Household coping strategies for food insecurity during droughts in northern Namibia

Abstract

As droughts grow more frequent and severe, understanding how households respond to food insecurity is essential for shaping effective humanitarian and adaptation programs. Based on a household survey of 730 respondents across four regions in northern Namibia, this study explores who experiences food insecurity, the coping strategies households use to manage drought, and how these strategies relate to household characteristics and perceptions of future risk. The survey collected information on household demographics, income sources, access to food aid, and food security status one year after a drought along with coping strategies used to manage drought and food shortages, and perceptions of future food insecurity risk, which were analyzed using descriptive statistics. One year after the most recent drought, 42% of households reported experiencing food insecurity. The most common drought coping strategies were diversifying income (24%) and borrowing money (14%). Households with lower incomes, agricultural livelihoods, and female heads were more likely to experience hunger. Drought coping patterns varied by region with nearly half of respondents in Ohangwena Region (primarily subsistence agriculture) reporting use of at least one coping strategy, compared to 13% in Kunene (arid region with primarily pastoralists). Fear of future food insecurity was widespread, regardless of current food security status or coping strategy use; however, receipt of food aid was associated with lower household hunger. These findings indicate that many households are already taking action to manage drought impacts, yet the effectiveness of these strategies remains uncertain. Rather than applying generic drought solutions, humanitarian and development investments in drought-prone settings should build on existing community practices that are associated with positive outcomes and target those who are most vulnerable. For example, a small percent of households were actively engaging in resource conservation; these practices could be shared and replicated by other households.

1. Introduction

Drought is one of the most pervasive and costly natural hazards globally, affecting nearly every region and threatening food systems and livelihoods. According to the Intergovernmental Panel on Climate Change (IPCC), recent decades have seen increasing drought frequency and severity linked to climate change, disrupting agricultural productivity, accelerating land degradation, and driving humanitarian crises across Sub-Saharan Africa, South Asia, and the Americas (IPCC, 2022). Globally, drought-induced food insecurity now affects hundreds of millions of people, highlighting the urgent need to understand how households cope and adapt in contexts of recurrent climate stress (Kogan et al., 2019).

Many parts of Southern Africa, including Namibia, have a semi-arid climate and reliance on rainfed agriculture, making drought particularly detrimental to food systems and increasing rates of food insecurity (Liu and Zhou, 2021; van Rensburg and Tortajada, 2021). A prolonged dry spell in southern Africa in early 2024 scorched crops and threatened food security for millions of people, due to widespread crop failures and livestock losses (Severe Drought in Southern Africa, 2024). Humanitarian needs surged, as evidenced by a $5.5 billion humanitarian aid appeal to fight the effects of the drought (Dube, 2024). The Namibian government declared a state of emergency and took extreme actions such as culling wild animals, including elephants and hippos, to relieve stress on land and water resources and distributed the meat via drought relief programs for human consumption (Reliefweb, 2025). At the same time, at the community level, households and individuals used their own coping strategies to secure food and ensure their other needs were met during and after this period of inadequate rainfall.

Understanding which strategies households employ to cope with drought, and whether these strategies are effective in maintaining food security, is critical for designing relevant and proactive policy and interventions. Food Stability is defined in the Food Security Cluster Handbook as a cumulative pillar that encompasses all the other three dimensions (availability, access and utilization) ensuring consistency over time regardless of economic or climatic crisis (Global Food Security Cluster, 2023). In addition, coping strategies are defined by literature as responses taken to alleviate the short-term impacts of extreme weather events and are distinguished from adaptation by the focus on short-term solutions (Ashraf et al., 2014; Gichangi and Gatheru, 2018; Savari et al., 2021; Yilma et al., 2014). Adaptation is generally recognized as actions that require long-term adjustments and promote resilience (Owen, 2020). However, in practice, the line between strategies categorized as coping versus adaptation are blurred and, empirically, the ways researchers classify these actions and activities overlap (Orlove, 2022). For this reason, we do not distinguish between coping strategies and adapting activities but rather define all actions participants reported taking to reduce the impacts of the drought as coping strategies. We focus drought related coping strategies. Examples of strategies used to cope with food shortages include reducing the number of meals eaten, changing types of food consumed, and borrowing food from family or neighbors. Examples of drought related coping strategies include diversifying income sources, adapting farming practices, or migrating.

