The economic activities among mining-induced displacees in Arda Transau, Zimbabwe

Abstract

Mining is one of the leading causes of displacement worldwide and leads to disruptions of household economic activities. With the increase in demand for minerals, the world over by manufacturing industries such as those of jewellery and cellphones and therefore rise in mining activities, displacement of people becomes inevitable. After displacement, households still have to survive and therefore are forced to find alternative ways to make ends meet. The households explore several choices including what to engage in, when to do specific activities and in which combination. This study sought to identify the determinants of the choice of livelihood activities among Arda Transau internally displaced persons (IDPs) using a multinomial model. It also sought to determine the extent of agricultural diversification among these IDPs. The study ascertained that gender of household head, economic activities done by households before displacement and marital status of household head are significant determinants of choice of economic activity among the IDPs. It showed limited agricultural diversification among IDPs in Arda Transau and observed that the age of household head, gender, religion and nature of previous economic activity are important determinants of the activity chosen. The study recommends crop diversification to minimize risks and calls for reduced use of erosive coping strategies.

Introduction

Mining is one of the leading causes of displacement worldwide (Teminisk, n.d) whose attendant consequences lead to disruption of household economic activities. With the increase in the quest for minerals and therefore mining activities, displacement of people becomes inevitable. Although a considerable amount of literature exists showing that mining-induced internal displacement forces households and individuals to find other ways to sustain themselves other than those used prior to displacement (Barbelet 2017; Abraham et al. 2018 a troubling aspect of mining-induced displacement perhaps is its general approval in several parts of the world.

There is evidence that economic activities among internally displaced persons (IDPs) differ. Among the many economic activities done by IDPs, agriculture usually dominates (Adeniyi et al. 2016). In some instances, trading (for example, firewood selling) and asserts disposal become common coping strategies (Adeniyi et al. 2016). These activities at times may be gendered (Amisi 2006; Dolan 2004) with male and female IDPs employing different livelihood strategies. Several reasons have been given for this disparity, for example, DRC women refugees in South Africa reported that they could not choose working as security guards because of the inherent risk and working conditions (Amisi; 2006). Dolan (2004) also noted that differentials in access to productive resources also contribute to differences in livelihood strategies among male and female-headed households.

It can, therefore, be noted that after displacement, households must find other ways of survival because the ones used prior to displacement may no longer be possible. Possible reasons may be the difference in the environment, soil and climate, for instance when agriculture is to be employed as the main economic activity since the climate and soil type must be suitable for that activity. In such instances, other means of survival have to be devised. Therefore, in recent years, there has been a worldwide increased interest in internal displacement and its consequences and the choices of livelihood activities among IDPs.

Along with the aforementioned claims, it can again be noted that since displacement is inevitable in development project areas and its consequences are inevitable as well, the necessary question to be asked is “How do the IDPs make a livelihood faced with these consequences?”

Research interest among the majority of existing literature concentrated on coping strategies of camp residents excluding that of IDPs who are there to stay like those in Arda Transau. Among the majority of research that has concentrated on economic activities of IDPs in IDP camps include Bozzoli et al. (2011). However, Bozzoli et al. (2011) acknowledged that little attention has been paid to economic activities of the displaced households. The other avenue concentrated on, by previous researchers, is the coping strategies of those who have experienced conflict, for instance, civil war like those in Mozambique (Bruck; 2004 and Bruck and Danzer; 2007) and Uganda, (Bozzoli et al. 2011). And even more, literature also concentrates on coping strategies of refugees, for instance, Amisi (2006) and Khawaja et al. (2008).

The case of Arda Transau is unique in that the IDPs are mining-induced and are not in camps but are in the newly resettled area with no possibility to return or for being displaced again in the near future. It is, therefore, the intention of this study to add to existing literature by focusing on economic activities of this previously excluded and rare group of IDPs since little is known about them. Also, most of the existing literature on coping strategies has largely ignored the determinants of choice of particular activities but concentrated more on how the strategies are employed for instance, Rashid et al. (2006), Bello et al. (2014) and Israel and Briones (2014). Thus, the other contribution of this study is on what exactly determines the choice of use of particular livelihood activities among mining-induced IDPs in Arda Transau. Several features distinguish mining-induced displacement from other development-induced displacements. One such feature is that mining-induced displacement sometimes takes a step-wise approach to displacement (Downing 2014). Owen and Kemp (2015) also noted that mining-induced displacement and resettlement is unique in that the displaced and the mining companies may cohabit. It was also noted that because of resource constraints, most governments relegate the responsibility of managing resettlement to the same companies that led to displacement of people.

Additionally, this study contributes to existing literature in that mining-induced displacement has been understudied with dams and other development-induced displacements having been well researched, despite the fact that mining-induced displacement is significant and also has consequences that affect the IDPs (Hoadley 2008). Additionally, in Zimbabwe mining-induced displacement has led to a large number of displacees and constitutes a major forced migration problem worldwide (Terminski n.d.).

Therefore, the main objective of this study was to identify and examine the determinants of choice of economic activities or coping strategies among internally displaced families in Arda Transau from an individual household livelihood perspective. Additionally, the study intended to establish the extent of diversification and determinants of agricultural coping strategies among the IDPs.

Understanding the economic activities of IDPs is important for two main reasons. Firstly, unfettered acceptance of mining-induced displacement may not only lead to economic risks such as unemployment but also disruption of household economic activities like agricultural activities. Secondly, the knowledge of type and combination of economic activities among the displaced enables policy-makers to assess the probability of rehabilitation of the displaced.

Literature review

Displacement has consequences that stretch for generations including joblessness, landlessness, marginalization, food insecurity, social disarticulation and homelessness, among others (see Cernea 1997). The strategies used to overcome these consequences of displacement may have been acquired prior to displacement but may also be acquired in the post-displacement period. Thus, in such instances where the formerly used means of survival cannot be used in the post-displacement location, adoption of new strategies is imminent thereby bringing in coping strategies which may be considered as risk management approaches.

Deressa et al. (2010) indicate that risk management under life-stressing situations like drought mostly come in two forms, viz, at household-level and public-level. At the household level, coping strategies include disposal of assets (divestment) including productive assets and animals, employing child labour, depending on relatives and migration, among others. Of note is that some of these strategies have long-term effects especially the impoverishment of IDPs. For instance, animal disposal is a form of disinvestment whereby the family may not be able to buy-back the once owned animals (Bird et al. 2002) and hence may live in abject poverty forever. Child labour may deprive the children of education (Loewenson 1991; ITUC 2008; ZimSTats 2015) thereby affecting the household’s future human capital. The human capital deficiency will furthermore affect the household’s ability to cope in the long run. In this context, child labour takes place if the child is engaged in an activity which contributes to household income unlike where the child engages in household chores. Therefore, Deressa et al. (2010) indicate that household-level strategies are mostly ineffective because they have proved to be costly yet they achieve partial insurance. It is possibly considered to be partial insurance in the sense that the household level strategies are unsustainable and impoverish the said household in the long run.

At government level, strategies such as free food distribution and food for work programmes, among others, have been used. In such cases where free food distribution is used, the affected households may end up in a dependency syndrome where they will expect the government to undertake actions without them taking positive roles, thus the households end up using passive coping strategies. The free food distribution, however, may be applied well to the old aged households who if required to do work, for example under the food for work programmes, may no longer be able to do so. Deressa et al. (2010) noted that, in Ethiopia, food aid has become the most widely used strategy to cope with drought. On the other hand, the government level risk-mitigation approach suffers from the limitation of coverage and dependency on foreign aid, among others. However, if applied well, government level strategies may benefit more individuals than household-level strategies.

The risk management strategies used by IDPs can also fall into two broad groups viz. ex-ante and ex-post strategies (Bruck and Danzer 2007). “Ex-ante mechanisms address what households (and to that extent, public and private instruments) can do to reduce or prevent the occurrence of risks and mitigate the impact of risk if an adverse event occurs” (Acevedo 2016, 2) and is a forward-looking risk management concept (Mukarami 2017). Acevedo (2016) indicates that this approach can be considered as income smoothing. One example of income smoothing is income diversification where individuals may find ways to get income from various activities such as engaging in agriculture as the main economic activity but at times working in other people’s fields thereby supplementing income earned from sale of agricultural produce. At the same time, purchasing of animals such as cattle for sale when faced with a shock can be considered as an ex-ante livelihood diversification strategy. In this regard ex-ante strategies are considered to be proactive in nature because they are instituted well before the unforeseen event occurs therefore, Mukarami (2017) refers to ex-ante mechanisms as forward-looking strategies.

On the other hand, ex-post strategies may be used. In this approach, households may build asserts in ‘good’ years so as to sell in ‘bad’ years (Dercon 2000). Ex-post strategies are meant to deal with the consequences at hand hence may be considered short-term in nature. Another example of ex-post strategies is reliance on relief agencies such as Care International and Oxfam. Dercon (2000) highlights that some ex-post strategies are used as ways of earning extra income. For instance, individuals may gather wild fruits for own consumption as well as selling. Also, increased labour supply may be used as a coping strategy which may earn an income to the household. Another way of using ex-post strategy is through self-insurance by precautionary saving. In this instance, households may save for unforeseen eventualities such as crop failure. Consequently, households will have something to fall back on when faced with such eventualities. However, Dercon (2000) highlights that self-insurance is far less useful because of the covariance between assets and income. Covariance between assets and income is noted in that when incomes are high, generally, economic agents acquire wealth and vice versa.

On the other hand, consumption smoothing can be applied, particularly as an ex-post risk-management strategy, under forced migration circumstances. Along these lines, households stretch their food stock to defend the current consumption levels by reducing meal size and being involved in various activities (Lekprichakul 2009). However, this meal portion reduction may work as a short-term approach. Another way of smoothing consumption may be taking children to relatives so that the available food in the household can take the particular household longer. During that period other ways of surviving will be explored. This is a form of transferring risk of food shortage as well as risk-sharing. The other example of consumption smoothing is where a household may engage in sale of assets with the intention to make resources available to maintain a certain level of consumption.

Acevedo (2016) highlights that missing or incomplete markets may become obstacles to the use of ex-post strategies. According to Acevedo (ibid), this makes decisions by economic agents to be suboptimal. Accordingly, Acevedo (ibid) noted that households may depend on income smoothing rather than its counterparty consumption smoothing (an ex-post strategy). For instance, in Arda Transau, access to credit from even microfinance institutions is not available. Therefore, should any household want to borrow they have to rely on their neighbours and relatives, a type of loan that Rashid et al.( 2006) calls unsecured loans, therefore reliance on social capital becomes more important. The absence of markets, therefore, implies that even if households want to use credit to smooth consumption they cannot do so.

