Large-scale land acquisitions, state authority and indigenous local communities: insights from Ethiopia

Abstract

The convergence of diverse global factors – food price volatility, the increased demand for biofuels and feeds, climate change and the financialisation of commodity markets – has resulted in renewed interest in land resources, leading to a rapid expansion in the scope and scale of (trans)national acquisition of arable land across many developing countries. Much of this land is on peripheral indigenous peoples’ territories and considered a common property resource. Those most threatened are poor rural people with customary tenure systems – including indigenous ethnic minority groups, pastoralists and peasants – who need land most. In Ethiopia large areas have been leased to foreign and domestic capital for large-scale production of food and agrofuels, mainly in lowland regions where the state has historically had limited control. Much of the land offered is classified by the state and other elites as ‘unused’ or ‘underutilised’, overlooking the spatially extensive use of land in shifting cultivation and pastoralism. This threatens the land rights and livelihoods of ethnic minority indigenous communities in these lowlands. This article argues that recent large-scale land acquisitions are part of state strategy for enforcing political authority over territory and people. It examines the implications of such strategy for indigenous ethnic minority groups, focusing particularly on the Benishangul-Gumuz region.

Keywords:

• Large-scale land acquisition
• state authority
• land-based social relations
• indigenous ethnic minority groups
• livelihoods
• displacement
• Ethiopia

Introduction

Since 2008 a sizeable body of research by scholars and activists has shown the scope, context and implications of contemporary global land grabbing in greater depth. Several of the studies have addressed the enormous scale and speed of expansion of (trans)national land deals, and emphasised its impact in changing agrarian structures, rural social relations and rural livelihoods.¹ In most cases much of the appropriated land is on peripheral indigenous peoples’ territories and is considered a common property resource. This article argues for viewing contemporary large-scale land acquisitions as part of state strategy for consolidating and enforcing political authority and control over people and territory where states have historically had limited reach. It examines the implications of such strategy for indigenous ethnic minority groups in Ethiopia, with a particular focus on the Benishangul-Gumuz region.

Over the past decade the acquisition of arable land has expanded rapidly in many African and other developing countries. Although land has always been central to the livelihoods of millions of smallholders, the recent convergence of diverse global factors – such as food price volatility, the increased demand for biofuels and feeds, climate change, and the financialisation of commodity markets – has resulted in a renewed and growing interest in land resources. This has led to a rapid expansion in the scope and scale of (trans)national acquisition of arable land across many developing countries – mainly in sub-Saharan Africa but also in Southeast Asia, the former Soviet countries and Latin America – for the cultivation of food crops and biofuels.² This growing interest in land resources, both domestic and transnational, has increased the pressure on their availability, suggesting an increased competition between different interests in and claims to the same resource. Those most threatened are poor rural people – including ethnic minorities, indigenous people, pastoralists and peasants – who, in fact, need land the most.³ The land rights of these subaltern groups rarely extend beyond use rights, which are often unprotected and weak – making them highly vulnerable to being dispossessed and forced off their land more easily. In particular, the very fact that the land acquisitions are taking place mainly in countries where the state is the de jure or de facto owner of most land puts the subaltern rural groups with customary tenure systems at risk.⁴ This means the state can easily expropriate land legally for ‘the public interest’ or for ‘development’ purposes, meaning, in some cases, that there is no obligation on the part of the state to pay compensation to anyone.

In most cases African governments have welcomed such large-scale land investments, considering them an opportunity to transform their agricultural sector – seen as ‘backward’, subsistence-based smallholder farming – particularly through technology transfer, the expansion of local infrastructure and rural employment generation. Moreover, large-scale investments are also seen as a means to ostensibly achieving national food security.⁵ As such, host governments in many African countries have responded by promoting investor-friendly land-market policies, such as low land rents, tax waivers, and limited restrictions on production and exports.