Coping is especially relevant when household needs are high and conventional means of obtaining food sources are disrupted. In such contexts, households may rely on a range of coping strategies from selling animals and other assets, borrowing food and other resources, and getting support from the government or NGOs to less-studied practices such as harvesting wild fruits or pulling children out of school (Huho and Asokan, 2024; Rees, 2021). Fear may also drive individuals to plan or take a specific action, and fear has the potential to be paralyzing and lead to inaction (SANDUA, D., 2024, p. 57). Nava et al. (2025) found that experiencing climatic events and heightened risk perception drives short-term survival strategies and creates “climate traps,” or cycles of survival decision making that actually lead to the reinforcement of longer term vulnerability and instability both economically and ecologically (Nava et al., 2025). Understanding this psychological dimension of coping, how fear of scarcity shapes decisions, is essential to interpreting whether household actions reflect resilience, desperation, or both.

Drought coping strategies remain undocumented and poorly understood. High-cost development projects alongside climate change literature have promoted diversification of income sources or diversification of crops as major adaptation strategies (Amfo and Ali, 2020; Makate et al., 2016; Mulwa and Visser, 2020), yet it is not clear whether people utilize these strategies across geographies. Coughlan de Perez et al. (2024) found that individuals are not diversifying crops, but rather reducing agricultural areas and inputs, when they expect a drought (Coughlan de Perez et al., 2024). This literature suggests a divergence between the strategies promoted in policy and development discourse and those actually practiced by households.

Bhattarai et al. (2015) found that agricultural diversification efforts often increase women's workloads, and many studies have shown that coping strategy use and type vary by gender in general (Bhattarai et al., 2015; Myeni and Wentink, 2020; Rao et al., 2020). For example, women may be more likely to adopt labor-intensive strategies, while men may migrate for wage labor (Bhattarai et al., 2015; Myeni and Wentink, 2020). Another study found that while the strategies men and women take aren't always different, women have less decision-making significance due to cultural barriers, meaning their ideas are not implemented, or they may face barriers in implementation, unless their male counterparts agree that they are useful means of coping with drought (Huho and Asokan, 2024; Ndlovu and Mjimba, 2021).

While policy discourse and investment strategies emphasize diversification as a key adaptation pathway, less is understood about how households actually engage with these promoted strategies and the extent to which they influence food insecurity. This study addresses these gaps by examining the connections between demographic characteristics, coping strategy use, food security outcomes, and fear of future food insecurity in drought-affected communities in Namibia. It contributes to scholarly debates on vulnerability and resilience, and offers practical insights for the design of drought relief interventions and social protection programs offered in Namibia. By situating household experiences within both local and global adaptation discourses, we aim to move beyond binary distinctions between coping and adaptation, emphasizing instead how short-term survival actions intersect with long-term structural vulnerability. In doing so, this study seeks to document the strategies already in use among drought-affected households, encouraging humanitarian and development actors to build on these existing practices rather than introduce externally designed solutions.

This study addresses four research questions:

1. What are the distinguishing characteristics of individuals who experience food insecurity in Namibia?

2. What are the range and frequency of coping strategies households reported using to cope with drought?

3. What are the distinguishing characteristics and perceptions between households who did and did not use coping strategies during the most recent drought?

4. Is the use of coping strategies associated with food security one year after a drought?

By asking who is at risk of food insecurity in drought-affected communities in northern Namibia, we aim to identify household characteristics that distinguish food-insecure from food-secure households. The Integrated Food Security Phase Classification has shown that food security in Namibia varies by season and geographic location (IPC - Integrated Food Security Phase Classification, 2025); we therefore hypothesize that both contextual factors such as urban versus rural geography and regional location, and sociodemographic characteristics, including livelihood type and income, will correlate with food security status.

