This research was informed by the Theory of Risk and Uncertainty and the Conflict Theory of change. The theories jointly offer explanation for the relationship between the cattle rustling related insecurity and community livelihood. Whereas theory of Risk and Uncertainty explains how households cope with and make livelihood choices under conditions of insecurity and uncertainty, Conflict Theory provides the rationale why insecurities thrive. They together can offer a logical link for assessing how insecurity affects community livelihoods.

The Risk and Uncertainty theory, originally developed by Frank Knight in 1921, separates risk, which is the situation where probability of occurrence of events in the future can be estimated, from uncertainty, where probabilities are largely unknown [14]. Households in pastoral environments usually make decisions under circumstances that are insecure and uncertain in relation to access to resources and environment. According to Ibrahim [15] repeated cattle rustling in these areas affects the existing systems of livelihood and makes it difficult to predict the status of livestock ownership and sustenance of livelihoods. Similarly, Shanguhyia [16] has reported that continued cattle rustling and banditry in North West Kenya makes the households insecure, which affects decision-making process in the utilization of resources.

Relevance of the theory to the present study emanates from its capability to describe the behavioural reactions of pastoralist families in face of insecurity situations. Families who have a high perception of the risk of being attacked tend to avoid places that they consider risky such as some grazing lands, migration pathways or markets. Similarly, insecurity may limit the movement of families hence making it difficult for them to access water, pastures and other means of livelihood. The level of safety perceptions affects the degree to which families participate in productive activities. However, although the theory describes the response of families to insecurity situations, it fails to describe the circumstances under which insecurity is created. This creates the need for inclusion of Conflict Theory.

According to Conflict Theory, which was advanced by Karl Marx in 1976, conflict is as a result of competition over scarce resources and differences in power and opportunities [17]. In modern research, the theory has been used to examine insecurity in pastoral settings. For instance, Marigat [18] discovered competition for scarce resources, poverty, unemployment, small arms, and cultural issues to be some of the causes of cattle rustling and insecurity in North West Kenya. Likewise, Bello and Abdullahi [19] observed that conflicts caused by livestock rustling and resource competition are highly related to struggles over grazing and water resources.

The Conflict Theory, within the scope of this study, helped explain the circumstances under which insecurity from cattle rustling is perpetuated. On the other hand, the Theory of Risk and Uncertainty will be used to explain how households respond to such insecurity during their livelihood practices. The competition for grazing areas, water sources and livestock results in circumstances under which cattle rustling continues to exist, resulting in fear, lack of mobility and lack of safety perception. Ultimately, this impacts livelihood including herd size, access to water and grazing sources, market access and marriage culture. The convergence of the two theories thus explains and offers a complete explanatory framework of the link between livestock rustling-related insecurities and community livelihood diversity of pastoral households.

The literature reveals that cattle rustling in Kenya has gone through a remarkable change in terms of development from being a cultural activity to a multi-faceted problem relating to security and survival. The history of cattle raiding is linked to some social and economic activities such as restocking of the livestock, wealth distribution, payment of bride wealth, and enhancement of social prestige within the pastoral communities. As indicated by Akama and his colleagues [20] the above activities used to be conducted through customary structures that defined how and to what extent raids were supposed to be done. Colonial administration led to the disruption of customary grazing arrangements through the introduction of borders, tribal reserves, and pastoral movement.

One of the areas that the literature shows agreement on is that there was a fundamental change in the nature of cattle rustling in the latter half of the twentieth century. As pointed out by Kimani [21], the 1990s saw the rise of organized cattle rustling, which was accompanied by violence and involved the sale of stolen livestock through the market. In a similar vein, Amutabi and Hamasi [22] contend that cattle rustling became increasingly economic rather than cultural and formed part of a broader network of criminal activities. The rise in small arms due to the instability of neighboring countries and the resulting increase in scarcity of resources led to cattle rustling becoming a far deadlier crime involving death and destruction of property [18]. However, despite their agreement, there are some differences in regard to the causes of cattle rustling in the contemporary times. While the authors view the continuation of cultural practices like the payment of dowry, circumcision, and other cultural issues associated with cattle keeping as being responsible for the persistence of cattle rustling. Kimani [21] views commercialization as an increasingly influential cause of cattle rustling. In a similar vein, Kamais and Mosol [23] point to political incitement, poor governance, resource competition, and collapse of traditional institutions as key causes of cattle rustling in the modern times.