Another coping strategy used by IDPs is as explained by the ‘added worker effect hypothesis’. “The added worker effect hypothesis states that in the eventuality of a shock on the primary earner in the household, secondary workers would enter the labour market as imperfect substitutes to smooth consumption profile at the household level” (Martinoty 2015, 2). In this case, households may turn to, say, child labour should a household find itself in a shock such as food insecurity and unemployment. For instance, households relying on firewood sales, in Arda Transau and elsewhere, may engage children in the ferrying of the firewood. One other particular instance where child labour may be applied is where households may sell their labour (labour exchange) for instance cultivation of other people’s fields in which case the family members take the children, who are not yet in the labour market, along with them. The added worker hypothesis has been found to apply in Bangladesh where culturally women are not allowed to wage labour but because of displacement they were found as workers (Islam 2008). However, the ‘added worker effect hypothesis’ is likely to be small because, for instance among IDPs in Arda Transau, loss of employment by the household head will not necessarily translate to the other members getting equally paying jobs. Additionally, in a life-cycle model, the added worker effect may not arise. This is because, in the life cycle model, the loss of income by the breadwinner will only affect household consumption if it affects permanent income. So, failure by the loss of income to reduce the permanent income means transitory reduction in income will not have effects on household consumption hence added worker effect falls short. However, given that duration of stay in Arda Transau is probably permanent, unless someone voluntarily migrates, and that income has fallen significantly, it is most likely that permanent income of the IDPs is affected hence the need for added worker effect to at least reduce the magnitude of effects of displacement.

Empirical evidence from the Northern provinces of Mozambique by Bruck (2004) reveals that displaced households concentrated more on food crop production at the expense of cash crops as a coping strategy in the post-war era. This strategy was attributed to experience of having practiced that during the war period therefore experience is important in determining an economic activity to engage in after displacement. Those households with more livestock were also found to have increased income from agricultural activities because of the synergistic effects from animal ownership and crop yields. Bruck and Danzer (2007) also found out that having cattle increased agricultural income and noted that having cattle was a coping strategy since cattle can be used to increase income in several ways such as selling and providing manure for crops that can then be sold. In the same study by Bruck and Danzer (2007), land tenure was found to have a strong impact on farming and subsistence activities among IDPs in Mozambique.

Owning the land, one is using is likely to improve sustainable farming strategies because one understands the importance of that land to their current and future livelihoods. Leased land may be used in unsustainable manner because the user may be concerned with short-term benefits thereby paying less attention to the long-term ones. Therefore, land tenure type is likely to have a significant impact on the type of farming as noted by Bruck and Danzer (2007).

Social capital and social support have been found to be other common coping strategies among displacees especially African refugees (Schweitzer et al. 2007; Mukwada 2012 Levron 2013; Northcote 2015; Barbelet 2017; Kiboro 2017). A number of studies have shown that social support as a coping strategy is very useful among Sudanese displacees (Schweitzer et al. 2007; Khawaja et al. 2008 and Bello et al. 2014. While Schweitzer and colleagues concentrated on Sudanese outside Sudan, Bello et al. (2014) considered IDPs in Khatoum, Sudan, specifically women and concluded that social capital is an important and common coping strategy.

Schweitzer et al. (2007) found out that Sudanese refugees in Australia use mostly three strategies, to cope with trauma, which include religious beliefs (a component of social capital) and social support. Again, Khawaja et al. (2008) found that Sudanese IDPs in Brisbane, Australia, relied more on religious beliefs, inner resources and social support, as forms of coping with displacement during the pre-migration, transit and post-migration phases. In Zimbabwe, Mukwada (2012) established that most households in Mufurudzi who were displaced during land reform programme depended on social capital. The same can be said among IDPs in Arda Transau who happen to be connected by their background. By the same token, Abdulai (2016) noted that family networks at destination as a form of social capital are important in migration decisions. Therefore, because of social capital IDPs could not opt for other areas other than Arda Transau where their neighbours have been placed.

Social capital has also been used by migrants in South Africa where forced migrants were found to use ethnicity and community connections to get employment (Northcote; 2015). However, ethnicity also led to few women benefiting from connections because the women were left to do mostly child-minding. Again, Barbelet (2017) found that refugees of Central Africa Republic origin in Cameroon relied more on family and friends for food and clothes and even their spouses did so, a conclusion also reached by Levron (2013) when they indicated that Cote D’Ivoire displacees took refuge in friends’ and relatives’ homes in urban centres. However, Levron (2013) and Barbelet (2017) noted that in as much as social support is helpful it has a limit beyond which it can no longer be used. For instance, in the study by Barbelet (2017), one participant indicated that they could not be accommodated beyond 10 months because of the host members’ capacity. Also, Levron (2013) indicated that sheltering displacees places a burden on hosting households and risk of food insecurity increases thereby limiting its use in longer durations.

In a study of Eritrean female refugees in Norway, Abraham et al. (2018) found that social support, especially among fellow Eritreans, was used as a coping strategy. Jabeen et al. (2010) also found that social networking was the first port of call to get help for the slum residents of Korail in Dhaka, Bangladesh. Through investigating the influence of social capital on livelihood outcomes, Kiboro (2017) found out that the majority of households who belonged to local level associations obtained what they needed which influenced their livelihoods and hence were insulated from risks. In the current study, therefore social capital is considered as a coping strategy that can be heavily relied upon by IDPs. Social capital has the ability to reduce transaction costs especially related to information.

Jabeen et al. (2010) found that the slum residents in Korail used saving as a coping strategy. It was found that 50% of the IDPs saved with NGOs and savings groups. This was viewed as a preventative coping strategy, because of its ex-ante nature, since the residents would use those savings to access loans during times of hardship. In the Philippines, 41.8% of respondents from Leyte area used family savings as a coping strategy against climate change effects (Israel and Briones; 2014). In as much as saving is a good strategy, the savings may be affected by inflation thereby affecting the use of saving as a strategy. However, from the intertemporal consumption theory, saving increases future consumption. If a household currently saves, their future consumption will increase because they will use future income combined with previous period consumption to finance their consumption.

In the Indonesian city of Ambon, Adam (2008) established that informal market activities were common coping strategies after the conflict-caused displacement. In the same research Adam (ibid) argued that these coping strategies must be appropriately termed coping mechanisms, due to their short-term nature and that these mechanisms have caused more impoverishment. However, the displaced individuals, especially women, showed a proactive attitude because of the access they had to the informal sector than their male counterparts. These findings were corroborated by the works of Borghans et al. (2009) who found out that women are more risk-averse than men. The implication is that women are more risk-averse than men, therefore, would prefer to use ex-ante strategies.

Horn (2009) argue that coping strategies used by Northern Ugandan IDP camp residents mainly consist of assistance from others, work and income generation and social support. It was established that the residents had to resort to social support so as to make practical assistance possible when it really was needed, a finding that is in line with that by Helgeson et al. (2013) where the majority of farmers obtained funds from family and friends. Among the work and income generation strategies, cultivation was impossible in the Ugandan case because of movement restrictions (Horn 2009). A strikingly different finding in Northern Uganda was by Helgeson et al. (2013), who, among weather disaster-induced IDPs, found that asset disposal, especially livestock, was mostly used as a coping strategy. Disposing assets for instance land, however, exposes households to vicious circles of extreme poverty.

Bello et al. (2014) also found street vending, buying low priced food items and second-hand clothes as other coping strategies used by female IDPs. Vending was used as an income-generating activity while the other two were cost-cutting measures. These findings seem to corroborate Dercon’s (2000) statement that sometimes coping strategies are used as income-generating activities.

Time also plays a pivotal role in decision-making for purposes of coping. Bozzoli et al. (2011) studied the effects of living in an internally displaced people’s camp in Northern Uganda on activity choices and found that camp duration is a significant determinant of activity choices. This was also found among Central Africa Republic refugees in Cameroon by Barbelet (2017). The time that someone has spent at a place makes them familiar with the area such that even the type of crops to be cultivated will be better known to them. Bozzoli et al. (2011) also found that Northern Uganda displacees living in camps were more into cultivation and trade than their counterparts who returned, that is returnees. Although Bozzoli et al. (2011) and Barbelet (2017) concluded that duration of stay in an area by IDPs plays a significant role in determining choice of economic activities, one may also note that because of social cohesion new and old families (residents) may share ideas such that duration of stay may not really have an effect on choice of coping strategy.

Education has also been cited as a determinant of choice of coping strategy (Rashid et al. 2006). The transmission from education to coping strategy is through income. It is argued that households with high levels of education tend to have stable incomes and have more income sources than the less educated ones. Thus, income diversification reduces the need to use coping mechanisms where other individuals will apply it. Rashid et al. (2006) also indicate that the pattern of coping strategies also depends on the nature of the shock and household characteristics whereby household characteristics include level of education. It was also found that among Syrian refugees in Jordan, high education led to high and satisfaction with income. Therefore, the attained education was used as a coping strategy to get a job and earn a living (Alzoubi et al. 2017). Those with higher education had a high probability of getting employed and fend for the family.

One coping strategy that Rashid et al. (2006) and Israel and Briones (2014), among other authors, identified is adjustment of meal portions that is, reducing household food consumption. In Bangladesh, Rashid et al. (2006) found that 72.4% of households used reducing meal sizes as a coping strategy. The adjusting of meals includes reducing the number of meals eaten such as going the whole day without any meal and or reducing the size of each meal. The adjustment of meals was also found among Sudan female IDPs in Khatoum (Bello et al. 2014) where 56.7% of respondents restricted meals to one per day. This was necessitated by the inadequacy of their income to buy the necessary food items. Additionally, Israel and Briones (2014) found that 47% of non-agricultural households and 76% of landless households in East Laguna Village, Philippines, use reduction of food consumption as a coping mechanism while 79% of all flood-affected households cope by eating less preferred food.

Another possible coping strategy employed by IDPs is livelihood diversification with those IDPs mostly familiar with agricultural activities using crop diversification. Livelihood diversification is the process by which households construct a diverse portfolio of activities and social support capabilities for survival and in order to improve their standard of living (Ellis 1998). Ellis (1998) accepts that livelihood is more than just income and notes that “A livelihood encompasses income, both cash and in-kind, as well as the social institutions (kin, family, compound, village and so on), gender relations, and property rights required to support and to sustain a given standard of living” (Ellis 1998, 4).

Livelihood diversification strategy can be used by both the poor and the rich albeit from different perspectives. The rich use diversification from the perspective of accumulating wealth which is termed opportunity-led diversification. From the poor people’s side, it is viewed as a strategy for survival hence the term desperation-led or distress-push diversification (Motsholapheko et al., 2012). Motsholapheko et al. (2012) therefore assert that the economic status of each household determines the diversification approach that the household uses. Since most households were found to earn as little as less than US$100 per month (Mandishekwa and Mutenheri 2018) and also because of the nature of this study, livelihood diversification in this study is mostly viewed as desperation-led diversification.