We also hypothesize that households adopt multiple coping strategies simultaneously to manage the impacts of drought. Some of the strategies we expect at the individual level due to existing literature describing their explicit use in Namibia include pastoralists changing to drought resistant livestock or converting to crops, migrating people or animals, changes in planting timing, reduction in food consumption, and reliance on different income sources including remittances or aid, among others (Inman et al., 2020; Schweiger, 2025). We expect both demographic and contextual factors to influence whether, and how extensively, households employ such strategies. We then explore which coping strategies are associated with food security outcomes one year after the 2023–2024 Southern Africa drought. Finally, throughout this study, we include perceptions on future fears for food security.

Together, our four research questions highlight the relationship between structural vulnerability, household decision-making, and future fears. As depicted in Fig. 1, we link demographic factors, future fears, and coping responses to food insecurity, within the context of drought.

Fig. 1.

Relationship between drought and variables of interest.

(source: created by the author)

2. Methods

2.1. Study area, population, and timing

This study is based on data collected from 730 household level surveys from four regions of northern Namibia: Kavango West, Kunene, Ohangwena, and Oshana. These four regions were selected because of their high vulnerability to drought alongside heavy reliance on rainfed agriculture which can be affected by lack of rainfall, among other vulnerability factors. The regions were proportionately sampled based on their population density (Namibia—Subnational Population Statistics ∣ Humanitarian Dataset ∣ HDX, 2025). Participants were randomly selected using the lottery method based on lists provided by community leaders, which included all households containing permanent residents over the age of 18. While an effort was made to speak directly with the household head, if the household head was unavailable, the next empowered adult was surveyed, assuming they met the inclusion criteria. Refusal rate was approximately 1%. Table 1 provides an overview of the respondent characteristics.

Table 1.

Survey respondent demographic characteristics.

	Region
	Kavango West	Kunene	Ohangwena	Oshana
# surveyed	111	120	300	199
Geography
Rural	111	120	254	136
Urban			46	63
Average age	42.5	39.5	48.8	43.6
Gender				58%
Female	72%	48%	72%	42%
Male	28%	52%	28%
Marital status
Single	23%	33%	39%	39%
Married	41%	43%	30%	43%
Widowed	23%	7%	18%	18%
Other	13%	17%	13%
Most prevalent livelihood	Agriculture/crop production	Livestock production	Agriculture/crop production	Self-employed

The survey collected demographic characteristics, the use of coping strategies to manage drought impacts, two validated food security indicators as well as future concerns about food insecurity. The food security indicators utilized included the food consumption score (FCS) and the household hunger scale (HHS) which are both described in more detail below.

The drought itself, which prompted the declaration of a state of emergency from the government, occurred throughout the 2023–2024 rainy season and left 1.15 million people or 38% of Namibia's population in a state of phase 3 acute food insecurity or worse due to livestock losses and crop failure, according to the Integrated Food Security Phase Classification (IPC - Integrated Food Security Phase Classification, 2025). Survey data was collected by local enumerators in March and April 2025, when IPC classifications showed improvements in food security, especially in northern Namibia, where rains from the new rainy season had begun to alleviate some of the driest conditions. We measured food insecurity in the non-drought period to better understand how people were faring after the drought given the use of coping strategies which may have been utilized before, during, or after the onset of the drought. Importantly, the goal was not to document food security impacts during the drought itself, which are likely to have been much worse. Surveys were verbally enumerated in the local language of each constituency that was surveyed. These languages included Otjiherero, Khoekhoegowab, Zemba, Rukwangali, Oshiwambo, and Oshikwanyama. Responses were documented in English by enumerators using tablets.

