drought-induced income shocks together constitute the primary sources of economic exposure. This finding underscores the vulnerability of commercial livestock farmers not only to climatic stress but also to market instability, reinforcing the notion that drought vulnerability encompasses both production and market risk phenomena.

TABLE 5

Table 5. KMO and Bartlett’s Test of Sphericity for EVI-IPCC.

TABLE 6

Table 6. Factor loadings for vulnerability indicators.

In the sensitivity dimension, market access constraints (0.750) and lack of access to resources (0.745) emerge as the strongest contributors, highlighting the major role structural constraints play in shaping farmers’ economic vulnerability. These results suggest that farmers with limited access to production resources and markets are less equipped to absorb economic shocks. The relatively lower loading for debt levels (0.490), though still significant, may imply that debt alone does not necessarily heighten vulnerability unless it is coupled with other structural disadvantages such as poor market integration or limited productive assets.

Regarding adaptive capacity, the strongest contributors are adaptive farming practices (0.778) and on-farm diversification (0.774), followed by off-farm diversification (0.562) and financial safety nets (0.554). These findings emphasise the crucial role of diversification and proactive drought management strategies in mitigating economic vulnerability. Farmers who adopt multiple production activities and drought adaptation practices seem better positioned to withstand financial shocks. The results also suggest that while financial safety nets contribute to resilience, practical on-farm adaptation strategies may play an even more significant role in bolstering economic resilience.

To analyse the determinants of economic vulnerability among commercial livestock farmers, this study employs a fractional logit regression model by Papke and Wooldridge [29]. This model is adopted because the economic vulnerability index is bounded between zero and one (see Figure A1). The fractional logit model produces consistent and efficient estimates compared to ordinary least squares regression (OLS), which may generate biased predictions outside the feasible range. A Tobit model was later estimated as a robustness check, given the nature/distribution of indices, with 0 to 1 bound.

The fractional logit model is a quasi-likelihood method that does not assume a distribution but only requires the conditional mean to be correctly specified, yielding consistent parameter estimates with an identical likelihood function that has a structure similar to that of a Bernoulli distribution [30]. The quasi-log likelihood for observation i is similar to the logit binary response model and presented as:

$l_i(a)=y_i\log [h(x_{i}a)]+(1-y_1)\log [1-h((x_{i}a)];\text{ with }0\le y_i\le 1$ (8)

where h(X_iα) the logistic Cumulative Distribution Function and y_i∈[0,1] (that differs from binary logit, which limits y to values of 0 and 1). The expectation form of the model is specified as:

$E(y_1|X_i)=\frac{\exp (x_{i}a)}{[1+\exp ((X_{i}a)]}=h((X_{i}a)$ (9)

With the boundary probabilities for the dependent and independent variables specified as:

$\mathrm{Pr}((y_i=0\left|X_i)\right.〉0\text{ and }\mathrm{Pr}((y_i=1\left|X_i)\right.〉0$ (10)

To confirm the suitability of the fractional logit model, a double limit Tobit regression for a robust check with a latent variable setup with the possibility of two corner solutions, as 0 for no vulnerability and 1 for complete vulnerability, was estimated. The Tobit model assumes that there is a latent variable Y* such that

$y_i^{*}=X_{i}a+f(y_{i-1})\rho +k_i+{\epsilon }_i$ (11)

${\epsilon }_i|(X_i,y_i-1,\dots \dots ,y_{i0},k_i)~N(0,{\delta }_{\epsilon }^2)$ (12)

$y_i\{\left\{\begin{array}{c}0if{y}_i^{*}\le 0\\ y_i^{*}if0<y_i^{*}<1\\ 1if{y}_i^{*}\ge 1\end{array}\right.$ (13)

where X_i represent independent variables, k_i represent the unobserved effects and ε_i is the normally distributed error term. Before estimation, multicollinearity was tested using the Variance Inflation Factor (VIF), with an average VIF of 1.5, which is below 10, confirming no serious multicollinearity is observed (see Table A2). Variables with VIF values above 10 were examined for potential removal. Heteroskedasticity was addressed using robust standard errors.