Diversification, as a strategy to cope with stressful life events, comes in various forms. For instance, in the Ogbozor (2016) study, households diversified their activities by engaging in vertical diversification through fishing, herding and farming, among others, as a coping strategy against displacement by Boko Haram insurgency. Horizontal diversification through crop diversification has also been used to cope with climate change effects (Mulinya 2017). Mulinya (2017) found that farmers were growing more than one crop, such as drought-tolerant crops at a time, agroforestry and even crop rotation all in the name of horizontal diversification strategy. In Botswana, households planted an average of five different crops on a subsistence basis (Motsholapheko et al. 2012). Another form of horizontal diversification is plot diversification as used by IDPs in Northern Mozambique (Bruck and Danzer (2007). In the plot diversification approach, a household will spread their risk by having fields that are distant apart. This has the effect of reducing crop failure through various ways. One such way is by virtue of soil type differences hence moisture requirements. Another way is a way of minimizing effects of pests. Some pests are prevalent in any area while they may not be in another area. Common ways of pursuing plot diversification strategy are through renting plots and or buying fields in other areas.

In Okavango Delta of Botswana, households used livelihood diversification when faced with extreme floods (Motsholapheko et al. 2012). The diversified activities include livestock farming, dryland arable farming and social welfare, among others. In addition, it was found that the main livelihood activity was livestock farming undertaken by 73% and that cattle acted as savings for these households. It was also found that the choice of diversification strategies was determined by socio-economic status with high income earners adapting to flood effects more than poor ones.

Coping and livelihood strategies also differ significantly according to gender. With the objective of finding differential effects of gender on livelihood diversification in Uganda, Dolan (2004) found that female-headed households had distinct constrains from male-headed ones. Chief among these constrains were cultural norms and access to productive resources. Renner and Salem (2009) established that while women, among asylum seekers and refugees, concentrated on taking care of children and indoor activities, men looked for employment to feed the family and also socialized more. Additionally, Seguin (2016) found that gender differences exist among coping strategies used with regard to problem-solving and support seeking, among others. This strand of literature may be viewed as an indicator that gender really plays a significant role in determining how one copes with stressful situations, thereby showing that some strategies are masculine in nature while others may be viewed as feminine. However, it is undeniable that coping can be cross-cutting across gender where males may employ those strategies that may be associated with women.

It can, therefore, be concluded that choice of coping strategies among IDPs is determined by various factors such as gender with women preferring ex-ante risk management strategies while men prefer ex-post strategies. It also appears that women are more risk-averse than men. Various coping strategies have also been outlined in literature such as social networks, religion, consumption smoothing, income smoothing and duration of stay in an area. It has also been established above that little is known about economic activities among IDPs with small probabilities of being displaced again like in Arda Transau.

Methods

This section takes the reader through the various methods used in gathering data for the research. Firstly, an overview of the area is given followed by data gathering techniques and methods of analysis used.

The study area

Arda Transau is a former government farm where households displaced from Chiadzwa diamond fields were relocated. The farm, which is about 40 km from the City of Mutare, is found in Mutare rural district in the Eastern Border of Zimbabwe. The 12,000 ha (Nyawo et al. 2012) farm is now divided into five major areas now named according to the company responsible for displacing households from Chiadzwa. The largest area, in terms of area and population size, comprising Arda Transau now is Anjin followed by Jin An. Of all the five areas making Arda Transau DMC is the smallest.

Displacement of households started in 2009 when the Zimbabwean government ordered the households to be forcibly relocated to pave way for mining. At the time data was gathered, a total of 930 households had been allocated houses implying that the minimum number of displaced households was 930. This figure marked the population size for sample size determination in this study. When households were displaced, they were promised to be catered for by the companies concerned. Such assistance was to be in the form of compensation for lost assets, employment for the young men, food hampers, farming inputs and a portion in the irrigation scheme. During the first days, food hampers were given but they suddenly disappeared, signalling the beginning of other impending problems. The irrigation scheme required a joining fee which most households could not afford implying that they will be out of the scheme. The promised free piped water now required payment and this was argued to be payment for electricity to pump the water. Those households who could not afford to pay for the water had their water disconnected. On the other hand, some areas such as Mbada Diamonds and DMC were given community boreholes unlike others with reticulated water supply.

Despite losses experienced due to displacement (landlessness, marginalization, food insecurity, among others), compensation was never given to the households except for only 1 ha of arable land on which the houses were built. The only assistance they got was a disturbance allowance amounting to US$1000 per family. Unlike in Chiadzwa where relief agencies were allowed to assist households, according to the IDPs, no relief agency is allowed in Arda Transau.

Another problem with the new settlement is overcrowdedness. Households were given four-roomed houses irrespective of the family size. One other thing that was not considered is membership of the households. For instance, one family man with four married sons was given four rooms despite the fact that each son had a standalone family in Chiadzwa.

Data collection

Data was gathered using simple random sampling technique during the period 2016 to 2017 with the understanding that the study is a case-crossover design as propounded by Maclure (1991). The control period is the period prior to displacement while the exposure period is the period after displacement. The design is appropriate since it reduces control selection bias because of self-matching of participants in the sense that each participant serves as their own control thereby adjusting for confounding.

A sample of 274 participants was selected determined using Krejcie and Morgan (1970) methods from a population of 930 households. For agricultural diversification, (a form of horizontal diversification), three options were given to respondents to choose and these choices were guided by the various economic activities undertaken by households in Zimbabwe and Africa at large, for instance, food crop cultivation being dominant in Zimbabwe (Bird et al. 2002), cash crop cultivation and livestock rearing. Agricultural diversification, however, is sometimes defined as “...the increasing allocation of household resources to the production of non-staples relative to food staples...” (Dzanku and Sarpong 2011, 189). In this study, however, it is defined as the number of agricultural activities undertaken as perceived from the context of horizontal diversification. Thus, in line with this, Dzanku and Sarpong (2011) alternatively define agricultural diversification as implying “...increased cultivation or the adoption of cash crops” (Dzanku and Sarpong 2011, 192). Further to that, these two authors iterate that food self-sufficiency is a strategy towards food security where food self-sufficiency may be interpreted as entitlement in line with Sen (1981).

Two key questions were asked relating to major economic activities undertaken by households prior to and after displacement. This set of questions guided this study as to whether that activity can be considered as a coping strategy or not. This is in line with the observation made by Bozzoli et al. (2011) that the data on activity choices must reflect what was done prior to displacement. For agricultural activities, it was noted that most IDPs may practice more than one strategy such that there was a need for study participants to choose as many options (from the given list) as they use. This was motivated by the significance of agriculture in Zimbabwe’s economic activities. To this end, the question allowed farmers to indicate even more than one activity used, therefore, allowing this study to determine if ever there is livelihood diversification from agricultural activities perspective borrowing from horizontal diversification concept. Again, from horizontal diversification perspectives, households were asked a question relating to plot diversification. Added to this was the analysis of other economic activities undertaken by IDPs other than agriculture related ones. Frequency tables were used to determine the approach mostly used and the level of diversification.

Choice of a coping strategy depends on specific household characteristics (Rashid et al. 2006). Therefore, following the works of Bruck and Danzer (2007) and Deressa et al. (2010), this study used a multinomial logit (MNL) model to analyse the determinants of the choice of main economic activities among households. As far back as 1984, Hausman and McFadden (1984) indicated that the multinomial logit model is the most often used among discrete choice models. One of the most restrictive assumptions, though not always plausible, of the MNL is the Independence of Irrelevant Alternatives (IIA) assumption. This assumption states that increasing the number of alternatives available to the choice-maker will not change their odds of making their choice (Jones; 2005). Thus, using the example of consumers, Baltas and Doyle (2001) states that “...the odds of the consumer choosing j over k remains the same regardless of the composition of the choice set” (Baltas and Doyle; 2001,116). While it is an important assumption, Dow and Endersby (2004) argue that this IIA “... is a logical property of decision-making not a statistical property such as consistency” (Dow and Endersby 2004, 112). Basically, the two are saying the importance of IIA must not be exaggerated. To verify the independence of irrelevant alternatives this study carried out the Hausman type test, the same way that Bruck and Danzer (2007) did in Mozambique. However, a suest-based Hausman type was used in this study. The preference for the suest-based Hausman type over the ordinary Hausman test is because the Hausman IIA test depends mostly on choice of base category.

The multinomial logit (MNL) model applied in this study is derived in the context of choice models based on economic theories of utility maximization. In a case where a displaced individual household has to make a choice among various coping strategies or economic activities such as employment and petty trading, among others, one can assume the individual household i’s utility for economic activity j, is a function of both the attributes of the individual household and the activity itself (Maddala 1993) and an error term which is stochastic. In that case, then the conditional logit model as propounded by McFadden is applicable (Maddala 1993). However, in this study, the focus is on household characteristics as determinants of choice, therefore, giving rise to multinomial logit model. The multinomial logit model used in this study is specified as follows:

$$ {U}_{ij}={\beta}^{\ast }{X}_{ij}+{\alpha}_j^{\ast }{Z}_i+{\in}_{ij} $$
(1)

Where Uij is household i’s utility derived from economic activity j, Xij is a vector of attributes of economic activity j as perceived by household i. Zi is a vector of the individual household characteristics, usually proxied by household head’s characteristics. βand αare the respective coefficients to be estimated.

The Additive Random Utility Model variant of eq. 1 shows that from an m-choice model, the utility of the jth choice (Uj) is specified as follows (Cameron and Trivedi 2005, 504):

$$ {U}_j={V}_j+{\in}_j,\kern0.5em j=1,2,3,....,m $$
(2)

“Where Vj and ∈j are deterministic and random components of utility” (Baltas and Doyle 2001, 116), respectively. The individual household will thus choose an alternative that yields the highest utility since utility maximization is assumed within the context of household choice models.