When exploring the main issues embedded in recent large-scale land acquisitions, debates around their contemporary political economy have tended to follow one of two main lines of argumentation, the choice between one or the other influencing how a range of interest groups perceive and contest the politics around large-scale land acquisitions. The first line of argument focuses on the implications for local communities, arguing that such acquisitions threaten the livelihoods and food security of millions of poor rural people, as well as raising the risks of environmental destruction and social and political upheavals. This narrative is in stark contrast to the positions taken by mainstream international financial institutions. The former firmly argues that such acquisitions lead to a type of agrarian structure that produces the processes of accumulation ostensibly predicated upon mechanisms of what the agrarian political economy literature calls social differentiation.⁶ Peasant differentiation and expanded marginalisation will, in turn, lead to de-peasantisation and massive proletarianisation, forcing peasants to subsist mainly by selling their labour. Such corporate-driven agricultural structures exemplify agrarian capitalist accumulation, which ‘rules out a place for peasants, physically expelling them from the land, and epistemologically removing them from history’.⁷

The second line of argument is mainly spearheaded by international financial institutions and development agencies, and constitutes the mainstream development discourse. It argues that, if managed well, large-scale land investments have considerable potential to contribute to multiple development objectives in the case of developing countries.⁸ While acknowledging the challenges and risks associated with the wave of these investments, this line argues that said investments can be minimised and regulated to ensure that local communities are not adversely affected. To this end, mainstream arguments propose the need to improve the transparency and accountability of the deals and processes culminating in such investments, in order to translate the anticipated opportunities into a ‘win-win-win’ deal – one in which benefits will be shared equitably between local communities, host governments and investors. From this optimistic standpoint, the World Bank and others (FAO, the International Fund for Agricultural Development (IFAD) and UNCTAD) have proposed the Principles for Responsible Agricultural Investments (RAI), which aim to regulate these investments for a better result. However, these attempts to ensure a ‘responsible investment’ have received wide-ranging criticism.⁹ More recently the Committee on World Food Security has come up with more far-reaching, but still voluntary, guidelines (Voluntary Guidelines on the Responsible Governance of Tenure).¹⁰

A recurring theme in recent debates about the wave of large-scale land acquisitions is the key role of the state. The debates inherently involve profound questions about state authority, governance and property.¹¹ The state is conceived as a complex and differentiated actor with internal inconsistency in its agenda and practice at different levels; moreover, the state itself is a site of struggle over resources, power and authority among its various constituents.¹² As Watts put it, ‘the state [can] be opened as a theatre in which resources, property rights, and authority are struggled over’.¹³ Thus, there is a need ‘to unbundle the state, to see government and governance as processes, people and relationships’ to gain a better insight with regard to the role of the state in the land acquisitions.¹⁴ More specifically, Wolford et al have emphasised the ‘need to know more about the nature of states themselves…the motivations of particular actors as well as the capacity of governments and the political cultures that shape the path from policy to practice’.¹⁵

In Ethiopia such large-scale land acquisitions – both by domestic and foreign investors – have been taking place over the past few years, mainly in the country’s lowland regions. The state has already made ‘available’ millions of hectares of land for commercial agricultural investment. The government claims that the land offered is ‘underutilised’, ‘unused’ or ‘idle’, which implies that such land neither belongs to, nor is used, by anyone. However, this claim tends to overlook existing land-use types and different categories of users in these areas. The use of land mainly by pastoralists and shifting cultivators in the targeted lowland areas is contested by the state, which perceives the former’s land uses as essentially ‘unsustainable’ or inefficient.¹⁶ Such a claim highlights the weak recognition of land rights and livelihoods of indigenous ethnic minority groups. This official perception and image of existing land uses in the lowlands has proven formative in the design of state policy that focuses on leasing vast tracts of land to investors in these areas. For the state farmland investment provides an opportunity for extending its reach, exerting its power over peripheral areas and peoples, as well as for retaining and expanding the extraction of resources from such areas on a grand scale. The land investment, which has primarily been driven by the state, can thus be seen as a mechanism of retaining, consolidating and expanding federal state authority and control over territory and people.