RESULTS AND DISCUSSIONS

Table 7 displays the socio-economic features of the participants. In South Africa, the norm is that males dominate commercial farming. This is supported by the results, which show that 98.4% of the farmers are male and 1.6% are female. The dominance of male farmers can be attributed to the physical work involved as well as social and institutional factors, such as land tenure structure. Similarly, Ngaka (2012) [31] found that farmers are predominantly male.

Most participants were married (89.4%), followed by single (9%), with widows/widowers and divorced at 0.8% each. Being married can add value in times of drought, as partners can help deal with stress, or, if needed, can seek other work to provide additional income. Bahta (2021) [32] reported somewhat different results among smallholder farmers, where only 67% were married, and 11% were widowed.

Education can be critical in deciding what steps should be taken to cope with drought. Few (6.5%) participants left school before Grade 12, with 26.8% finishing Grade 12.8%. Most participants, however, had a degree or diploma (48.8%) or a postgraduate qualification (17.9%). Education can assist with making better decisions on the farm. In contrast, Mdungela and Bahta (2017) [15] found that most smallholder farmers were only educated to a primary school level, and only 3.68% obtained tertiary-level qualifications.

Farmers’ primary occupation is not necessarily full-time farming, where some might have another primary job and farm on weekends or when time allows. In the current study, farming was significantly the main occupation (93.5%) compared to other formal employment (4.9%) and businesses (1.6%). Thus, most participants depended primarily on farming as an income source, making them more vulnerable to the impact of drought. This concurs with DAFF’s [33] findings where most households depended on livestock farming as the only income source.

The youngest and oldest participants were 24 and 84 years old, respectively; the average age was 53 years. The highest age group, 50 years and older, accounted for 72.4% of the participants. Similar results were found by Mdungela et al. (2017) [15], who found that 42% of farmers were older than 55 years. The years of farming experience ranged from 3 to 58 years, with an average of 28 years. The groups of 20 to 29 years of farming experience and 30 to 39 years of experience comprised 24.4% and 25.2%, respectively. The age trade-off is that younger farmers are more likely to take risks and test new methods, while older farmers stick to what they know [34]. Conversely, more years of farming experience imply that farmers have more knowledge to apply to specific situations, which can be important during times such as drought.

TABLE 7

Table 7. Socio-economic characteristics of livestock farmers.

Farm size included farms owned, rented, and the total area used for production. The smallest area was 1000 hectares, and the largest was 33,000 hectares, with an average of 9436 hectares. Farm sizes are considerably larger than in other provinces, as the carrying capacity is much lower. Thus, farmers need much larger farms to carry the same number of animals and make it financially feasible. Over two-thirds (82.1%) of the farms’ sizes were larger than 6,000 hectares. The vast area is supported by Meissner et al. [35], who state that more land is needed in semi-arid to arid areas than in any other area in South Africa to make farming profitable.

The economic vulnerability index was first computed using the factor scores generated using the regression method. The distribution of EVI, as presented in Table 8, shows that 64.2% of commercial livestock farmers fall within the high vulnerability category, while only 7.3% exhibiting low vulnerability. This may be driven by a lack of adaptive strategies, diversification options, high sensitivity of product prices, and constrained access to markets. This shows that most farmers lack sufficient financial buffers, diversification strategies, and market resilience to withstand drought shocks. The findings indicate prevalent exposure to financial risk among commercial livestock farmers, largely driven by income losses, price volatility, and limited adaptive capacity [36]. This outcome further invites multidimensional assessment of vulnerability.

TABLE 8

Table 8. Economic vulnerability index distribution.

The Economic Vulnerability Index distribution based on the IPCC framework in Table 9 indicates that most commercial livestock farmers (84.6%) are categorised as moderately vulnerable, with only 5.7% classified as having low vulnerability and 9.8% as highly vulnerable. This distribution indicates that while economic vulnerability is widespread among the sampled farmers, it is typically not severe. The high prevalence of moderate vulnerability suggests that while farmers experience significant financial stress during droughts, they possess a limited coping capacity that prevents them from falling into severe economic hardship. This may be attributed to the commercial nature of their farming systems, where farmers characteristically have some productive assets and management skills but remain susceptible to drought-induced financial pressures such as increased feed costs, reduced productivity, and unstable market prices [37,38].