While the literature is abundant with information on the history and causes of cattle rustling, some issues have yet to be clarified. The current body of literature emphasizes the causes and security impacts of cattle rustling and does not provide sufficient information on how these changes in history affect livelihoods at the household level. Furthermore, while there are numerous studies on cattle rustling at more regional levels, there is a lack of specific evidence from places like Baragoi, which continues to experience insecurity. Literature provides abundant proof of the fact that cattle rustling is a major security issue nowadays, but little evidence is provided regarding the influence of such insecurity on livelihoods because of fear, lack of mobility, and sense of insecurity.

Recent scholarship increasingly demonstrates that the consequences of cattle rustling extend beyond the direct loss of livestock to encompass broader forms of insecurity that affect community livelihood. Rather than functioning solely as an economic crime, cattle rustling has evolved into a complex security challenge that influences access to resources, market participation, mobility, social relations, and overall livelihood wellbeing. Across the contemporary literature, three interrelated themes emerge consistently: fear of attack, restricted mobility, and perceived safety. Although scholars generally agree that cattle rustling-related insecurity undermines livelihood outcomes, there are notable differences regarding the mechanisms through which these effects occur and the dimensions of livelihood most affected.

Fear of attack emerges as one of the most significant consequences of cattle rustling-related insecurity. Ibrahim [15], in a study examining cattle rustling and livelihood transformation in Nigeria, found that recurrent cattle raids disrupt traditional pastoral livelihoods and generate uncertainty that affects household wellbeing and livelihood adaptation. Similarly, Ojo et al. [24], in their study about armed banditry and insecurity in North-west Nigeria, established that persistent attacks contribute to displacement and declining agricultural productivity by exposing communities to continuous threats. Both studies converge on the view that fear associated with recurrent violence constrains livelihood activities and weakens household resilience. Comparable findings were reported by Chemase and Muhindi [25] in their research about the implications of banditry on socio-economic development in Baringo South Sub-County, Kenya. The authors observed that contemporary cattle rustling has evolved into a serious security threat characterized by loss of life, destruction of property, and prolonged insecurity. Their study identified poverty, unemployment, competition over scarce resources, proliferation of small arms, and cultural practices as factors sustaining cattle rustling and insecurity in the region. Unlike Ibrahim [15] who emphasizes livelihood transformation, Chemase and Muhindi [25] place greater emphasis on the broader security implications of cattle rustling and its threat to human security. Despite these differences, all three studies agree that fear generated by recurrent attacks significantly disrupts normal livelihood activities.

Restricted mobility is yet another prominent theme in the literature. Free mobility is an essential attribute of the pastoral way of life since access to grazing grounds, water sources, migration routes, markets, and social networks is dependent on free movement of the households. Defere et al. [26] observed that insecurity resulting from cattle rustling is responsible for restricting household mobility and limiting access to productive resources. The same was observed by Anno and Ameripus [27] in their analysis of resource-based conflicts experienced by livestock traders in pastoral settings of Kenya and Uganda where the conflict had a significant impact on access to grazing resources, water sources, and livestock production possibilities. A clear trend emerging from these studies is that insecurity causes restricted mobility and, as such, is a factor limiting access to productive resources for pastoral livelihoods. What differentiates the two studies is that while one emphasizes displacement and reduced production as major effects of restricted mobility, the other focuses on disruption of livestock production systems and market integration.

The literature further identifies perceived safety as an important determinant of livelihood wellbeing. Ibrahim [15] argues that persistent insecurity weakens household wellbeing by creating uncertainty regarding the protection of productive assets and future livelihood prospects. Similarly, Amutabi and Hamasi [22] found that insecurity discourages market participation by creating uncertainty among traders and producers, thereby limiting access to economic opportunities. Both studies imply that perceptions of safety influence the willingness of households to engage in productive activities and invest in livelihood strategies. A recurring pattern across the literature is that communities characterized by low levels of perceived safety tend to exhibit poorer livelihood outcomes than communities operating within relatively secure environments. Nevertheless, a notable limitation is that most studies treat safety as an indirect outcome of insecurity rather than measuring perceived safety as a distinct analytical construct. Consequently, the role of subjective security perceptions in shaping livelihood decisions remains insufficiently explored.