In Fig. 2, seasonal conditions depict the expected climatic conditions, classified as either dry or rainy seasons. A dry season is defined as months without rain or surface water while a rainy season is defined as months with rain and surface water. The 2024–2025 environmental conditions presented in Fig. 2 refer to the actual conditions northern Namibia was presented with in the timeframe represented (Jan 2024-Jun 2025). These conditions were generally similar in all four regions, although the extent of impact differs, often associated with geographical features of the region. For example, Kunene, a mountainous landscape, would expect to have visible surface water in the valleys where people are more likely to reside. The impacts of flash floods may have been different in Oshana, a relatively flat region.

Fig 2.

Calendar of seasonal events relative to research timeline.

2.2. Food security indicators

As mentioned, the two food security indicators used in the study were the Food Consumption Score (FCS) and the Household Hunger Scale (HHS). The FCS measures household food security by assessing the diversity, frequency, and nutritional value of foods consumed over the past seven days. The frequencies are summed and weighted by food group, then the household is classified as having “poor,” “borderline,” or “acceptable” food consumption using guidance provided by the World Food Programme (Weismann et al., 2009). For the purpose of this study, we have classified individuals scoring “poor” or “borderline” on the traditional FCS scale as having “unacceptable” food consumption. The acceptable category uses the standard scoring.

The HHS is an experience-based measure of food insecurity that assesses food deprivation and one of the main indicators used in the Integrated Food Security Phase Classification System. This scale inquires about extreme scenarios such as a household running out of food, members going to bed hungry, or going a whole day and night without eating over a 30 day period. Using guidance from Ballard (2011), the frequencies reported are totaled and households are categorized as having “little to no hunger in the household,” “moderate hunger in the household,” or “severe hunger in the household,” (Ballard, 2011). For the purpose of this study, we have classified households categorized as “moderate” or “severe” on the traditional Household Hunger Scale as having “moderate to severe hunger in the household” to create a binary variable for analysis.

2.3. Analysis

Descriptive statistics were first calculated for all variables used in the analysis, including household demographic characteristics, livelihood type, access to food aid, and food security indicators. Means and proportions are reported to summarize the characteristics of the survey population. For demographic and sociodemographic characteristics, results were disaggregated by region to capture variation across the four survey sites (Table 1).

Next, we examined the distribution of drought coping strategies across households. Drought coping strategies included diversifying income, borrowing money, forms of social support, resource conservation such as reducing meals or water, adapting farming practices, migration, and asset management. Comparative analyses were conducted to assess differences in food security outcomes (FCS and HHS) across household characteristics, including gender of the household head, age, primarily livelihood, region, marital status, and access to food aid, using t-tests for continuous variables and chi-square tests for categorical variables. To identify which factors influenced food insecurity, we estimated multivariate logistic regression models with the HHS as the outcome variable.

Finally, we present odds ratios from the regression models to assess the association between household coping strategy use and food insecurity outcomes. Each coping strategy was included as a predictor variable, and odds ratios with 95% confidence intervals were calculated to estimate the likelihood of moderate to severe hunger given coping strategy use. Odds ratios less than one indicate lower odds of moderate to severe hunger, while values greater than one indicate higher odds.

All analyses were conducted in R version 4.5.1.

3. Results

3.1. Who is at risk of food insecurity?

One year after the drought, 42% of people reported moderate to severe hunger in the household based on the HHS. 37% reported food insecurity based on the FCS metric. During the drought, 38% of the population of Namibia were categorized IPC as phase 3 or greater, meaning they were experiencing either a food crisis, food emergency, or famine. The results of the HHS and FCS show that this level of food insecurity was sustained for at least one year after the drought (ACAPS, 2025). Table 2 summarizes the distinguishing characteristics of individuals who experience food insecurity in northern Namibia based on the HHS (see SM1 for FCS results).

Table 2.

Associations between household characteristics and the Household Hunger Scale.