Although new opportunities for national growth may be created from increases in land investments, fundamental questions still remain on the implications of these investments for the land rights of poor indigenous local communities who are barred from using the leased land. This is especially problematic given the scope and scale of the land acquisitions. Empirical evidence is sparse with regard to actual and potential impacts of land-use change on the poor. While large-scale land acquisitions do not always result in local communities losing their land or having their livelihoods subverted, many recent acquisitions have, in fact, entailed the dispossession and displacement of rural households, and the damaging of local livelihoods, food security and access to key natural resources.¹⁷

In Ethiopia the state has the ability to determine who gets to use land and for what purpose. Given its current policies of promoting large-scale export-oriented agriculture based on foreign capital and technology, it is common for local communities to suffer as a result of large-scale land transfers. Typically this is because they cannot effectively negotiate with or defend their rights against state and corporate actors in a situation of wider inequalities in bargaining power.¹⁸ The state formally owns the land and has extraordinary power to determine who does (and who does not) get to use land resources, how much, in what ways, and under what circumstances, making it particularly difficult for weakly organised indigenous ethnic minority groups to negotiate or secure adequate compensation from corporate investors or state actors.

Under such circumstances this study argues that land acquisitions threaten the economic, cultural and ecological survival of indigenous local communities. The Gumuz ethnic groups – who depend on customary forms of land access and control, and whose livelihoods are based heavily on access to natural resources – are being particularly affected. Through a case study in some selected administrative woredas (districts) of the Benishangul-Gumuz region – one of the focus areas of the current land acquisitions in the country – this article examines the implications of land acquisitions for indigenous local communities.¹⁹

The Benishangul-Gumuz region is currently one of the nine administrative regions under Ethiopia’s federal political system, based on ethnic regional administration. This region is located in the northwest of the country, sharing an international border with Sudan and South Sudan in the west. Inside Ethiopia it shares borders with the Amhara and Oromiya regional states (Figure 1). It occupies an estimated total area of 50,380 sq km,²⁰ and has a total population of 670,847.²¹ The population consists of the indigenous ethnic minority groups of Berta, Gumuz, Shinasha, Mao and Komo. The region is also inhabited by settlers from other regions with diverse ethnic origins. The indigenous groups depend on a customary land tenure system of communal ownership and rely mainly on shifting cultivation for their livelihood. This is supplemented with other subsidiary activities, such as hunting, gathering, fishing, livestock raising, traditional alluvial gold mining, and honey production and collection.

Figure 1. Map of the Benishangul-Gumuz regional state, showing the locations of the woredas studied. Sources: This map was created by Nigussie Abdissa, a colleague who works in the Tana-Beles Integrated Water Resource Development Project, Assosa, using ArcGIS 10.1 by Esri (© Esri). Boundary data were obtained from the Benishangul-Gumuz Bureau of Finance and Economic Development (BoFED).

The article’s empirical analysis draws on information collected during intensive fieldwork in the Benishangul-Gumuz regional state from April to June 2012.²² The data comes from semi-structured, in-depth interviews with key informants, focus group discussions (FGDs), direct field observation and document review.²³ The in-depth interviews were conducted with 17 key Gumuz informants and 14 selected government officials and experts at various hierarchical levels. In-depth interviews with three informants from the Berta ethnic group were also conducted.²⁴ In addition, in the selected case study villages, a total of seven FGDs were conducted with the communities affected by land acquisitions. Interviews were also conducted with five managers of investment projects operating in the study areas. The article also draws from the author’s recent short field visit to the same study areas in April 2015.

The remainder of the paper comprises three sections. The first presents a brief overview of large-scale land investment in Ethiopia and sheds light on the role of the state as a key driver of the investment. The second section analyses the implications of these investments for indigenous ethnic minority groups. The third and final section draws conclusions.