Further, the literature points out that perceived safety is one of the significant factors driving livelihood wellbeing. The contention of Ibrahim [15] is that chronic insecurity negatively impacts household well-being through uncertainty about the safety of one’s productive assets, and livelihood future. Similarly, Amutabi and Hamasi [22] discovered insecurity in their study results as the deterrent of participation of both producers and sellers to local markets because it instills uncertainty of business. In northern Kenya, a recent study conducted by Schilling and Werland [1] indicates that there are overlapping threats affecting pastoral communities’ perceptions about security status in the area with decreased mobility and reduced access to important water and grazing resources, as well as eroded security needed for sustenance of livelihoods.

These analyses signal that the level to which communities assess security impacts the behavior of engaging in productive activities and securing livelihood resources. Therefore, security and peace are crucial determinants of livelihood sustenance. In most studies reviewed there is overwhelming evidence indicating that, generally households based in insecure environments has lower livelihood wellbeing compared with those inhabiting the relatively safer contexts. Nevertheless, many scholars have either regarded security or, viewed perceived security as the outcome or by-product of insecurity instead of as a significant factor affecting the decisions made regarding livelihoods. The implication of this perspective is the dearth of evidence that provides a clear answer on what shape household decisions regarding livelihood, and wellbeing by communities experiencing recurrent Cattle rustling have as per their level of their perception of safety.

Several areas of agreement emerge across the reviewed studies. First, there is broad consensus that cattle rustling-related insecurity affects livelihoods through mechanisms that extend beyond livestock losses. Second, the studies consistently demonstrate that insecurity disrupts access to productive resources, limits economic participation, and undermines household wellbeing. Third, there is agreement that cattle rustling has increasingly evolved from a traditional livestock acquisition practice into a more complex security challenge characterized by violence, displacement, and livelihood disruption. These patterns are evident across studies conducted in Nigeria, Kenya, and cross-border pastoral regions.

Despite these areas of convergence, important disagreements and inconsistencies are also apparent. Ibrahim [15] emphasizes livelihood transformation and declining wellbeing as the principal outcomes of cattle rustling, whereas Anno and Ameripus [27] concentrate on resource access and livestock production, while Amutabi and Hamasi [22] emphasize market access and economic participation. Similarly, Chemase and Muhindi [25] frame cattle rustling primarily as a national security challenge. These variations suggest that the consequences of cattle rustling-related insecurity are context-specific and may differ according to environmental conditions, livelihood systems, and conflict dynamics.

There are also notable limitations in terms of research methodology. Current studies are largely based on review-based, descriptive and secondary sources of information, which prevents them from exploring the experience of households related to insecurities and consequences of insecurities on livelihoods. Most of the research is focused on the changes in livelihoods, resource conflicts, or livestock production, while paying little attention to such aspects of insecurity as the fear of being attacked, restrictions on movements, and safety. Overall, the current literature gives an incomplete picture of the relationship between insecurities and pastoral livelihoods.

A further limitation concerns geographical context. Most of the reviewed studies were conducted in Nigeria or in broader pastoral settings within East Africa. Although these studies provide valuable insights into the relationship between insecurity and livelihood, their findings may not adequately capture the realities of Baragoi, where cattle rustling is shaped by unique historical, cultural, ecological, and socio-economic factors. Moreover, empirical studies explicitly examining fear of attack, restricted mobility, and perceived safety as dimensions of cattle rustling-related insecurity remain limited within northern Kenya.