Variable	Test	p-value	Effect size*	Interpretation
Female household head	Chi-square	0.0069	0.1170	Female heads of households are more likely to report hunger than male heads of households.
Primary livelihood: agriculture vs not agriculture related	Chi-square	0.0042	0.1062	Households relying on agriculture related livelihoods are more likely to report hunger than those relying on non-agricultural livelihoods.
Region	Chi-square	<0.000	0.4507	Hunger prevalence differs significantly across regions.
Region duration: 0–3 years, 4–9 years, 10+ years	Chi-square	0.1587	0.1610	Duration of residence in the current region showed no significant association with household hunger.
Age of household head	T-test	<0.000	0.48	Household heads reporting little to no hunger tended to be older compared to household heads reporting moderate to severe hunger.
Income	T-test	<0.000	0.42	Lower-income households are significantly more likely to experience hunger than higher-income households.
Marital status	Chi-square	0.0136	0.12	Marital status is significantly associated with lower household hunger but the strength of the association is small.
Food aid access	Chi-square	0.0146	0.0906	Households with access to food aid are less likely to experience hunger compared to those not receiving food aid.

^*Effect sizes are reported as Cramer's V for chi-square tests and Cohen's d for t-tests.

There were several significant associations between demographic characteristics and the household hunger scale. Female-headed households were more likely to report hunger than male headed households (p = 0.0069) and households relying primarily on agriculture-based livelihoods were more likely to report hunger than those relying on non-agricultural livelihoods (p = 0.0042). Additionally, access to food aid was associated with lower reported hunger (p = 0.0146). However, for these comparisons, the effect size was small. We found a more substantial effect when considering region, age of household head, and income, which were all found to be significant at the 1% level.

Similar tests were run with the food consumption score (see SM1), but associations between household characteristics and the food consumption score were generally weak, with only two statistically significant variables. However, higher income and access to food aid were strongly associated with better food consumption (p < 0.000 and p = 0.0005, respectively), suggesting that access to money and external support may be important determinants of household food security in the study regions.

Regression results indicate that household income is associated household hunger (OR = 0.9997, p = 0.02), with lower income increasing the odds of household hunger. Households with marital status classified as other, which included separated, divorced, cohabitating, or otherwise non-traditional relationships, had significantly higher odds of experiencing hunger than single individuals (OR = 5.29, p = 0.01). Region can also be used to predict household hunger, with households in Ohangwena and Oshana, both regions with urban representation, having lower odds of hunger than Kavango West (reference region) (See Table 3).

Table 3.

Predictors of food insecurity based on the Household Hunger Scale.

Variable	Estimate (β)	Std. Error	OR = (exp(β))	95% CI (OR)	p-value
Intercept	2.956	0.911	19.22	3.85–95.79	0.001**
Household head	−0.632	0.456	0.53	0.23–1.23	0.165
Age	−0.005	0.012	0.995	0.97–1.03	0.694
Income	−0.0003	0.0001	0.9997	0.9995–0.99996	0.020*
Livelihood: Non-Agriculture	−0.169	0.375	0.845	0.42–1.68	0.652
Marital Status: Married	0.349	0.503	1.42	0.55–3.67	0.488
Marital Status: Widowed	−0.106	0.627	0.90	0.27–3.03	0.866
Marital Status: Other	1.666	0.646	5.29	1.54–18.13	0.010**
Region: Kunene	−1.228	0.683	0.29	0.08–1.03	0.072
Region: Ohangwena	−3.625	0.598	0.027	0.008–0.091	<0.001***
Region: Oshana	−3.611	0.625	0.027	0.007–0.097	<0.001***

Significance codes:

^***p < 0.001

^**p < 0.01

^*p < 0.05

Reference categories: Household head = male; Livelihood = Agriculture-based; Marital status = single; Region = Kavango West

3.2. What are people doing to cope with drought?

We found that the most commonly used strategy to cope with drought was diversifying income (24% by either starting a business, getting a part time job, or looking for informal sources of revenue such as casual labor or day jobs). Borrowing money (14%) was the second most commonly used strategy followed by social support (11%) which included applying for forms of government assistance, remittances, and sharing among community members (See Fig. 3). Resource conservation (i.e. skipping meals, storing food, and conserving or reducing water use), adapting farming practices (i.e. altering planting seasons or changing crops), migration, asset management (i.e. destocking or investing in drought tolerant seeds), and all other strategies were used less by less than 10% of the respondents per category. Importantly, 60% of respondents used at least one coping strategy and 17.5% of respondents used two or more coping strategies.