Overview of large-scale land investment in Ethiopia

Since 2005 Ethiopia has emphasised private sector investment in agriculture to accelerate growth and eradicate poverty. At present the government is pursuing a strategy of promoting large-scale, export-oriented agricultural investment as a major part of its overall development strategy, which envisages making Ethiopia a food-secure middle-income country by 2025. To this end, land from the lowland regions has been leased on a grand scale to both domestic and foreign investors. So far large swathes of land have already been transferred. Various support and incentive mechanisms aimed at attracting foreign direct investment (FDI) have been put in place since 2002–03;²⁵ these were further amended in 2008.²⁶ One particular example was the keen promotion and support of private sector investment in land resources, primarily to generate greater foreign exchange earnings. This measure was part of the government’s last five-year Growth and Transformation Plan (GTP), which had been implemented between 2010–11 and 2014–15. By way of this measure the government expected to generate a total of US$6.58 billion from the agricultural export market by the end of the GTP period (2015). For this to materialise, it also expected to transfer an estimated 3.3 million hectares of land (in addition to land already allotted before the GTP) over this same period to large-scale agricultural investors.²⁷

To accelerate the transfer of land to investors, an Agricultural Investment Support Directorate (AISD) was set up at the federal level in 2009. This federal body was reorganised in 2013 as the Agricultural Investment Land Administration Agency. Its role was to transfer the agricultural investment lands identified into a centralised pool referred to as the Federal Land Bank, which it would also administer.

Although estimates vary and are, at times, unreliable, the World Bank report notes that the total amount of land transferred to investors in Ethiopia between 2004 and 2008 amounted to 1.2 million hectares.²⁸ Yet another report from the Oakland Institute estimates the total amount to have reached more than 3.6 million hectares as of January 2011.²⁹ Rahmato puts the estimated amount of already transferred land at 3.5 million hectares, and estimates the total amount to reach seven million hectares by the end of 2015.³⁰

Recent figures from the government’s annual progress report on GTP implementation show that, during the 2012–13 fiscal year, about 3.31 million hectares of large-scale investment land were identified and transferred to the federal land bank.³¹ Nevertheless, over the past three GTP years, no more than 473,000 hectares of land were transferred to investors, of which only 11% was reportedly developed.³² This is far below the GTP’s plan to lease 3.3 million hectares by 2015. In this regard, the government has pointed out some of the major factors that have resulted in low performance, which include delays in investors’ participation and capacity limitations when developing the land transferred to them.³³ Recently the government has terminated the land lease contract of some of the high-profile land investment projects that have failed to deliver on their promises to the government in terms of actually developing the land leased; thus, the land has been restored to the federal land bank.

Ethiopia is said to have large uncultivated arable land that can potentially be developed for agricultural purposes.³⁴ The country has about 51.3 million hectares of arable land, out of which only about 11.7 million hectares are currently being utilised.³⁵ This agricultural land potential is assumed to exist in the peripheral lowland areas.³⁶ According to the Ministry of Agriculture and Rural Development, the lack of capital and technology is the major factor that has constrained the utilisation of the country’s large investment potential for agricultural development.³⁷ This has been a key assumption put forward to justify recent commitments to promote and expand FDI in the land sector.

Seemingly sparsely populated, the country’s lowland regions are the major areas where much of the current land acquisitions in Ethiopia are taking place.³⁸ It is expected that the expansion of large-scale agricultural investments will benefit the country in a number of ways: increasing foreign exchange earnings from the export of crops; creating local employment opportunities; expanding local infrastructure and social services; and creating opportunities for technology transfer, particularly to local farmers.³⁹ Nevertheless, the purported benefits do not always materialise. If they do, they may not necessarily benefit the same people who have lost their land resources or livelihoods as a result of the land acquisitions. Key concerns have been raised over the possible adverse consequences of large-scale agricultural land acquisitions for subaltern groups, particularly for poor, marginalised and vulnerable rural groups.⁴⁰ In the context of these key concerns, the article now sets out to examine the implications of large-scale land acquisitions for local communities.