The reviewed literature therefore demonstrates a strong consensus that cattle rustling-related insecurity constitutes a significant livelihood challenge. However, important conceptual, methodological, and contextual gaps remain. In particular, limited empirical evidence exists on how fear of attack, restricted mobility, and perceived safety collectively influence community livelihood among pastoral households in northern Kenya. This study addresses these gaps by examining the association between these dimensions of cattle rustling-related insecurity and community livelihood in Baragoi Sub-County, thereby contributing context-specific evidence to the growing literature on pastoral insecurity and livelihood outcomes.

The conceptual framework was instrumental in guiding the study by providing a basis for investigating the relationship between insecurity associated with cattle rustling and community livelihood in pastoral communities. The conceptual framework for this study draws on Theory of Risk and Uncertainty [14] and the conflict theory [17], which combined together guide us how the insecurity occasioned by cattle rustling affect household livelihoods among the pastoral communities as shown in Figure 1Figure 1. Theory of Risk and Uncertainty defines household approach towards decision making under insecurity and unknown circumstances while conflict theory define the how contest over scarce resources lead to conflict which gives rise to cattle rustling and in security it triggers.

Figure 1.Conceptual framework.

The framework defined cattle rustling related insecurity as an independent variable by capturing its elements such as fear of being attacked, limited mobility, and perceptions about security while community livelihood was defined as a dependent variable through elements like herd size, water and pasture availability, market access, and marriage practices. The framework was helpful in defining the research objective, in designing instruments for data collection, and the variables used in the analysis.

This study was anchored on the assumption that insecurity associated with cattle rustling affects livelihood through various channels. Fear of being attacked prevents the community from engaging in any livelihood practice, limited mobility affects access to necessary livelihood resources and opportunities, while perceptions of being safe enhance productivity. Thus, household members with high levels of insecurity would be predicted to have poorer livelihood outcomes than those that had low levels of insecurity. The model hence acted as a tool that was used in the analysis of how fear of attack, restricted movement, and perception of security affected livelihood in communities.

The socio-demographic factors were investigated to give a description of the research sample and to ascertain whether the respondents had the requisite experience in households and livelihoods to give credible responses regarding cattle rustling and insecurity.

The data collected in Table 1 indicates that the respondents were mostly male and majority represented young age groups. More than half (46.1%) of the respondents were 18 - 26 years old, and 48.9% were married; therefore, many of them had responsibility for household affairs and livelihood activities. Furthermore, most of the respondents were from medium and large families, as about three quarters indicated that their family size was four members and above. This is typical to pastoral communities because the respondents had experiences and information on the issue of cattle rustling insecurity.

Table 1. Socio-demographic characteristics of respondents (n = 321).

Community livelihood served as the dependent variable and was measured as a composite variable composed of four indicators, which include: herd size, water and pasture resources, market access, and marriage customs. The respondents provided details about these indicators based on their household’s experiences within the study community. The four indicators were coded through structured answers and expressed as standard scores for purposes of comparisons. The scores of the four indicators were combined to create a Community Livelihood Index (CLI), which was created by calculating the average of the four indicators per respondent. The method involved giving equal weight to all the indicators of community livelihood.

Community Livelihood Index (CLI), computed from the four livelihood indicators outlined in Chapter 3 was used as the dependent variable in this research. Households fall into 3 livelihood groups, high, moderate and low. Table 2 presents frequency distribution of respondents in respective livelihood groups based on CLI. Later CLI was used to carry out crosstabulation chi-square analysis, ordinal logistic regression analysis, in exploring the relationship between cattle rustling and community livelihoods.

Table 2. Distribution of community livelihood levels.

According to the findings, the majority of those sampled (45.2%) belong to the moderate livelihood category. Those belonging to the low livelihood category are represented by 30.5%, while those belonging to the high livelihood category are represented by 24.3%. This distribution helps in analyzing the differences in livelihood outcomes with respect to insecurity caused by cattle rustling.

Participants were asked to articulate their perceptions of the insecurities associated with cattle rustling in terms of fear of attack, lack of movement, and safety concerns. Responses were then classified into three ordinal levels (low, medium, and high) depending on intensity and frequency. From the distribution of answers, it can be clearly seen that there is a significant pattern of high insecurity within the region under study. As for the fear of attacks, 44.2% of the respondents showed high levels of fear, while only 18.1% showed low levels, implying that insecurity from cattle rustling is widespread. In terms of restricted movements, the same pattern emerges, with 42.1% of the respondents showing high levels of restricted movements and 17.7% reporting low restriction. Perceived safety is also shown to have a similar trend, with 46.1% of the respondents reporting low safety in their day-to-day activities (see Table 3).