Fig. 3.

Actions respondents took to cope with drought.

Table 4 shows that the use of coping strategies during drought varied across some household characteristics. Coping strategy use did not differ significantly by gender, food aid receipt, income, or age of the household head. However, households engaged in non-agricultural livelihoods were more likely to report using coping strategies compared to agricultural households (p = 0.020). Single people were least likely to use coping strategies while married individuals were most likely to be using coping strategies. We also found regional differences (p < 0.001), with the highest coping strategy use reported in Ohangwena. For context, Ohangwena primarily engages in subsistence agriculture while Kunene is an arid region with primarily pastoralists.

Table 4.

Coping strategy use.

	Used coping strategies1	Did not use coping strategies1	p-value2
	N = 439	N = 291
Gender
Female	283 (64%)	190 (65%)	0.9
Male	156 (36%)	101 (35%)
Livelihood
Agriculture	150 (34%)	75 (26%)	0.02
Non-agriculture	289 (66%)	216 (74%)
Region
Kavango West	83 (19%)	28 (9%)
Kunene	56 (13%)	64 (22%)	<0.001
Ohangwena	216 (49%)	84 (29%)
Oshana	84 (19%)	115 (40%)
Region duration
0–3 years	23 (5%)	30 (10%)	0.018
4–9 years	44 (10%)	35 (12%)
10 years or more	372 (85%)	225 (78%)
Food aid
Received food aid	154 (35%)	113 (39%)	0.3
Did not receive food aid	285 (65%)	178 (61%)
Monthly income (NAD)*	1200 [450, 1900]	1000 [310,2200]	0.7
Martial status
Single	141 (32%)	125 (44%)
Married	166 (37%)	80 (28%)	0.052
Widowed	80 (18%)	43 (15%)
Other	57 (13%)	38 (13%)
Age of household head	58 [47, 72]	58 [48, 73]	>0.9

^*1200 NAD = ~$66 USD, 1000 NAD = ~$55 USD at the time of data collection (3/21/2025). Source: Authors' survey data.

¹n(%); Median [Q1, Q3]

²Pearson's Chi-squared test; Welch Two Sample t-test

3.3. Which strategies are associated with better or worse food security one year later?

Here we aim to be more specific about the usefulness of coping strategies to protect people against food-related impacts of drought. We find that coping strategies not always associated with protection against the food related impacts of drought.

Odds ratio analysis showed that while some coping strategies such as borrowing money and conserving resources, were associated with significantly lower odds of moderate to severe hunger, others were associated with higher hunger risk (See Table 5). For example, households that migrated as a strategy for avoiding the worst impacts of the drought had significantly higher odds of experiencing hunger. These findings suggest that while drought adaptation strategies are used, they vary in effectiveness.

Table 5.

Odds Ratios for the Association Between Coping Strategies and Household Hunger.

	respondents using the strategy (%)	Odds Ratio	Std. Error	p-value	95% CI Lower	95% CI Upper
Diversify income sources	23.6%	1.180	0.18	0.347	0.83	1.66
Borrow money**	13.7%	0.556	0.23	0.012	0.35	0.87
Social Support	11.0%	0.643	0.25	0.079	0.39	1.04
Resource Conservation***	8.6%	0.182	0.39	<0.000	0.08	0.37
Adapt farming practices*	8.0%	0.487	0.31	0.021	0.26	0.88
Migration**	6.6%	2.256	0.31	0.007	1.25	4.17
Asset management	5.9%	0.737	0.33	0.353	0.38	1.38
Other	4.5%	0.595	0.39	0.179	0.27	1.24

^*p <0.05

^**p <0.01

^***p <0.001.