The implications of land acquisitions for local livelihoods and the environment

While Ethiopia is in general categorised as one of the most land-constrained countries in Africa, a few of its regions still have relatively abundant land that can be used for sustainable cultivation.⁴¹ As already noted, the Benishangul-Gumuz region is one of these few areas, provided due recognition is made of its different ethnic groups’ distinct land-use practices and its fragile ecological contexts. During previous political regimes the region was viewed as peripheral and neglected. Over the past few years a significant number of investors have acquired land across it. However, despite claims of generating high economic and social returns, the rising land acquisitions appear to have had adverse impacts on local land-use practices and land resources, including land dispossession, declining access to resources and environmental destruction. This section demonstrates how current large-scale land acquisitions have created adverse impacts on local livelihoods and the environment.

Land dispossessions and declining access to land resources

One of the major adverse implications of current land acquisitions is the loss of local land rights and land-use practices. As already indicated, most of the indigenous ethnic groups in the region mainly depend on shifting cultivation. Both the Berta and Gumuz people (the dominant groups in the region) and the Mao and Komo are shifting cultivators who practice slash and burn agriculture. Natural resources are the source of basic livelihoods for these groups, on a common ground that provides for gathering forest foods, hunting, fishing, honey collection and traditional alluvial gold mining.

Among the Gumuz land resources are ideally communal property, and rights to these resources are derived from the community. In their customary communal tenure system patrilineal kin groups or clans own all the resources inside the clan territory, which is marked by land features such as rivers, hills, big trees, roads and footpaths. The community makes decisions regarding the overall utilisation of natural resources, while individual members possess only usufruct rights. However, recent increases in large-scale land acquisitions have been affecting this traditional system, leading in turn to the deprivation of local people’s right to access their traditional sources of livelihoods.

The Gumuz, who originally and predominantly inhabited the Metekel area, have suffered continuously from encroachments by different interest groups across political regimes.⁴² In addition to the highlanders’ long established encroachment, the 1980s state-sponsored resettlement schemes and later state farm expansions have further impeded the Gumuz from practising their traditional livelihood activities. In so doing, they have pushed the Gumuz further down to the peripheral lowlands.⁴³ The current rapidly emerging trend of land acquisitions for commercial agricultural investment has created additional challenges, by exerting intensified pressures on them. The Gumuz overwhelmingly perceive land acquisitions by investors as inimical to their local livelihoods and the environment. Although land scarcity is not a problem – at least for the moment – the Gumuz interviewed in all the study kebeles invariably felt that this would soon become a reality as a result of the enclosure of large land resources they had previously had access to under their traditional system of tenure. A Gumuz farmer from the Qotta kebele in the Dangur woreda, for example, explained the current situation as follows:

There are six investors who have acquired land in our kebele. These investors claimed very large tracts of land, leaving our village in the middle. For example, one investor [S&P Energy Solutions, an Indian company] alone took 50,000 hectares of land. Can you imagine how much land is left for us? As we are totally encircled by these investors, we don’t have a hope of expanding our farm lands or to continue practising our traditional farming practices as we used to do in the past, before the arrival of the investors. As to my knowledge, the farmlands of ten households have been taken by one of these investors. We are now left with little land and we are very worried for our children. Because of this situation we are even forced to keep our goats within the village and at the edges of our crop fields. Because of this, our goats often encroach into the crop fields, causing damage to our planted crops. In the past, we used to keep our animals in the fields opposite the village community’s cultivated lands, but this is no longer possible. (Interview, May 2012)

According to an expert in the Guba woreda’s Agriculture Office, some efforts are usually made to protect villages in the process of land allocations (interview, June 2012). One of these measures is that the lands allocated to investors must be beyond a five km radius from villages. This is quite limited, as the farmers are shifting cultivators and also depend on multi-niche livelihood sources accessed from the forest. As part of their customary land practices the Gumuz people do not cultivate their fields intensively. Rather, they cultivate a given plot of land for some three to five years, then leave it fallow when a decline in yield is perceived. Within their clan territory new land is then cleared and cultivated in the same way, until the yield is again seen to decline. In the process the whole or part of the village may also be abandoned if the newly acquired lands are too far from the initial village location. However, the Gumuz do not move to new places each time, but rather move around, returning to the abandoned lands that had been left to regenerate. Now, however, this land practice appears to be changing as a result of the increase in land acquisitions.