Table 3. Descriptive results for cattle rustling-related insecurity indicators.

Analysis of the frequency of responses shows that there is a common trend in terms of insecurity. With regard to the fear of being attacked, 44.2% of respondents have reported experiencing high fear, while 18.1% of them have experienced low fear. This shows that people in Baragoi sub-county experience insecurity due to cattle rustling. In terms of restricted movements, 42.1% have experienced high restrictions, while 17.7% have experienced low restrictions. With regards to the sense of security, 46.1% of respondents have shown low sense of security. From the above analysis, insecurity is commonly experienced in the area due to cattle rustling.

To analyze the association between insecurity due to cattle rustling and the livelihood of the communities, a cross tabulation test and Chi-Square Test of Independence were conducted where the livelihood levels of the community were classified into low, moderate, and high livelihoods depending on the composite livelihood score.

Table 4 presents the distribution of respondents across livelihood categories according to their reported levels of fear of attack.

Table 4. Cross-tabulation between fear of attack and community livelihood.

χ² (4) = 31.824, p < 0.001.

The findings show clear disparities in the livelihoods experienced by various levels of fear of attack. Of those respondents who indicated high levels of experience of attack situations, 44.4% were found in the low livelihood group, while only 12.1% of those respondents who indicated low levels of experience of attack situations fell in the low livelihood group. Likewise, respondents who indicated high levels of experience of attack situations were found to be in the high livelihood group at 15.5%, while those who reported low experience of attack situations were in the high livelihood group at 41.3%.

A further concern in this study is whether the observed differential outcomes are explained by differential movement restriction due to cattle rustling-related insecurity. The results for this analysis are in Table 5.

As can be seen in Table 5, there were major differences in livelihood outcomes for different degrees of mobility restrictions. Close to 44% (43.7%) of the participants with high degrees of movement restrictions belonged to the low livelihood category while only 14% (14.0%) of the participants who reported having low mobility restrictions belonged to the same category. On the other hand, those that were categorized into the high livelihood category increased from 15.6% to 43.9%. The results indicate that higher mobility restrictions led to relatively poor livelihood outcomes. Chi-square showed a statistically significant association between the two variables.

Table 5. Cross-tabulation between restricted mobility and community livelihood.

χ² (4) = 34.271, p < 0.001.

The last dimension considered was perceived safety. Table 6 indicates the distribution of livelihoods categories based on respondents’ perception of safety in their communities.

Table 6. Cross-tabulation between perceived safety and community livelihood.

χ² (4) = 29.115, p < 0.001.

The result shows that there is a similar trend in relation to fear of attack and restricted mobility. The percentage of the people who perceived low levels of safety and had low livelihood outcome status was 43.9%, while the percentage of the people who reported high levels of safety and had low livelihood outcome status was 13.2%. On the other hand, the percentage of people who perceived low level of safety and had high livelihood outcome status was 14.9%, while the percentage of the people who reported high levels of safety and had high livelihood outcome status was 37.7%.

From the findings, a clear trend has been observed from all three dimensions of insecurity in relation to cattle rustling. Those respondents who recorded high levels of fear of attacks, more movement restrictions, and less perception of safety have often been found in the low livelihood group. On the other hand, those respondents who recorded lower levels of insecurity often belonged to the high livelihood group. The significant Chi-square findings show that there is an association between cattle rustling insecurity and livelihood in the community and need further investigation using multivariate analysis. Among the three measures, restricted mobility showed the highest association with livelihood in the community, followed by fear of attack and safety perception.

From the bivariate analysis, there were some strong relationships established between the following factors—fear of attack, restricted mobility, perceptions of safety, and community livelihood. But these analyses did not take into consideration the joint impact of all these insecurity dimensions. For that reason, ordinal logistic regression was carried out to test the relationship of one factor with community livelihood.