Now based on the argument that the choice of coping strategy depends on the individual household characteristics, the study employs the multinomial logit model, which is alternatively termed “the characteristics of the chooser model” (Jones 2005, 34). The reason for ignoring the effects of attributes of the economic activity on choice was motivated by the fact that these activity attributes do not usually vary across individual households (Davidson and MacKinnon 1999) and that the choice of a coping strategy by a household depends on specific household characteristics (Rashid et al. 2006). Thus, the only determinants of choice are the characteristics of the individual household which, however, do not vary across alternatives but enter the equation only in such a way as to create differences in utility over the available choices. Eq. 2 can, therefore, be restated in such a way as to capture the characteristics of the decision-maker and thus is specified as follows:

$$ {s}_i=\alpha +\beta {\chi}_i $$
(3)

The left-hand side (LHS) of Eq. 3 is a polychotomous nominal variable showing whether a certain coping strategy was employed by household i. On the other hand, the RHS shows the household demographic characteristics such as family size, household head’s age and education and gender of household head, among other characteristics, as captured by Xi. Of note is that the response variable is nominal in nature, that is, unordered. MNL models require that one of the categories has to be considered the base category against which all others are compared (Chatterjee and Hadi 2006; Kleinbaum and Klein 2010). Again, one has to evaluate the probability of one alternative against another where the two alternatives are not the same, thus finding out how a change in one characteristic affects the response probability (Perez-Tuglia 2009).

The data gathered for the main economic activity engaged in by a household was screened into five main groups. Each activity could fall into one of these main activities viz., employment (self or otherwise), petty trading (e.g firewood trading, among others), farming, asset disposal and any other activities besides the four identified. Employment was coded as one (1), petty trading as two (2), farming as three (3), asset disposal as four (4) and others as five (5). Thus, this creates an unordered polychotomous variable with 5 levels.

For analysis of determinants of diversification in agricultural livelihood activities, the ordered logit model was used since the dependent variable had a natural ordering. Agricultural livelihood diversification in this study is defined as engaging in more than one agricultural activity. Agricultural activities were classified into three namely, food crop production, cash crop production and animal rearing. The number of agricultural activities undertaken was recorded implying that an ordered categorical variable was created. The more the number of activities undertaken, the more diversified a household is. The specification of an ordered logit, however, is similar to the one for MNL except that the dependent variable is no longer nominal but ordered thereby implying that eq. 3 is still useful with that small modification.

Variables

Chambers and Conway (1991) argued that “Many livelihoods are largely predetermined by accident of birth” (Chambers and Conway 1991, 6). This statement implies that some livelihood choices may be determined from the day that a person was born. Implied again in the statement by Chambers and Conway (ibid) is the fact that males and females employ different livelihood activities as determined by their gender. Based on literature reviewed, several characteristics of households have been identified as being possible determinants of choice of a particular activity. In this section of the study, these variables are outlined and justified.

Household head’s age

As age of household increases, the motive to adapt to change reduces. This may be aligned to the old adage that ‘You cannot teach an old dog new tricks’. Thus, from the IDPs perspective, households led by elderly IDPs may try by every means possible to apply the livelihood strategies used prior to displacement, for instance, agricultural practices, used in Chiadzwa while in Arda Transau. To this effect, Mulinya (2017) found a negative correlation between age of farmer and adaptation strategies to climate change in Kenya. In this regard, this study assumes that as age of household increases, the probability of engaging in new economic activities reduces. It is also assumed that as household head’s age increases, the household is less likely to rent a farm (alternatively termed plot diversification in this study) as well as working in other people’s fields but is more likely to dispose off livestock as a coping strategy. This is reasonable because the first two of these activities require more labour and may, therefore, be mostly engaged in by active members of the family. However, disposal of assets such as animals may increase as age of the respondent increase. Along this line of reasoning, Helgeson et al. (2013) found that households headed by older members were more likely to sell livestock as a coping strategy because households headed by elderly members are considered more risky, for instance, to access credit as a coping method. The three claimed that older members of the households are more risky borrowers than young ones hence could not access lines of credit thus they resort to livestock sales. The probability for this level of risk may be explained by the fact that elderly households will be on the declining part of their income curve, ceteris paribus. However, among rural women in Oyo state of Nigeria, Adeniyi et al. (2016) found that age was an insignificant determinant of livelihood diversification just like Spence et al. (2007) in the USA among IDPs displaced by Hurricane Katrina. Spence et al. (2007) also highlighted the possibility of differences between coping strategies attributed to age differences when these three authors state that older people may feel they (older people) have no control over events hence may take passive approaches compared to young ones who take active approaches to coping with stressful life events.

Gender of household head

Most decisions regarding what to grow and at what time in Africa are relegated mostly to men (Mulinya 2017) possibly because of tradition where men are generally considered to be the head of the family. In this assertion, it is implied that agricultural coping decisions are a preserve of the males. Another perspective is from the risk side where women are considered to be more risk-averse than men (Borghans et al. 2009; Nelson 2012) such that women are more likely to engage in familiar coping strategies than their male counterparts usually because risk-taking is considered a preserve of men. Thus female-headed households in the study area are more likely to engage in activities that they used to do in Chiadzwa than engage in new ones. With such a presumption women are most likely to engage in agriculture and petty trading activities since the women IDPs in Arda Transau are familiar with those activities from experience acquired in Chiadzwa. Again, empirical findings by Dolan (2004), Spence et al. (2007) and Renner and Salem (2009) indicate that livelihood activities differ between male and female-headed households. From all these arguments, it is plausible to suggest gender as a determinant of choice of coping strategy among IDPs in Arda Transau.

Marital status

Marital status has been noted to affect individuals’ risk aversion with married individuals being more risk-averse than the non-married ones Roussanov and Savor (2014). Roussanov and Savor (2014) noted that usually unmarried individuals are more ambitious hence may tend to be more risk-tolerant. Among women, being married has been found to be statistically negatively related to household livelihood diversification (Adeniyi et al. 2016). Therefore, married individuals have lower probability of having better livelihoods because they take less risk since the adage ‘no pain no gain’ may hold. One such argument may emanate from the fact that married individuals may engage in what they consider to be safer (from experience) activities than those done by their non-married counterparts. On the other hand, married individuals may be assumed to be more diversified since the couples share ideas as to what to engage in. Also, Martinoty (2015) iterates that marriage is a form of risk diversification, so married individuals are likely to be more diversified than non-married household heads. Therefore, in Arda Transau, one would expect married household heads to be more diversified than their non-married counterparts.

Household size

According to Mulinya (2017), family size plays a pivotal role in determining strategies to use to adapt. The author claim that large family sizes may strain family resources thereby hindering adaptation to change. To this end, the study by Mulinya (2017) found that family size is negatively correlated to adaptation to climate change. Along this line, the current study assumes that large households are likely to venture into familiar territories, that is, use activities that they were using in Chiadzwa. With regard to livestock sales, Helgeson et al. (2013) found that large households are more likely to sell livestock to cope with disasters. They further allege that this strategy increases vulnerability of households. Awotide et al (2010) in Adeniyi et al. (2016) argued that an increase in household size increases the probability of farmers being poor.

Religion

In this study, religion is used as a proxy for social capital. It is assumed in this study that religion plays a role in disseminating information on household livelihood activities. Thus, being a religious household head is likely to increase the probability of choosing a particular economic activity versus others because one will have ideas gathered through social capital. Halcon et al (2009) indicated that women employed social capital as a coping strategy through discussing their problems with friends. In this regard, it is argued that when it comes to diversification strategies more religious individuals may be highly diversified than their counterparts. In short, religion in this study is a proxy for social capital.

A plethora of literature reveals that many livelihoods depend on social capital. For instance, Mukwada (2012) found that in Mufurudzi village, Zimbabwe, most households depended on social capital when they were displaced during land reform. Authors like Khawaja et al. (2008), Amisi (2006), Northcote (2015), Barbelet (2017) and Kiboro (2017) also established that social capital is a determinant of coping with displacement.

Education

Rashid et al. (2006) have indicated that education determines coping strategies used. Households that are educated tend to diversify their incomes better hence are less likely to need coping strategies when faced with shocks because income diversification on its own is an ex-ante coping strategy. In Cameroon, Central Africa Republic (CAR) refugees had to change their economic activities because they could not be employed and teachers were not allowed to teach at the same level because their diplomas were not accepted (Barbelet 2017). Thus, teachers were denied teaching jobs because in Cameroon teachers are supposed to be highly educated as compared to that in CAR. The same happened to IDPs in Sri Lanka whose skills were not used because of lack of demand for these skills in the host community (Amirthalingam and Lakshman (2009). However, among Syrian refugees, Alzoubi et al. (2017) found that education on its own was a coping strategy because it enabled IDPs get jobs. On education as a determinant of livelihood, Adeniyi et al. (2016) established that education is a significant and positive determinant of livelihoods. Also, farmers who are better educated may have a better understanding of livelihoods.

Duration of stay in Arda Transau

The choice of economic activity may depend on time spent in the post-displacement area. For instance, among Central African Repulic refugees in Cameroon, new refugees mostly relied on manual jobs while those who had been there for long sought to diversify activities (Barbelet 2017). Time is also associated with learning. In the event that one strategy fails, the displaced will try another strategy until one strategy prevails. Thus duration of stay plays a critical role thus the study expects that those IDPs who have been displaced earlier are likely to be more diversified. Also, Bozzoli et al. (2011) showed that coping strategies differ according to period that the IDP has stayed in a given place.

Economic activity prior to displacement

In this study, this variable was used to represent experience with an economic activity. Along the lines of thinking of Amisi (2006), previous experience may determine current economic activity-. In fact, those who had been in farming may be inclined to get into farming even after displacement while those who are experienced in petty- trading will likely to be doing so. Individuals usually have a fear of the unknown hence would want to play in familiar grounds, even though the situation or conditions may have changed. Thus, economic activity before displacement is likely to have an influence on the current activities undertaken by households.

Results

Descriptive statistics were used in the analysis. The main statistic employed by this study for analysis of descriptive statistics is the mode. Two main methodologies were employed viz. multinomial logit for determinants of main economic activity and ordered logit for determinants of diversified agricultural activities. Odds ratios are reported and are the method of analysis used in this study.

Summary statistics

To deduce how diversified livelihoods are in agricultural activities terms, the study employed frequency tables. The number of agricultural activities undertaken was recorded. Three major groups of agricultural activities were used viz, cash crop farming, livestock production and food crop production. Findings to this effect show that the majority (60%) of households are not diversified with 56% of households undertaking only one activity while 4% are not even involved in agriculture. Of the diversified households, 94% engage in only two agricultural activities while the remainder engage in all three. The next most common (38%) number of activities was two where households engaged in two agricultural activities. With 4 % of displacees indicating that they did not engage in any agricultural activities after displacement, only 2 % of the households indicated that they were engaged in all three. Table 1 shows the number of agricultural activities undertaken grouped by gender.

Table 1 Agricultural Diversification by Gender

Table 1 shows that very few households (2%) are highly diversified in their agricultural activities. The most common group (56%) is that of households who engage in only one agricultural activity of which slightly more than half (51%) are males.