The Gumuz farmers interviewed indicated that the lands that had been lying fallow, abandoned villages and forestlands have been largely converted into permanent farmlands by investors. A Gumuz farmer in his late thirties from the Qotta kebele explained how the situation changed:

The government told us that we should stay in a permanent village, farming fields close to our villages so that we will be provided with schools, health posts and water pumps. Recently, a lot of people have been relocated to our village from various scattered places, and the kebele allotted these households with small plots of land for their survival. Because of this situation we cannot continue practising our traditional farming practices anymore, unless we totally move to very remote areas that cannot be reached easily and which aren’t suitable for the cars [tractors] of the investors. (Interview, May 2012)

Another farmer, who had recently been relocated to a newly established village under the regional government’s ongoing programme of villagisation – in which scattered small villages are collected into designated settlements – expressed his sentiments thus:

It is so bad!…land was abundant in the place where I was living before we were relocated to this village. If you are strong, you can clear and cultivate as much land as you can. Here, they gave me a piece of land because there is not much land left for us here. In any direction you go from this village, you will encounter investors’ land. I was about to go back to my previous village, but one of the investors who have been there for the past two years has now taken over all our previous lands. We want our land. As a Gumuz, land is what we have. Now, we are aware that the campaign of collecting our people into big settlements along the main road is meant to take our lands and give it away to investors. Nobody cares about us…What did these investors do for us since they came here? Nothing! What we have seen is destruction, nothing else. (Interview, May 2012)

According to the regional government’s villagisation plan for 2011–12, 19,763 households are expected to be resettled from their scattered settlements to designated villages across most of the woredas within the region. In the Dangur and Guba woredas studied alone the government planned to settle nearly 3000 households into 14 permanent villages.⁴⁴ A regional government official interviewed admitted that, while the major objective of the villagisation programme was to deliver basic infrastructure and services to deprived indigenous communities, an implicit objective was to smooth the expansion of commercial agricultural investments via ‘planned relocations’ (interview, Assosa, April 2012). Gumuz informants interviewed in the Dangur woreda – who had been recently relocated to newly designated villages – were explicit in indicating that most of their previous lands had already been transferred to investors; they expected that the remainder would also inevitably be given away soon (interview, Qotta kebele, May 2012). Some relocated people complained that the provision of infrastructure and services has been minimal in the newly resettled villages, thus challenging the motives behind the programme (interview, Ayicid kebele, 6 June 2012). A Gumuz informant living in a village that was created in the Ayicid kebele (Guba woreda) under the villagisation program explained the situation as follows:

We came to this new village without our consent. The woreda officials instructed us to leave our previous village. Initially we tried to refuse, asking the officials: why do we need to leave? Why are you forcing us? The officials told us that this was an order from the federal government that we cannot refuse, and threatened us to accept it. They said that, if we refuse, they will bring the federal police forces so that we will face the consequences. It was because of this high pressure that we decided to leave. We finally complied with the government’s order, but since we came here we have been faced with a lot of problems. Especially the lack of water is a serious problem. We even managed to construct the houses that you see here from the materials that we brought from our previous villages, by demolishing the previous ones. We are not even provided with water facilities as they promised. It might surprise you to hear, but women now fetch water from our previous village, as it is not far from here. (Interview, Ayicid kebele, 6 June 2012)

Local communities perceived the land acquisition trends as destroying their traditional livelihoods and entailing displacements from their homes. As the field investigation suggests, the threat of displacement has been on the rise in the region. For example, the land deal between Tracon Trading and the federal government – involving 5000 hectares of land in the Dangur woreda – is to displace the villagers from both their cultivated lands and their homes. This is because the land was allocated based on satellite imagery of the area, without undertaking proper verification – the typical process through which investment land is currently identified. A Dangur woreda council official explained that the land in question had actually been cultivated by the Gumuz people of the Dachigeri kebele (interview, Manbouk town, 29 May 2012). He further explained that most of the investors who came to the woreda through the federal government were allocated land in the Qotta kebele, for reasons he did not understand. This concentration created overlaps not only with the local communities’ cultivated lands but also between the lands transferred to investors, leading to several land disputes.