The estimate for changed jobs is unreliable due to the small number of respondents reporting use of this strategy.

Odds ratios <1 indicates lower odds of moderate/severe hunger with that strategy; >1 indicates higher odds.

Examples of “resource conservation” include but are not limited to: reducing the number of meals or amount eaten in a day and reducing water use. Examples of “other” include but are not limited to: accessing savings or pension funds, purchasing more food from stores, and starting a garden.

3.4. Does fear of future food insecurity shape coping?

Regardless of current food security status or use of coping strategies, when asked about concerns for the future, respondents expressed concern for future food insecurity.

Fig. 4 illustrates that concern for future food security was widespread among both food-secure and food-insecure households. Among respondents classified as food secure by the food consumption score, 91% still expressed concern about future food access, while 88% of those not food secure reported the same (Fig. 4). Similarly, 95% of households experiencing moderate to severe hunger and 86% of those with little to no hunger based on the household hunger scale reported concern about their future food security (Fig. 4). These results suggest that fear of future scarcity is prevalent regardless of current food security status.

Fig. 4.

Concern for future food security by food consumption score and hunger status categories.

Fear of future food insecurity remained high regardless of whether households employed coping strategies. The highest levels of concern were observed among those who used two or more coping strategies, with 94% expressing worry about future food insecurity. Concern was similarly high among households not using any strategies (90%) and those using one (88%) (See Fig. 5).

Fig. 5.

Concern for future food security by coping strategy.

When examining individual strategies, households that diversified income, practiced resource conservation, managed assets, relied on social support, borrowed money, or migrated generally reported greater concern about future food insecurity; however, only the association with resource conservation was statistically significant. Those who adapted their farming practices were slightly less likely to be concerned (84.5% versus 90.3% among non-users), though fewer than 8% of respondents reported doing so. “Other” strategies, such as using savings or pension funds, purchasing more food, or starting gardens, were associated with the lowest levels of concern (69.7% versus 90.8%). While “other” was also significant, fewer than 5% of respondents reported using these approaches.

4. Discussion

Our study highlights that access to financial resources is an important factor shaping both coping strategies and food security outcomes in drought-affected households in northern Namibia. Income and food aid access were significantly associated with household hunger (p ≤0.000 and 0.0146 respectively), while the ability to borrow money was a commonly used coping strategy associated with lower household hunger (14% of households reported borrowing; OR = 0.556, p = 0.012). These findings suggest that strengthening households' access to financial resources may be useful in building resilience. This aligns with broader literature emphasizing financial access serves as a buffer that enables households to avoid coping strategies with negative long term impacts such as asset depletion (Gichangi and Gatheru, 2018; Savari et al., 2021). Relatedly, our findings show that certain coping strategies, including borrowing money, conserving resources, and adapting farming practices, were associated with reduced odds of hunger (OR/p = 0.556/0.012, 0.487/<0.000, 0.487/0.021 respectively), suggesting that households with the knowledge and resources to adapt flexibly can better buffer themselves against the negative impacts of drought. However, migration was associated with increased odds of hunger (OR = 2.256, p = 0.007), reinforcing that some strategies may exacerbate vulnerability. This may reflect the possibility that migration separates individuals or households from their established support networks, such as family, neighbors, or community-based assistance. Further research is needed to understand patterns of migration and its efficacy in buffering against climate-related shocks.

Demographic and regional differences also shaped vulnerability to food insecurity. Female-headed households were more likely to report hunger than male-headed households (67% vs. 53%, p = 0.0069), and agriculture-dependent households showed greater food insecurity than non-agricultural households (50% vs. 38%, p = 0.0042), echoing literature on gender barriers and the fragility of agriculture-based livelihoods in drought-prone areas. Regional variation in coping strategy use further illustrates the need for locally tailored interventions that consider the contexts of each community. Given the variation in coping strategy use across the regions, for example 49% of respondents in Ohangwena used coping strategies compared to 13% in Kunene, policies that assume a one-size-fits-all approach to adaptation are unlikely to successfully address the needs of all households, even within a geographic context. Integrating this regional nuance into drought policy design suggests that interventions like cash transfers or food aid may work better in regions with strong market linkages, such as in Ohangwena where there is more urban access, while community-based resource conservation programs could be more effective in areas with higher social cohesion, such as Kunene where reliance on social networks was high (Burchi et al., 2018; Della Guardia et al., 2022).