Whilst the number of male-headed and female-headed households engaged in only one agricultural activity is almost the same, the proportion differs with a higher proportion of females (69%) compared to males (47%). 47% of males engage in two agricultural activities, a proportion that is equal to that of those engaging in one agricultural activity only. For those who diversify by engaging in only two activities, 74% are male-headed while 26% are female-headed households. Table 1 also shows that the majority (60%) of those displacees who are not into agriculture are males while almost an equal proportion among males and females engages in only one agricultural activity, (51% and 49% respectively). For the females only sample, 24.5% diversify by engaging in two activities, a proportion slightly more than half that of males (47%) who engage in the same number. It is, therefore, most likely prudent to conclude that diversification is associated with being a male-headed household.

To determine the main economic activities undertaken by displacees, frequency tables were also employed as reported in Table 2. Table 2 also shows economic activities done by households grouped by gender of household head. This was guided by the idea proposed by Jacobsen (2002) that “Displacement can result in new forms of gender and age vulnerability” (Jacobsen 2002, 98). Therefore, the different vulnerabilities may lead to different economic activities employed as coping strategies. Findings indicate that slightly more than half (53%) of the displacees are engaged in agriculture as a livelihood activity while 8% depended mostly on employment as their economic activity. This finding confirms the assertion that mostly IDPs engage in agriculture as a livelihood activity (Niger 2018) and that employment is less relied on after displacement (Adeniyi et al. 2016). Findings to the effect that farming dominates among rural households corroborate those found by Adeniyi et al. (2016) in Nigeria and that by Beyene (2012) in Ethiopia. Slightly more than half of the male-headed and that of female-headed households engage in agriculture as a coping strategy (54% and 51%, respectively). Of the 53% of households who depend on agriculture as a coping strategy, 62% are male-headed while slightly more than half of that percentage (38%) are female-headed households. The implication is that being male is more likely to be associated with engaging in agriculture as an economic activity than being female. This is in line with the findings by Amisi (2006) that there is usually a gender discrepancy between males and females when it comes to economic activities after displacement.

Table 2 Major Economic Activities by Gender

Of those who consider employment as their main economic activity, the majority (68%) are males. The least used coping strategy was found to be asset disposal with 4% of displacees using it as the main economic activity of which the majority (78%) are male-headed households. This finding may be as a result of IDPs knowing the long-term consequences of asset disposal that is it may exacerbate their impoverishment and retard their rehabilitation. Any other activities, besides the aforementioned, used as coping strategies comprise 17.33%, a figure slightly less than that for petty trading which was reported to be being used by 19% of IDP households. Petty-trading was found to be the next most frequently used strategy among IDPs in Arda Transau and findings are in line with those by Adeniyi et al. (2016) that petty-trading follows behind farming. While petty-trading as a coping strategy was used by 19% of respondents, 50% of the 19% were males, thereby seemingly indicating no difference between males and females in use of petty-trading. The result also corroborates those by Magaramombe (2010) among fast track land reform displacees in Mazowe and also Omata (2012) found the same.

However, for the males only sample, only 15% engage in petty-trading as a main livelihood activity unlike 23.5% for females, thus again corroborating the assertion that males and females employ different strategies. Again that petty-trading is prevalent is not unique to this study since Mukwada (2012) also found that in a study among resettled farmers in Zimbabwe and Levron (2013) found that among women IDPs in Cote d’ivore. That petty-trading is common among displacees was also argued for by Omata (2012) where it was indicated that displacees, refugees, in particular, engaged in a variety of activities including small business activities, such as vending fruits, to sustain themselves. Also, Adeniyi et al (ibid) found that formal employment is less used by rural households as a livelihood activity, a finding similar to that of this study.

Pre- and post-displacement main economic activities compared

Barbelet (2017) argued that after displacement, skills that IDPs have gained may become obsolete or not relevant. To that effect, this study sought to determine whether households still engage in what they did before displacement. Table 3 reports the results of this study on that particular aspect.

Table 3 Comparison between Pre-and Post-Displacement Economic Activities

Table 3 shows that while 111 (45%) households relied more on petty trading prior to displacement only 46 (19%) do so after displacement. This is justified because the main items they relied on in Chiadzwa are not available in Arda Transau such as selling Nyii, mauyu and even gardening. While agriculture is the mainstay of Zimbabwe, only 26% of families relied on it Chiadzwa while the figure slightly more than doubled (53%) in Arda Transau. In line with this finding again Niger (2018) found that most IDPs in Niger rely on agriculture as an economic activity. The group of people who relied on other activities was the lowest before displacement being a paltry 6% while it almost tripled to 17% after displacement. Thus, after displacement, more households found themselves relying more on other things other than employment, petty-trading, farming and asset disposal. This is similar to a situation where Central African Republic refugees found conditions in Cameroon different from those in their home country hence their experience was no longer useful. Congolese refugees in Durban were also found in that dilemma where they were then forced to review expectations and had to do what is there for them to survive (Amisi 2006).

After combining asset disposal and others with agricultural activity because the first two had small frequencies (also as a way of complying with results of the tests performed later in the study, Table 5), a Stuart-Maxwell test was used to test marginal homogeneity between economic activities done before and after displacement. Findings show a statistically significant difference,χ2(2, N = 248) = 52.73, p ≺ 0.001), implying that these activities may be considered as coping strategies in line with the Niger (2018) assertion that once a person does new things in the post-displacement era, it can safely be considered a coping strategy.

Besides the main economic activity done by each household, several other coping strategies are used. These involve exchanging goods for goods (barter trade), engaging in working in other people’s farms (selling labour) especially during the farming season and even livestock disposal on its own. Some of the most commonly cited coping strategies used by IDPs are summarized in Table 4.

Table 4 Animal disposal and barter trade

In Arda Transau, animal disposal as a coping strategy has become common with 58% of respondents indicating that animal disposal has increased after displacement. This is despite the fact that most IDPs have lost their animals due to displacement (Mandishekwa and Mutenheri 2018). That sale of livestock among IDPs has increased is in line with the sentiments made by Justino (2012) that animal sales is mostly used by rural households in times of crisis. Also shown in Tables 4, 49.60% of respondents have increased their degree of barter trade after displacement. From the livelihood perspective, the increase in livestock sales represents the cash earnings component of income (Ellis 1998). Therefore, earnings from animal sales have significantly increased between the pre-and post-displacement era.

McNemar’s test was performed to find out if really there is a significant difference between those who used livestock sale as a coping strategy before and after displacement. It was found that surely livestock sales increased after displacement implying that this activity can safely be considered as a coping strategy employed after displacement, χ2(1) = 7.12, p = 0.0076. This means that animal sales as a coping strategy have increased significantly after displacement. However, this approach of survival leads to higher levels of poverty in the future since it leaves households with few assets that they can safely rely on. This confirms that self-insurance is an unsustainable approach (Dercon 2000).

Again, McNemar’s test was used to determine if barter trade among IDPs increased after displacement. Results indicate that there is no statistically significant difference in terms of use of barter as a coping strategy between the pre- and post-displacement periods, χ2(1, N = 248) = 2.56, p = 0.1098. Thus despite, about 50% of respondents indicating that using barter as a coping strategy has increased, this was found not to be statistically different from its use in the previous period.

Statistical analysis of determinants of coping strategies among IDPs in Arda Transau

One of the study’s objectives was to explore the determinants of choice of coping strategies and livelihood diversification strategies among internal displacees in Arda Transau, Zimbabwe. In a bid to attain this broad objective, quantitative analysis of data was used. This study reports odds ratios and p values and variables are considered significant using the 1%, 5% and 10% levels of significance. To find the determinants of choice of main economic activity undertaken by each IDP household, a multinomial logit (MNL) model was applied. Four main economic activities done by IDPs were found guided by literature on livelihood activities among IDPs and pilot survey. These activities include survival on employment wages, petty trading, farming and livestock disposal. However, other IDPs do not rely mostly on these hence a fifth group was created representing any other activities. Naturally, this kind of data creates a nominal variable indicating economic activity undertaken by IDPs. Therefore, the MNL logit model was used to find the socio-economic determinants of the choice of economic activity after displacement.

To find out if the categories of the dependent variable could be combined the likelihood ratio (LR) test was carried out and findings indicate that three categories can be combined into one and these categories are farming, asset disposal and others since the p values were statistically insignificantly different from zero. The reduction in categories of the dependent variable makes interpretation of results from a MNL model easier. To test for independence of irrelevant alternatives (IIA), a suest-based Hausman test was performed and findings indicate that the IIA was not violated since the obtained chi-square values were all not significant.

From Table 5, one can note that economic activity before displacement, gender of household head and marital status are statistically significant variables. It is reported in Table 5 that economic activities done by a household before displacement reduces the likelihood of a household to choose employment as a livelihood activity, thus its relative odds are 0.1664, with a p value less than 0.001. Again, being female increases the probability of engaging in petty trading. Female-headed households are 2.45 times more likely to be involved in petty-trading relative to farming than the male-headed ones, thereby implying a gender disparity in use of coping strategies between the two groups as noted by Seguin (2016).

Table 5 Determinants of Economic activities among IDPs

Similar interpretations may be made for marital status where married individuals are 6.8272 times more likely to engage in petty trading relative to farming than other non-married household heads, where petty-trading is an activity that was learnt from Chiadzwa (Nyawo et al. 2012).

To evaluate if female-headed and male-headed households’ choice of livelihood strategies are determined by different variables, the study used two equations, one for males only and the other for females only (Results not reported in table format here). Given economic activities done before displacement, women were found to be 0.06 times less likely to rely on employment as a survival strategy than being in farming(p ≺ 0.05). However, despite having only one variable found to be statistically significant, all the variables in the women-only model were found to be jointly statistically significant, χ2(28, N = 98) = 43.94, p ≺ 0.05). Equally, male-headed households are 0.2377 times less likely to use employment as a major livelihood activity than farming, (p ≺ 0.1). All variables used in the males only model were however found to be jointly marginally statistically significant at 10%, (χ2(28, N = 150) = 38.39, p ≺ 0.1).

Other strategies used by households as coping mechanisms include animal disposal. A binary logistic regression was used to find its determinants. No variable among those employed could significantly determine animal disposal. This finding seems sound given that families were required to dispose their animals before displacement. In that respect then, only a few households have animals, however, reduced in number because of the aforesaid requirement and also space restrictions. Mostly animals that were supposed to be disposed before relocation are cattle and goats such that the animal sale increase reported in Table 4 is mostly likely to be that for small livestocks like poultry.