In the Qotta kebele about 14 Gumuz farmers lost their cultivated lands, receiving neither financial compensation for their loss nor new land. As expressed by local informants and confirmed by the kebele Administrator, many other farmers had been cultivating lands already allocated to investment projects; these farmers had been told that they would be made to leave at any moment.

The field observation and discussions with community members also revealed that track roads leading to the various investment projects passed through some farmers’ lands, causing damage to their cultivated areas. The situation caused frustration and insecurity among the local kebele communities, thanks to the heightened pressure on the available land and other natural resources. Many farmers of the kebele felt they were at risk of losing their land; they were also aware of the disappearance of their once abundant land resources. Several of them indicated that they wanted all the investors to leave, persistently appealing to kebele authorities for this reason. In the Gimtiya kebele community leaders estimated that about 30 farmers had lost their farmlands because of the investment projects.

Additionally, the fieldwork in the Homosha woreda revealed that the farmland of eight Berta farmers was taken over by one of the domestic investors, Balzaf Alcohol and Drinks Factory Plc, which has acquired 1031 hectares of land in the Tsori Al Metema kebele. Similarly to the previously mentioned Gumuz farmers of the Qotta kebele, the Berta farmers had not been compensated for their losses. In general the indigenous ethnic groups were made more vulnerable by two factors: the fact that their de facto customary land rights have not been respected, and the fact that their use rights have not been formally registered.⁴⁵ In spite of this, the farmers asserted that their insecurities did not result from their existing customary land tenure system; instead, they stemmed from the increasing land acquisitions (interviews, April–June 2012).

In circumstances that involve overlaps and evictions, local authorities have generally tended to intervene in favour of the investors. This was illustrated by the cases of the Dachigeri and Qotta kebeles, where displacing the entire villages was ‘inevitable’, only postponed to the following year. Moreover, when the disputes are between investors, the demands of those investors who acquired land through the federal government prevailed over those who acquired it through the regional government.

It is also interesting to note that the land investment projects examined almost totally depended on agricultural labourers recruited from other regions in the country, particularly the Amhara region. As these jobs are mostly seasonal in nature, many of the labourers choose to stay in the area after their contracts’ completion. They tend to encroach on the forest to acquire land, which they cultivate for about year; after this period they bring their families to their new settlements. These ‘illegal’ settlements have generated additional challenges for local communities, as they further increase the pressure on the available land resources. This has mainly been the case in the Dangur and Guba woredas, where local authorities see these new settlements as major peace and security challenges, likely as they are to fuel land conflicts.

The Gumuz, in particular, are developing hostile attitudes towards the investment projects. This is not only because they perceive them as instrumental in the destruction of their traditional land-use practices and of the natural environment, but also because they see neither economic or social benefits, nor any future discernible benefits from said projects. Although it has been widely argued by the government that the land investments will offer many economic and social benefits to local communities, few, if any, benefits have been gained by the studied communities. One anticipated immediate benefit for the local communities had been the employment opportunities. However, it seems unlikely that indigenous local communities could become plantation workers, as labourers from the highland areas – mainly from the Amhara region – have filled almost all seasonal wage employments.