The relatively small percentage of highly vulnerable farmers indicates that extreme economic fragility is not widespread but remains a concern for a specific group facing compounded challenges like high debt exposure, limited diversification, and weak financial buffers. These farmers likely represent priority targets for policy interventions aimed at enhancing financial resilience and risk management capabilities. The small proportion of low vulnerability farmers further suggests that achieving robust economic resilience is challenging even for commercial producers [39]. This finding may reflect that increasing climate variability and rising production costs are narrowing the margin of economic safety, even for relatively well-resourced farmers. These insights suggest that policy interventions should focus not only on highly vulnerable farmers but also on the large proportion of moderately vulnerable farmers who could potentially become highly vulnerable under severe or prolonged drought conditions if adequate support mechanisms are not reinforced.

TABLE 9

Table 9. Economic vulnerability based on IPCC.

The descriptive statistics of the vulnerability indices in Table 10 and Figure 3 provide further understanding of the economic vulnerability structure among commercial livestock farmers. The exposure index, with the highest mean of 0.705, indicates that most farmers are significantly exposed to economic shocks associated with agricultural drought. This underpins the previous inference, also supported by Bell et al. (2021) [39], that price volatility and income fluctuations are major risk factors impacting financial stability in livestock production systems.

The adaptive capacity index also shows a relatively high mean value of 0.656, which appears counterintuitive given the moderate vulnerability observed. However, this may likely reflect the existence of coping mechanisms among farmers, such as diversification and adaptive management practices, despite the fact that these strategies may not fully counterbalance exposure and sensitivity pressures [36].

The sensitivity index, with a lower mean value of 0.482, suggests moderate structural constraints among farmers, which may suggest that while some farmers encounter resource and market access challenges, these constraints are not uniformly severe across all farmers. The moderate variability observed across all sub-indices reflects heterogeneity in farmers’ economic conditions, which suggests that vulnerability is not uniform but varies according to farm characteristics, financial structure, and management capacity [40].

The general Economic Vulnerability Index has a mean value of 0.510, which indicates a moderate level of vulnerability across the surveyed commercial livestock farmers. The relatively small standard deviation of 0.109 suggests limited dispersion, with most farmers clustering around moderate vulnerability levels. The observed range of 0.235–0.818 confirms the presence of both relatively resilient and highly vulnerable farmers, which underscores the need for differentiated policy strategies rather than uniform interventions. These results confirm that economic vulnerability among commercial livestock farmers is predominantly driven by exposure to drought-related financial shocks, moderated by varying levels of adaptive capacity and structural sensitivity.

TABLE 10

Table 10. Statistics for EVI-IPCC.

FIGURE 3

Figure 3. Vulnerability triangle diagram of the contributing factors of the Economic Vulnerability Index-IPCC (EVI–IPCC) for commercial livestock farmers in Northern Cape, South Africa. Sources: Authors.

To examine the factors influencing the economic vulnerability of commercial livestock farmers to agricultural drought, fractional logit and Tobit regression models were estimated. The Tobit regression is estimated as an alternative technique for robustness to the results. The consistency in coefficient signs and significance levels across both models suggests that the findings are stable and reliable. This is supported by the overall significance of both the chi-square (chi²) and F statistics, which indicate that the selected explanatory variables jointly explain variation in farmers’ economic vulnerability. Furthermore, the similarity in the direction and statistical significance of coefficients between the fractional logit and Tobit models strengthens confidence in the empirical results.

The results presented in Table 11 show that farmer age has a negative and significant effect on economic vulnerability at the one percent level, which shows that older farmers are less economically vulnerable to agricultural drought compared to younger farmers. This reflects the role of accumulated farming knowledge, financial stability, and stronger coping experience. As also discussed by Yeleliere et al. (2023) [41], older farmers may also have well-established networks and assets that help to buffer economic shocks related to drought.

TABLE 11

Table 11. Fractional logit and Tobit model regression results of determinants of economic vulnerability.