Two other logistic regressions were run, one for determinants of dependency on company support and the other for child support (remittances from children). It was found that being in DMC, other than being in other locations, is likely to increase the probability of one once relying on company for survival, odds ratio being 0.5323 and p ≺ 0.01. This finding is logical since the companies were nominally coded with DMC being one (1) and Anjin being coded five (5) making DMC the base category. Also, most respondents in DMC indicated that they got assistance in terms of food hampers in the first 3 months after displacement. However, at the time of study, no such food assistance was still being given to displacees.

Gender was also found to be a significant variable at 5% level with female-headed households being 2.67 times more likely to have relied on company support than male-headed ones indicating gender differences as also found by Seguin (2016). On the child support (remittances) variable, only age was found to be a significant variable, with households headed by older members being 2.144 times more likely to get support from children than young ones. This seems logical since children may feel more obliged to support households headed by their old aged parents than those still in the active age groups. The fact that IDPs may depend on support from their social networks has also been found by Mukwada (2012) among the most vulnerable land reform beneficiaries in Mufurudzi area of Shamva, Zimbabwe.

To find the determinants of agricultural diversification, the data gathered is ordinal in nature in that the number of agricultural activities undertaken by households determined how diversified the household is. Agricultural activities in this study were grouped into three categories, that is, food crop cultivation, cash crop cultivation and livestock production. In this instance, households engaging in all three were considered highly diversified in their agricultural activities. With such data, an ordered choice model is most useful. Table 6 shows the findings of the study on determinants of agricultural livelihood diversification.

The estimated model shows that all variables employed are jointly statistically significant as determinants of the degree of agricultural diversification because the LR chi-square statistic is 36.86, with a p value of 0.0001. Table 6 shows that four variables are significant as determinants of agricultural diversification. Three of the four variables are statistically significant at 5% (age, religion and major economic activity before displacement) while one is so at 10% (gender of household head).

Table 6 Determinants of Agricultural Diversification

Households with old aged heads are 1.325 times more likely to be diversified agriculturally than those headed by young adults (Table 6). This seems valid since in rural Zimbabwe most elderly-headed households usually engage in production of many crops though it may be on a small scale. Such activities include intercropping maize and beans, among others. This enhances food security of the family and elderly household heads are usually aware of such things from indigenous knowledge.

Religion as a determinant of diversification also plays a significant role. Findings indicate that being a religious household head increases the probability that the household engages in more than one agricultural activity, odds ratio being 4.404 with p value being 0.001. Religion as a social capital variable enhances information sharing. Thus, Christianity, being the main religion practiced by many Zimbabweans, IDPs in Arda Transau included where 90% of them are Christians, helps to communicate information about livelihoods. Also, when people gather for religious purposes, it is perhaps natural that they may engage in discussions about livelihoods and agricultural practices especially during the agricultural season. This will help disseminate information on drought mitigatory measures such as diversification.

The major economic activity done by a household before and after displacement was recorded as nominally and these activities include employment (coded 1), petty trading (coded as 2), farming (coded as 3), asset disposal (coded as 4) and other activities being coded as 5, the variable thus was regarded in this study as a nominal variable. Findings indicate that engaging in any other activity as a means of survival means that a household is 0.7113 times less likely to engage in agricultural diversification and the variable was found to be significant at 5%, (p = 0.008).

At 10% level of significance, gender was found to be a statistically significant variable in determining whether a household is agriculturally diversified or not. Females are 0.5765 times less likely to engage in diversified agricultural activities than males (Table 6). Since diversifying agricultural activities when one is still new in the area is more experimental than experience led, these findings are congruent with the assertion that women are more risk-averse than man (Borghans, et al. 2009, Nelson 2012).

To evaluate the assertion that women are less likely to diversify than males and whether determinants of such activities differ according to gender, the study modelled diversification models for males and females separately. From Table 7, one can note that two variables per each group are significant determinants of diversification. However, the determinants differ between male and female-headed households, thus supporting the assertion that gender of the household head determines the level of diversification. While older males are 1.3826 times likely to engage in diversification, older females are 1.1559 times likely to do so though insignificant for females and significant for males. While diversification reduces risk of food insecurity, females fear that they may engage in unfamiliar territories thus losing out in the process. Thus, females can conclusively be considered highly risk-averse than males in this study just like Borghans et al. (2009) and Nelson (2012) highlighted.

Table 7 Determinants of agricultural diversification by Gender

Males are 3.6138 times more likely to engage in diversification with their choice being influenced significantly by religion,p ≺ 0.05. Unlike for male-headed households, this variable is insignificant for female-headed ones, although the odds are almost double those for male IDP households. These results most likely imply that male household heads use religion as social capital from which they gather information on better ways to diversify so as to mitigate food insecurity. Female-headed households who have been displaced earlier are 1.53 times more likely to engage in diversification than their counterparts who have recently been displaced. Thus, this result seems to be in line with the assertion that most women are risk-averse (Nelson 2012), they do not just venture into activities without prior knowledge of its pros and cons. Those displaced earlier are likely to possess superior information as far as agricultural needs of the soil and farming activities suitable in the area, hence supporting findings by Bozzoli et al. (2011) that duration of stay in the post-displacement area plays a role in determination of choice of economic activities.

Unlike the first three variables, family size is less likely to influence female-headed households to diversify. Thus, large female-headed families are 0.8574 times less likely to diversify than smaller ones, p ≺ 0.1. In as much as diversification may reduce risk, larger female-headed households tend to venture mostly into familiar territories such as just planting one crop such as maize. Summary statistics reveal that for any household whose membership exceeds eight, the number of agricultural activities undertaken is likely to be one. This confirms the argument that large family sizes are less diversified than small ones, as reported in Table 6.

Discussions

The study reveals that most (60%) of the households are not diversified agriculturally, thus they engage in only one agricultural activity implying absence of horizontal diversification. Only 40% of households engage in at least two agricultural activities per period. While age was found to significantly increase the probability that a household will practice diversified agriculture, female-headed households were found to be less likely to practice diversified agriculture. Thus, those households who are headed by older male heads are likely to be diversified than female-headed ones. However, female-headed households were found to be more likely to practice diversified agriculture after a longer time in Arda Transau than man (Table 7). Thus, women are likely to take a longer time before engaging in more activities than just food crop production compared to men. What women IDPs may not understand are the benefits derived from a diversified portfolio of agricultural activities.

The results showing that most households are not diversified are congruent with Dzanku and Sarpong (2011) who indicated that risk-averse rural households may not diversify agriculturally especially if their food security is threatened. Since households in Arda Transau are food insecure (Mandishekwa and Mutenheri 2018), hence in line with the Dzanku and Sarpong (2011) argument, the IDPs may not diversify because they think the activity may exacerbate their food insecurity. Further to that, poorer households were noted to be less likely to diversify especially into cash crop because they cannot afford to ensure food security through food purchases. The study, therefore, supports that argument since Mandishekwa and Mutenheri (2018) established that households are poor since they earn way less than US$100 per month while this study found that households are not diversified agriculturally. Thus, poor households are usually not diversified agriculturally.

Also, the main livelihood activity done by IDPs in Arda Transau is farming which is practiced by 53% of households. They mostly practice food crop production. This finding conforms to the norm than most Zimbabweans are subsistence farmers. Thus, having been practicing farming prior to displacement, the IDPs continued to do so after displacement. However, given that land has been lost (Mandishekwa and Mutenheri 2018), this loss affected IDPs’ activities in as far as farming is concerned thereby again corroborating the works of Omata (2012) who found that limited access to land reduced reliance on farming as a subsistence activity among refugees in Uganda. Therefore, IDPs could not as much rely on farming as they could have done had they a larger land portion. However, animal disposal has increased and was reported by 58% of IDPs in Arda Transau. This poses the risk of future livelihood problems because of decapitalisation.

The main determinant of choosing petty-trading activity over farming was found to be marital status with IDPs who are married being more likely to engage in petty- trading than farming. On the other hand, being a female-headed household increased the probability that a family will choose petty-trading as a major economic activity than farming.

Findings also indicate that religion increases the probability of one engaging in diversified.

agriculture (Table 7). Thus, being a religious male-headed household implies that the household would engage in diversified agricultural activities. Since most of the IDPs are Christians, it means that being a Christian is likely to increase the probability of a family engaging in diversified agricultural activities through knowledge shared through social capital which is used to the benefits of the household. Thus, as households gather for religious purposes, they also engage in discussions on how to improve their livelihoods leading to use of diversified agriculture.

A seemingly striking surprise is the finding that family size reduces the probability that a household engages in diversified agriculture. This finding implies that large households, especially the female-headed ones, are more concerned with food crop cultivation alone because food crop cultivation dominated among other activities. This finding may be considered proper given too limited land available to the IDPs. However, it would have helped them by engaging in say both livestock production and food crop production since animal dung will be used as manure thereby lessening the burden for households’ demand for chemical fertilizers.

Furthermore, the findings show that married IDPs in Arda Transau are more likely to depend on petty trading other than farming compared to non-married IDPs. Possible reasons may be that the couples will help each other to scavenge for items to sell. For instance, if a family is to engage in firewood selling business, the husband may help by offering labour to cut the wood, draught power to carry the wood and even do the selling. On the other hand, if it is a non-married household head, it may be difficult to do this at a scale where two members can do. Also, if the household head is a woman, the probability of engaging in petty trading increases as compared to being a farmer. Possibly, this is so because petty-trading may be considered feminine activity compared to farming. Some men may feel their masculinity has been compromised by engaging in petty-trading, for instance, selling firewood. They would, therefore, spend most of their time in farming than selling small items.

An activity done by a household prior to displacement increases the probability that the household will engage in it after displacement than being a farmer. Thus, IDPs in Arda Transau are more likely to engage in what they did before displacement than just being farmers. One possible reason is that they the fear of the unknown hence they are using risk avoidance as a strategy. Venturing in new territories may be dangerous especially in an area like Arda Transau where IDPs are left on their own without support from government, NGOs and even the companies that displaced them.

This view then corroborates the notion that families are engaging in petty-trading since they have been in it prior to displacement (Nyawo et al. 2012), a finding similar to that by Amisi (2006) who indicated that displacees usually engage in what they did before displacement. Of note is that prior to displacement, some families were into sugar cane and vegetables selling businesses and also basket- and mat-weaving and selling business. Mostly these petty trading activities are associated with women hence the finding that women are more likely to be in petty-trading than men.

On the other hand, women IDPs in Arda Transau were found to be less likely to choose employment as a major livelihood activity than farming, a finding similar to the one found by Amisi (2006) and Islam (2008). This may be derived from the cultural norm that the man must be found to fend for the family by finding a job while the woman is mostly supposed to receive. However, this tendency is on the decrease. For Arda Transau residents it may also emanate from the fact that most IDPs are of the apostolic sect whose doctrine still upholds that men must be the only ones who may work. This may also explain why petty-trading may sprout in such an area though being less than what it was in Chiadzwa. On another note, women unemployment may be mostly influenced by the general tendency of general unemployment in the country such that they will be less motivated to rely on employment especially those whose husbands lost jobs from the mining companies.