Despite the widespread use of coping strategies, concern about future food insecurity was nearly universal across households, regardless of their current food security status. Those who reported using multiple coping strategies were slightly more likely to express worry about the future than those who used no coping strategies at all (94% vs 90%). The persistence of fear despite widespread coping suggests that the most widely used strategies may be more reactive responses to immediate pressures, and people might not believe that these mechanisms will have long-term protective effects. Coping strategies that were associated with less concern for the future were underutilized. This finding could be the result of the concept of “climate traps” described by Nava et al. (2025), in which short-term survival behaviors reinforce long-term vulnerability, helping explain why high fear levels persist even among households actively using coping mechanisms or it could be simply that people believe their coping mechanisms are insufficient (Nava et al., 2025). Regardless, fear of future scarcity may play an important role in shaping household decision-making, where short-term actions to secure resources reinforce longer-term vulnerability, pointing to the importance of incorporating psychosocial dimensions into both research and policy design.

Finally, our findings complicate existing adaptation narratives. While diversification of livelihoods or crops is often promoted as a meaningful adaptation strategy, our data did not show diversification to be significantly associated with reduced hunger (24% of households practiced diversification; OR = 1.180, p = 0.347), despite this being the most frequently practiced coping strategy. This misalignment between policy discourse and household outcomes raises questions about the appropriateness of certain adaptation strategies in practice, adding to a growing body of literature questioning the universal promotion of diversification as a resilience pathway (Amfo and Ali, 2020; Coughlan de Perez et al., 2024). This causes us to question whether funding and adaptation are appropriately building on the existing practices utilized by communities such as adapting farming practices, which is associated with both improved food security and reduction in future concern. Future research should investigate alignment of funding allocations with autonomous adaptation.

5. Limitations

This study has several limitations. First, while widely used, the food security indicators applied here may not fully capture the nuances of food access and experiences specific to this context. Both the FCS and HHS measure acute food insecurity. The prevalence of food insecurity based on these indicators was shared with community leaders during a community engagement session. Feedback from community leaders during these sessions suggested that food insecurity is suspected to be more prevalent than indicated by the FCS and HHS measures. Second, reliance on self-reported survey data introduces the potential for recall bias and subjectivity in respondents' perceptions. Finally, while the study design allows us to identify associations between coping strategies and food security outcomes, it does not permit causal inference.

6. Conclusions

This study shows that financial access is central to how households in northern Namibia can most effectively cope with drought and its impacts on food security. Policies and programs should prioritize expanding access to cash and credit, while also ensuring that interventions reach populations who remain underserved, such as female-headed and agriculture-dependent households. For example, program design could differentiate between pastoral and agricultural regions by creating livestock insurance schemes that help herders maintain herds in pastoral areas while investing in and linking programs that increase access to credit with climate-smart agriculture initiatives to support investment in irrigation in agricultural regions. Across both contexts, integrating financial inclusion with mental health support would strengthen both the economic and emotional resilience of vulnerable households, given that our findings highlight the universal fear of future food insecurity. Risk communication and psychosocial dimensions of drought response require greater attention and further investigation.

Future research should examine how fear shapes coping decisions and whether it reinforces cycles of vulnerability. Longitudinal studies, even with only a few repeated measures, would be especially valuable to capture how household strategies, food security outcomes, and fears evolve over time. Finally, preparedness should be framed not as a responsibility borne by individual households, but as a systemic challenge that demands stronger support structures capable of serving drought-affected populations more effectively.

Data availability

Data will be made available on request.