Diagnostic tests were performed before modeling to establish whether the data were fit for applying the ordinal logistic regression. Pearson correlation coefficients between the independent variables were carried out in an effort to detect the presence of multicollinearity among the predictor variables. The highest correlation between any of the predictors is between restricted mobility and perceived safety (r = 0.72) which is less than the recommended 0.80, therefore, indicating the absence of multicollinearity. In addition, the parallel lines test was conducted in order to validate the assumption of proportional odds resulting in non-significant results (χ² = 5.741, p = 0.453). Therefore, the assumption of proportional odds holds. For this study, higher values of fear of attack and restricted mobility denote higher levels of insecurity while higher values of perceived safety represent high levels of security.

The results of the analysis showed that the model was statistically significant (χ² = 47.686, df = 3, p < 0.001). This shows that there is a statistically significant association between cattle rustling insecurity and community livelihood. The model did not violate the assumption of proportional odds (χ² = 5.741, p = 0.453). The Nagelkerke R² was 0.196, meaning that the model explained 19.6% of the variance in community livelihood. Results of the ordinal logistic regression are presented in Table 7.

Table 7. Ordinal logistic regression results for community livelihood.

Model Fit Statistics: Model χ² = 47.686***; Degrees of Freedom = 3; Nagelkerke R² = 0.196; Cox & Snell R² = 0.138; McFadden R² = 0.117; Test of Parallel Lines (p-value) = 0.453. ***p < 0.001.

Results indicate that there is a significant relationship between fear of attacks and community livelihood (β = −0.684, OR = 0.50, 95% CI = 0.36 - 0.70, p < 0.001). Individuals with higher fear of attack had 50 percent lower odds of being classified under a high livelihood category compared to a low livelihood category relative to those who reported low levels of fear. Restriction of mobility is significantly associated with community livelihood (β = −0.741, OR = 0.48, 95% CI = 0.34 - 0.68, p < 0.001). Individuals with high restriction of mobility have 52 percent lower odds of being classified under a high livelihood category compared to a low livelihood category relative to those who had low restriction of mobility. Among all the predictor, restricted mobility was the most influential variable, as shown by the highest coefficient value and Wald statistic. Safety was positively and significantly related to community livelihood (β = 0.592, OR = 1.81, 95% CI = 1.32 - 2.49, p < 0.001). Respondents with high safety had 1.81 times higher odds of being classified under a high livelihood category than those with low safety.

Based on the ordinal logistic regression results, there appears to be a significant association between cattle rustling-related insecurity and community livelihood. This association varied depending on the type of insecurity measure considered. Increased fear of attack and movement restrictions were found to be associated with increased probability of being categorized in lower livelihood statuses, while increased feelings of safety were positively associated with better livelihood status. In terms of predictor variables, the variable of movement restriction was found to be the most significant correlate of community livelihood. The general conclusion from these results is that increased fear of attacks and movement restrictions led to the increased probability of falling into lower livelihood statuses while increased feelings of safety increased the probability of falling into higher livelihood categories.

This is further corroborated by the qualitative information provided by the key informants. For instance, one key informant pointed out that the social expectations about marriage are the cause of cattle rustling in the community because most young people are supposed to own enough cattle before getting married:

“Cattle rustling can by and large be attributed to marriage expectations; when the time for a moran to marry closes up, he is usually required to have a good number of cattle to show that he is a responsible young man capable of taking care of his family,” (KII02, Field Data, 2024)

This information proves that the continued practice of cattle rustling has perpetuated insecurity and fear of attack in the community. Also, another key informant stated that it is due to the cultural significance associated with the acquisition of cattle through cattle rustling that most of the people without cattle engage in the practice to acquire them to complete the ritual:

“The need to go through this rite encourages those without cattle to go raiding to get cattle to undergo it. Without a marriage ceremony, one becomes a laughing stock in the community.” (KII02, Field Data, 2024)

Combined, the results of the study suggest that insecurity caused by cattle rustling is closely linked to livelihood and that fear of being attacked, lack of mobility, and security are significant pathways through which livelihoods differ among the pastoral communities.