Conclusions

Three major conclusions of this study can therefore be spelt out. The first one is that choice of economic activities among IDPs is determined by what they did prior to displacement. Thus, those households who were into petty-trading are likely to engage in petty-trading as an economic activity even after displacement, while those who were mostly reliant on agriculture are likely to do likewise after displacement.

Secondly, male- and female-headed households’ choices of whether to engage in diversified agricultural activities are determined by different factors. Male-headed households’ decisions to diversify were determined positively by age and religion while female-headed households’ choices depended negatively on family size and positively on duration of stay. Therefore, it could be concluded that households headed by old aged religious male heads are likely to engage in diversified agricultural activities than their counterparts who may be young and non-religious. On the other hand, those households headed by female heads, with a large family size, were found to be less likely to diversify. However, female-headed households who have been in Arda Transau for relatively longer periods were found to be more likely to diversify their livelihoods. Thirdly, male and female IDPs employ different livelihood strategies after displacement. While men rely more on agriculture, females rely more on petty-trading as an economic activity after displacement. Possible explanations include that males may consider petty-trading a feminine activity hence will not like to be seen engaging in that activity even if it may be better paying than other economic activities.

Policy implications

Given that the IDPs, just like any other Zimbabwean peasant are mainly into food crop cultivation, the government is urged to supply agricultural inputs to the Arda Transau residents. Such assistance, even inputs from the presidential input scheme, has not been given to these IDPs. This will boost household food security and enhance economic activity of the displaced. This on its own is a sign of marginalization, a finding which was made by Mandishekwa and Mutenheri (2018). This assistance should not, however, be gender biassed.

IDPs in Arda Transau are advised, wherever possible, to avoid use of erosive coping strategies such as asset disposal without replacement. These approaches have an effect of reducing future subsistence base and hence future livelihood will be insecure. The same applies to use of abusive strategies such as indiscriminate cutting down of trees for firewood. The future implications are environmental degradation and other associated subsequent effects if no replenishment occurs.

Benefits from diversification seem to be less understood by IDPs in Arda Transau. The submission of this study is that farmers must diversify their activities to minimize risk of failure especially agriculturally. Intercropping may be practiced in Arda Transau where a farmer may grow maize concurrently with groundnuts. Very few farmers have been engaged in both livestock and food crop cultivation yet animals provide the much needed organic fertilizer. Given the level of incomes of the IDPs which makes them unable to afford artificial fertilizers, this may prove to be a useful alternative. However, this may prove difficult given that families have lost their animals. Be that as it may, most of the IDPs still can afford a few chickens which they may raise and from which they may get manure for application in their 1 ha of arable land. They can also dispose the chickens in times of need. Along this line of thinking, Ellis (1999) noted that diversified livelihood systems are less vulnerable and are likely to be more sustainable than undiversified livelihoods. Ellis (ibid) also noted that in the context of declining farm sizes, as in Arda Transau where households became landless after displacement (Mandishekwa and Mutenheri 2018), diversified livelihoods will be helpful.

It is recommended that households must desist from depending much on one agriculture activity. Ganiyu and Omotayo (2016) argue that this approach leads to higher levels of household poverty and food insecurity. Thus, households must diversify as much as possible to reduce risk associated with one crop, that is, use of horizontal diversification is advised. Again, vertical diversification can be used where households engage in farm and off-farm income-earning activities, an approach with a possibility of reducing diversifiable risk.

The study observes that the companies that displaced these households are no longer assisting the displacees. It is recommended that ZCMDC (the successor to the companies that displaced households) and even the government assist these households either through food for work programmes or free food handouts. This is in line with the Levron (2013) observation that humanitarian assistance has been critical particularly to households without robust social support. However, it was also noted that these activities must not substitute sustainable livelihoods activities. Therefore, these must be used as short term measures.

Religion has been found to be a significant determinant of diversified agricultural activities. With that in mind, using religion as a proxy of social capital, the study proposes that households in Arda Transau may engage in wider participation in such community associations so as to get necessary information on how to make a better living. In these associations, families share ideas which will benefit even the entire community through co-operative behaviour. This can also be used as a channel of disseminating early adoption techniques in agriculture from knowledge acquired through agricultural extension services.

The study proposes that in as much as households may want to make a living through firewood selling, this approach is not sustainable if no replacement of trees occurs. Households must find other ways of sustaining themselves that are sustainable. Indiscriminate cutting down of trees leads to well-known consequences such as deforestation whose long-term effect is desertification, soil erosion from desertification and a plethora of other consequences. Also, the common property resource (common forestry) will be depleted and the future generation will be deprived of access to that. Therefore, households must engage in sustainable livelihoods activities since “sustainable livelihoods will also provide the resources and conditions for enhancement and exercise of capabilities” (Chambers and Conway 1991, 6).

In future, the to-be displaced households are advised not to deplete their animals (where possible) when displacement is imminent. The study proposes that if IDPs had opted for moving their animals to other areas where they have relatives or friends, IDPs may have been able to sustain themselves through sale of animals in the worst-case scenario. This approach has also worked for drought-striken rural areas of Zimbabwe where animals (particularly cattle) were loaned to relatives during drought and retrieved in good seasons.

The findings from the study reveal that women, compared to men, are a bit sceptical about engaging in new activities. The study, therefore, recommends that women must be quick to adapt to change because they may be left behind in most cases if they maintain such wait and see attitude. This may elongate their impoverishment where male IDPs will have been rehabilitated after displacement.

Despite being practiced by a few households, plot diversification might be a lucrative livelihood strategy. Given reduced land size experienced by households, plot diversification might help in ameliorating the effects of land size, plant disease and as a pest control measure. In this regard, families are advised to employ the plot diversification strategy. A possible hindrance might, however, be affordability. In cases where rentals have to be paid for the plots; the use of plot diversification might be limited. In such cases, households must weigh the benefits of the use of the strategy versus the costs incurred.

Despite being largely insignificant in this study as a determinant of livelihood activities, education remains a key to sustainable livelihoods. There is a need for education for livelihood-linked capability. In this regard, the study proposes that the government (local or national) introduces these livelihood linked education programmes such as the once prominent Master Farmer programme. Though biassed towards agriculture, it can go a long way in disseminating information to IDPs in Arda Transau who are currently in the experimental stage of their agricultural life and even beyond that area. Thus, agricultural extension services available in other areas are supposed to be availed to these displacees.

This study does not profess to have covered everything as far as economic activities of the forcibly displaced people are concerned. It, therefore, acknowledges that there is a need to consider other aspects of economic activities among displaced households. Such areas include the need to capture the time dynamism in activities of the displaced. Thus, while this study is static in nature, future studies are encouraged to capture the time-varying activities with time between the time of displacement and a certain period after. This will capture the coping strategy use and possible rehabilitation as propounded by Davies (1993).

References

  1. Abraham R, Lien L, Hanssen I (2018) Coping, resilience and posttraumatic growth among Eritrean female refugees living in Norwegian asylum reception centres: a qualitative study. International Journal of Social Psychiatry 64(4):359–366

    Article  Google Scholar 

  2. Abdulai M A (2016) Internal migration determinants: Evidence from northern region of Ghana, Ghana Journal of Development Studies 13(1):1–17

  3. Acevedo MC (2016) What do we know about the effectiveness of ex-ante risk management strategies in Latin America and the Caribbean?, background paper for the World Bank book “aggregate shocks, household welfare and ex-ante risk management strategies in Latin America and the Caribbean” 2016

  4. Adam J (2008) Displacement, coping mechanisms and the emergence of new markets in Ambon, Conflict Research Group, Working Paper No. 9

  5. Adeniyi AB, Daud AS, Amao O, Omotayo AO (2016) Determinants of rural Women’s livelihood in Ibarapa north local government area of Oyo state, Nigeria. Journal of Human Ecology 56(1–2):84–90

    Article  Google Scholar 

  6. Alzoubi FA, Al-Smadi AM, Gougazeh YM (2017) Coping strategies used by Syrian refugees in Jordan. Clin Nurs Res 00(0):1–26

    Google Scholar 

  7. Amirthalingam K, Lakshman RWD (2009) Displaced livelihoods in Sri Lanka: an economic analysis. J Refug Stud 22(4):502–524

    Article  Google Scholar 

  8. Amisi BB (2006) An exploration of the livelihood strategies of Durban Congolese refugees. Centre for Civil Society Research Report No. 46, 1, pp 195229

  9. Baltas G, Doyle P (2001) Random utility models in marketing research: a survey. J Bus Res 51:115–125

    Article  Google Scholar 

  10. Barbelet V (2017) Livelihood strategies of Central African refugees in Cameroon. HPG Working Paper, March 2017

  11. Bello ARS, Daoud SAM, Baig MR (2014) Coping strategies of Darfurians displaced women in Khatoum. Journal of Agricultural Extension and Rural Development 6(5):168–174

    Article  Google Scholar 

  12. Beyene S (2012) Livelihood diversification among the pastoral and Agropastoral groups in the upper Awash Valley, Ethiopia. Journal of Human Ecology 39(3):241–253

    Article  Google Scholar 

  13. Bird K, Shepherd A Scott A Butaumocho B (2002) Coping strategies of poor households in semi-arid Zimbabwe, final technical report for natural resources systems Programme (NRSP), Project Number: R7545

  14. Borghans L, Golsteyn BHH, Heckmen JJ Meijers H (2009) Gender differences in risk aversion and ambiguity aversion, NBER Working Paper No. 14713

  15. Bozzoli C, Bruck T, Muhumuza T (2011) Activity choices of internally displaced persons and returnees: quantitative survey evidence from post-war Northern Uganda, Microcon Research Working Paper 53, Brighton: Microcon

  16. Bruck T (2004) Coping strategies in post-war rural Mozambique, HiCN Working Paper 02

  17. Bruck T, Danzer AM (2007) Coping with peace: household activity choices in post-war northern Mozambique. Institute for Economic Research, German

    Google Scholar 

  18. Cameron AC, Trivedi PK (2005) Microeconometrics methods and applications. Cambridge Press, Cambridge

    Google Scholar 

  19. Cernea MM (1997) The risks and reconstruction model for resettling displaced populations. World Dev 25(10):1569–1587

    Article  Google Scholar 

  20. Chambers R Conway GR(1991) Sustainable rural livelihoods: practical concepts for the 21st century, IDS Discussion Paper No 296, December 1991