Commercial livestock farmers with more farming experience, on the other hand, were observed to have a positive and significant relationship with economic vulnerability. While this appears to be contrary to expectations, it may suggest that more experienced farmers tend to remain heavily invested in livestock production and may therefore face greater financial exposure during drought events [40]. This is also connected to the fact that experienced farmers may operate larger and more commercially oriented enterprises, which increases their exposure to production and market risks [40,42].

Gender and household size were not found to significantly influence economic vulnerability, suggesting that economic exposure to drought among commercial livestock farmers may be driven more by structural and institutional factors rather than household demographic characteristics.

Farm-related attributes, such as farm size and herd size, showed a weaker statistical association with economic vulnerability, suggesting that simply owning larger farms or more livestock does not necessarily reduce economic vulnerability. This indicates that vulnerability depends more on management strategies and financial resilience rather than scale alone [43]. Similar results were observed for cooperative membership, although the negative coefficient suggests a potential shielding role. This outcome may reflect that existing cooperatives are not influential enough to effectively provide drought risk management support or financial protection services to their members.

Farmers with stronger social networks were found to be positively and significantly associated with economic vulnerability. This result may suggest that farmers experiencing higher economic vulnerability are more likely to rely heavily on their close relatives, family, or close-knit groups as coping mechanisms rather than social networks, which necessarily reduces vulnerability. Also supported by Zhang and Zhao (2024) [44], this means these social networks act more as ex-post reactive coping strategies of dealing with the economic fallout rather than preventive resilience mechanisms.

Government assistance to livestock farmers affected by agricultural drought showed a positive and statistically significant relationship with economic vulnerability across all models. This shows the targeting effects of the program, where more vulnerable farmers are more likely to receive government support, which underscores the case that government-supported programs may be reaching intended beneficiaries rather than implying that government support increases vulnerability. This is consistent with the vulnerability studies by Bahta and Myeki (2022) [16], and Sarmah et al. (2025) [45], in which support programs often respond to existing vulnerability conditions by targeting specific socioeconomic, structural, and environmental factors that limit farmers’ adaptive capacity.

Factors connected to climate exposure, market access, and credit do not significantly influence economic vulnerability. Although the number of droughts experienced had a positive coefficient, which may suggest that repeated drought exposure may increase economic vulnerability, the lack of statistical impact may reflect adaptation learning effects after repeated drought exposure. Similarly, distance to auction, abattoir, and feedlot were insignificant, which may show that commercial livestock farmers have developed logistical arrangements that lower the importance of distance.

Concerning the effect of farm wealth and asset effects, remarkably, the immovable asset index showed a positive and statistically significant relationship with vulnerability. This suggests that farmers with greater investment in fixed assets may face higher vulnerability due to the illiquid nature of these investments, which cannot easily be converted into cash in cases of financial stress, which heightens financial rigidity during drought shocks [46,47]. The insignificance of movable/vehicle assets, although positive, may suggest that transport assets alone are insufficient to reduce vulnerability unless combined with other financial or adaptive resources. Similarly, for access to credit, the lack of significance may imply limited access to effective agricultural credit or insufficient credit amounts to meaningfully reduce drought-related financial risks [48].

CONCLUSION AND POLICY IMPLICATIONS

Drought has a significant impact and continues to have long-lasting consequences on South Africa’s agricultural sector. Many factors contribute to the impact of drought, one of which is economic vulnerability. This study examined the economic vulnerability of commercial livestock farmers to agricultural drought in the Northern Cape province of South Africa using the IPCC vulnerability framework combined with Principal Component Analysis and fractional logit and Tobit regression models to determine the associated factors for farmers’ level of vulnerability. By integrating IPCC exposure, sensitivity, and adaptive capacity dimensions, the study provides a comprehensive understanding of how drought affects the economic stability of commercial farming systems. The findings reveal that economic vulnerability among commercial livestock farmers is largely moderate, with most farmers experiencing some degree of financial exposure to drought shocks but maintaining limited coping capacity that prevents extreme vulnerability. Results of PCA showed that economic exposure is largely driven by price volatility and income losses due to drought, while sensitivity is shaped primarily by structural constraints such as limited resource access and market participation. Adaptive capacity is strongly influenced by on-farm diversification and the adoption of drought-adaptive practices, highlighting the importance of management strategies in reducing vulnerability. Results from the fractional logit and Tobit regression models showed that age reduces vulnerability, whereas farming experience increases vulnerability. Moreover, government assistance and social networks are positively associated with vulnerability. Traditional structural variables such as farm and herd size were not found to significantly reduce vulnerability, whereas a positive relationship between immovable assets and vulnerability was observed, suggesting that capital locked in fixed investments may reduce financial flexibility during drought periods. This concludes that economic vulnerability among commercial livestock farmers is shaped more by financial structure, adaptive management capacity, and institutional dynamics than by production scale. These findings highlight the importance of strengthening adaptive capacity and financial resilience mechanisms rather than focusing solely on increasing production capacity.