  21. Chatterjee S, Hadi AS (2006) Regression analysis by example. John Wiley and Sons, New Jersey

    Google Scholar 

  22. Davidson R, MacKinnon JG (1999) Econometric theory and methods. Oxford University Press, Oxford

  23. Davies S (1993) Are coping strategies a cop out? IDS Bull 24(4):60–72

    Article  Google Scholar 

  24. Dercon S (2000) Income risk, coping strategies and safety nets, Centre for Study of African Economies , Oxford University, WPS/2000.26

  25. Deressa TT, Ringler C and Hassan RM (2010) Factors affecting the choices of coping strategies for climate extremes: the case of farmers in the Nile Basin of Ethiopia, IFPRI Discussion Paper 01032

  26. Dolan CS (2004) “I sell my labour now”: gender and livelihood diversification in Uganda. Canadian Journal of Development Studies 25(4):643–661

    Article  Google Scholar 

  27. Dow JK, Endersby JW (2004) Multinomial probit and multinomial logit: a comparison of choice models for voting research. Elect Stud 23:107–122

    Article  Google Scholar 

  28. Downing TE (2014) Does the Kosovo Power Project’s proposed forced displacement of Kosovars comply with International involuntary resettlement standards? The Kosovo Civil Society Consortium for Sustainable Development, http://allthingsaz.com/wp-content/uploads/2014/04/Final-Draft-Downing-Involuntary-Resettlement-at-KPP-Report-2-14-14.pdf (accessed 12.07.14)

  29. Dzanku FM, Sarpong D (2011) Agricultural diversification, food self-sufficiency and food security in Ghana-the role of infrastructure and institutions. In: Djurfeldt et al (eds) CAB International, African Smallholders: Food Crops, Markets and Policy, pp 189–213

    Google Scholar 

  30. Ellis F (1998) Household strategies and rural livelihood diversification. J Dev Stud 35(1):1–38

    Article  Google Scholar 

  31. Ellis F (1999) Rural livelihood diversity in developing countries: evidence and policy implications, Natural Resource Perspectives, No.40, ODI

  32. Ganiyu MO, Omotayo AO (2016) Effects of livelihood activities on the households’ food security in Ogbomoso south local government area of Oyo state, Nigeria. J Hum Ecol 56(1–2):107–113

    Article  Google Scholar 

  33. Hausman J, McFadden D (1984) Specification tests for the multinomial Logit model. Econometrica 52(5):1219–1240

    Article  Google Scholar 

  34. Helgeson J, Dietz S, Hochrainer-Stigler S (2013) Vulnerability to weather disasters: the choice of coping strategies in rural Uganda. Ecology and Society 18(2)

  35. Hoadley M (2008) Development-induced displacement and resettlement-impoverishment or sustainable development, School of Mining Engineering, University of Witwatersrand

  36. Horn R (2009) Coping with displacement: problems and responses in camps for the internally displaced in Kitgum, northern Uganda. Intervention 7(2):110–129

    Article  Google Scholar 

  37. IOM Niger (2018) Return intentions and socio economic opportunities of displaced populations in the Diffa region. Migration, UN

    Google Scholar 

  38. Islam MZA (2008) Survival strategies of the female displacees in rural Bangladesh: a study of two riparian villages on the right bank of Jamuna, http://www.irsa-world.org/XII/papers/2-1.pdf (Accessed 19.07.2019)

  39. Israel DC Briones RM (2014) Disasters, poverty, and coping strategies: the framework and empirical evidence from micro/household data-Philippine case, Philippine Institute for Development Studies, Discussion Paper Series No 2014-06

  40. ITUC (2008) Mini action guide, child labour, https://www.ituc-csi.org/IMG/pdf/guide_CL_EN_Final.pdf (Accessed, 19.07.2019)

  41. Jabeen H, Johnson C, Allen A (2010) Built-in resilience: learning from grassroots coping strategies for climate variability. Environ Urban 22(2):415–431

    Article  Google Scholar 

  42. Jacobsen K (2002) Livelihoods in conflict: the pursuit of livelihoods by refugees and the impact on the human security of host communities. Int Migr 40(5):95–123

    Article  Google Scholar 

  43. Jones AM(2005) Applied econometrics for health economists: a practical guide, 2nd ed, Prepared for the Office of Health Economics, 2005

  44. Justino P (2012) Resilience in protracted crises: exploring coping mechanisms and resilience of households, communities and local institutions, higher level expert forum on food insecurity in protracted crises, Rome, 13-14 September 2012

  45. Khawaja NG, White KM, Schweitzer R, Greenslade J (2008) Difficulties and coping strategies of Sudanese refugees: a qualitative approach. Transcultural Psychiatry 45(3):489–512

    Article  Google Scholar 

  46. Kiboro CN (2017) Influence of social capital on livelihood outcomes for the internally displaced persons in Kenya: a social capital approach. Eur Sci J 13(26):266–280

    Google Scholar 

  47. Kleinbaum DG, Klein M (2010) Logistics regression: a self-learning text, 3rd edn. Springer, New York

    Google Scholar 

  48. Krejcie RV, Morgan DW (1970) Determining sample size for research activities. Educ Psychol Meas 30:607–610

    Article  Google Scholar 

  49. Lekprichakul T (2009) Ex ante and ex post risk coping strategies: how do subsistence farmers in Southern and Eastern Province of Zambia Cope? www.chikyu.ac.jp/resilience/files/.../ResilienceProject_Report2009_16.pdf (Accessed25/09/18)

  50. Levron E (2013) Livelihoods, gender and internal displacement in cote d’Ivoire: between vulnerability and resiliency, Paper Prepared for the Brookings-LSE Project on Internal Displacement

  51. Loewenson R (1991) Child labour in Zimbabwe and the rights of the child. J Soc Dev Afr 6(1):19–31

    Google Scholar 

  52. Maclure M (1991) The case-crossover design: a method for studying transient effects on the risk of acute events. Am J Epidemiol 133(2):144–153

    Article  Google Scholar 

  53. Maddala GS (1993) Limited-dependent and qualitative variables in econometrics. Cambridge University Press, Cambridge

    Google Scholar 

  54. Magaramombe G (2010) ‘Displacement in place’: agrarian displacements, replacements, and resettlement among farm workers in Mazowe District. J South Afr Stud 36(2):361–375

    Article  Google Scholar 

  55. Mandishekwa R and Mutenheri E (2018) The economic consequences of internal displacement in Arda Transau, Zimbabwe, Paper presented at the 8th KESSA-AISA-Multimedia University International Interdisciplinary Conference on Development from below and from above in Africa held at Multimedia University of Kenya, 27-30 June 2018

  56. Martinoty L (2015) Intra-household coping mechanisms in hard times: the added worker effect in the 2001 Argentine Economic Crisis, GATE Lyon Sainte-Etienne, WP 1505, March 2015

    Google Scholar 

  57. Motsholapheko M R, Kgathi, D L and Vanderpost C (2012) Rural livelihood diversification: a household adaptive strategy against flood variability in the Okavango Delta. Botswana, Agrekon, 51:4: 4162

  58. Mukarami E (2017) Household consumption risk and coping strategies in Tajikistan: evidence from household surveys, ADBI Working Paper 738, May 2017

  59. Mukwada G (2012) Scales of vulnerability: resettlement and exposure to multiple stressors and shocks in Zimbabwe. J Hum Ecol 40(2):101–113

    Article  Google Scholar 

  60. Mulinya C (2017) Factors affecting small scale farmers coping strategies to climate change in Kakamega County in Kenya. IOSR Journal of Humanities and Social Science (IOSR-JHSS) 22(2):100–109

    Google Scholar 

  61. Nelson JA (2012) Are women really more risk-averse than men?, Institute for new Economic Thinking, INET research note #012

  62. Northcote MA (2015) Enterprising outsiders: livelihood strategies of Cape Town’s forced migrants, Unpublished Master’s Dissertation, The School of Graduate and Postdoctoral Studies, The University of Western Ontario London, Canada

  63. Nyawo VZ, Goredema D, King MD (2012) The socio-economic impact of Chiadzwa informal diamond mining on the lives of the people of Chiadzwa and its hinterland. The Dyke 6(2):6–21

    Google Scholar 

  64. Ogbozor E (2016) Resilience to violent extremism: the rural livelihood coping strategies in the Lake Chad Basin, Households in Conflict Network, HiCN Working Paper 237

  65. Omata N (2012) Refugee livelihoods and the private sector: Ugandan case study, Refugee Studies Centre, Working Paper Series No. 86, November 2012)

  66. Owen JR, Kemp D (2015) Mining-induced displacement and resettlement: a critical appraisal. J Clean Prod 87:478–488

    Article  Google Scholar 

  67. Perez-Tuglia R (2009) Applied econometrics using Stata, Lecture Notes

  68. Rashid DA, Langworthy M Aradhyula S (2006) Livelihood shocks and coping strategies: an empirical study of Bangladesh households, Paper presented at the American Agricultural Economics Association Annual Meeting, Long Beach, California, July 23-26, 2006

  69. Renner W, Salem I (2009) Post-traumatic stress in asylum seekers and refugees from Chechnya, Afghanistan, and West Africa: gender differences in symptomatology and coping. Int J Soc Psychiatry 55(2):99–108

    Article  Google Scholar 

  70. Roussanov N and Savor P (2014) Marriage and managers’ attitudes to risk, finance papers, paper no. 10-2024, University of Pennsylvania

  71. Schweitzer R, Greensdale JH, Kagee A (2007) Coping and resilience in refugees from Sudan: a narrative account. Aust N Z J Psychiatry 41(3):282–288

    Article  Google Scholar 

  72. Seguin ML (2016) Resource Loss and coping strategies used by internally displaced women in Georgia: a qualitative study, Unpublished Ph.D. Thesis. London School of Hygiene and Tropical Medicine, London

  73. Sen A K (1981) Poverty and famines: An essay on entitlement and deprivation,Oxford University Press, Oxford

  74. Spence PR, Lachlan KA, Burke JM (2007) Adjusting to uncertainty: coping strategies among the displaced after hurricane Katrina. Sociol Spectr 27(6):653–678

    Article  Google Scholar 

  75. Terminski B (n.d.) Mining-induced displacement and resettlement: social problem and human rights issue (a global perspective)

  76. ZimSTats (2015) 2014 child labour report, Government of Zimbabwe

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Robson Mandishekwa.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Mandishekwa, R., Mutenheri, E. The economic activities among mining-induced displacees in Arda Transau, Zimbabwe. Miner Econ 34, 51–70 (2021). https://doi.org/10.1007/s13563-019-00215-1

Download citation

Keywords

  • Mining
  • Internal displacement
  • Economic activities
  • Diversification
  • Zimbabwe

JEL

  • Q15
  • Q34
  • Q38