Therefore, with most farmers being moderately vulnerable, policy should focus on strengthening preventative resilience measures rather than only responding to crises. Programs promoting drought preparedness, enterprise diversification, and climate risk management could significantly reduce the likelihood of farmers transitioning into high vulnerability categories.

The critical role of diversification to adaptive capacity suggests that policies should actively promote both on-farm and off-farm diversification strategies. Support for mixed farming systems, alternative livestock enterprises, and complementary income sources could help reduce financial exposure to drought-related shocks.

While financial structure is observed to play a major role in vulnerability, improving access to financial risk management tools such as agricultural insurance, flexible credit schemes, and drought contingency financing is called for. This may include improving liquidity support mechanisms, mainly for farmers with high fixed asset investments who may face cash flow constraints during drought periods.

The positive impact of government assistance on vulnerability suggests that government support programs are largely reactive. This shows that there is a need to shift from reactive assistance toward proactive resilience-building interventions such as drought preparedness grants, adaptation investment support, and early warning response financing.

The importance of adaptive farming practices suggests that a larger emphasis on promoting drought adaptation strategies, such as conservation practices, improved breeding strategies, and feed management systems, should be encouraged via extension services. In addition to the African Union CAADP Strategy and Action Plan: 2026–2035, which aims to end hunger and improve sustainable food production, agro-industrialisation, and trade, this study contributes to accomplishing SDG 2, which calls for the eradication/reduction of hunger and poverty by 2030.

This study is limited in scope and generalisability as it was undertaken in one province, based on cross-sectional data, which falls short of capturing the economic dynamics, while falling short of discriminating the outcome based on farmer herd composition. Further research is therefore called to scale up the coverage to other provinces in South Africa, incorporating social and environmental vulnerability aspects which may alter the level of vulnerability, and the influence of herd heterogeneity. Additional research could determine whether specific adaptations (such as irrigation investment, solar-powered irrigation as an adaptation and recovery pathway, conservation practices, and financial instruments) reduce vulnerability and improve income resilience.

APPENDIX A. FIGURES AND TABLES DISPLAYING ECONOMIC VULNERABILITY INDEX DESCRIPTIONS, DISTRIBUTION, AND MULTICOLLINEARITY TEST

TABLE A1

Table A1. The Economic Vulnerability Index.

TABLE A2

Table A2. VIF for multicollinearity.

FIGURE A1

Figure A1. Economic vulnerability index distribution.

The Research Ethics Committee approved ethical clearance for the protocol in accordance with the General/Human Research Ethics (GHREC) guidelines and regulations. Reference number UFS-HSD2023/0373/3.

Participants volunteered after the ethical principles were publicly announced. All participants were informed, and they completed and signed a consent form before actively participating in the study.

DATA AVAILABILITY

Data will be available from the corresponding author (YTB) upon request.

AUTHOR CONTRIBUTIONS

All authors significantly contributed to the present manuscript’s preparation. AJV Collected the data and wrote the first draft of the article. YTB, FM, HNvN and SAN were the supervisors of AJV and aided in the review and writing of the final draft. All authors have read and agreed to the published version of the manuscript.

CONFLICTS OF INTEREST

The authors declare that they have no conflicts of interest.

FUNDING

This research was funded by the University of the Free State.