That being said, a major troubling development attributed to commercial land investments across the region has been the considerable environmental destruction that occurred in recent years. The arbitrariness of the land allocations, implemented without the necessary detailed socio-cultural, economic and ecological studies, has had adverse impacts on the environmental and natural resources. This in turn has had a considerably negative impact on the indigenous communities, whose livelihoods are heavily based on access to natural resources. In contrast to the widespread concerns and initiatives addressing environmental degradation in the country’s highlands, no attention has been paid to the environmental impacts of the current land investment trends in the Benishangul-Gumuz region. Information obtained from the regional Bureau for Environmental Protection, Land Administration and Use shows that none of the numerous investment projects scattered throughout the region had been subject to environmental soundness assessments (interview, Assosa, April 2012). Moreover, not even a single environmental impact assessment document was found in the responsible offices during the fieldwork.

All Gumuz and Berta informants were deeply unsatisfied with the ways in which the indigenous natural forests have been cleared to prepare farm fields. According to the informants, vast forest areas have been indiscriminately cleared, on occasion burning standing trees to ensure that that they would not regenerate. The local communities have already begun to feel the effects of these practices, including declining forest food sources, the deterioration of their livelihoods and the resulting growing need for food aid. This is because the forest has been a source of traditional food items for the Gumuz, including plant shoots, roots, leaves and fruits. Wild forest foods have not only been consumed in times of crop failure or food shortages, but also as part of the group’s regular diet. Some women interviewed explained that, before the arrival of the investors and the resulting deforestation and enclosures, wild edible leaves such as kaakima and roots like echa and cici could easily be collected close to the village (interview, Qotta kebele, May 2012).⁴⁶ Women are now required to travel far from the villages to gather these forest foods. They also complained that some area-specific forest foods have been disappearing, or could no longer be accessed as a result of enclosures. Some wild animals have similarly disappeared. As one informant in Dangur stated:

We were able to hunt many wild animals at a short distance [from the village]. But after these investors invaded us, the animals started to disappear and move to the remaining remote forested areas. I tell you; we will not see a single Guanja here by next year. (Interview, Qotta kebele, May 2012)⁴⁷

In interviews both experts and woreda, zonal and regional authorities unanimously confirmed the significant and rapidly increasing trend of environmental destruction caused by land investments (interviews, April–June 2012). The informants also emphasised that the lack of an appropriate regional land use plan – which would be the basis for land use and management decisions – has contributed to the environmental destruction. A widespread perception of land availability throughout the region has tended to undermine efforts towards efficient utilisation of land and natural resources. This notion has led to a land allocation process that has been based neither on a classification of the existing land uses, nor on a genuine participation of local communities. Thus, very little attention has been paid to protecting settlements, existing cultivated lands, or forest and protected areas that sustain environmental services and local livelihoods. The focus has instead been laid on attracting as many investors as possible to the region.

The natural resource management experts interviewed at the woreda, as well as regional offices indicated that vast land areas have been cleared under the ongoing indiscriminate act of deforestation. These areas are covered by indigenous forest species with economic value – including lowland bamboo, incense trees, gum trees and Zobi (interviews, April–June 2012). For example, the lowland bamboo tree – which is widely available in the region – is traditionally used by local people for a wide range of purposes, including the construction of houses, fences, cattle barns, baskets, furniture (stools, chairs, etc), granaries, tools, traditional beehives, traditional musical instruments, and for firewood.⁴⁸ Additionally, its sprouts serve as an important source of food. Finally, lowland bamboo trees have been shown to feature a high carbon sequestration capacity, providing critical ecological services. Despite its distinct importance, extensive areas of bamboo-covered land have been allocated to investors, something which has led to these areas’ deforestation. Through the federal government 3000 hectares of land with thick bamboo vegetation in the Guba woreda were recently allocated to the domestic investor Kehedem Trading. The land is located in the Ayicid kebele, on the left side of the main road between Gublak and Mankush, and bordered by the Beles River. According to informants, the area was designated as protected forestland because of its rich bamboo vegetation and wildlife biodiversity. The experts interviewed further emphasised that the land was not actually suitable for commercial agriculture, and that the project’s feasibility was deeply questionable because of the nature of the landscape. For these reasons the land transfer was contested by the local inhabitants – both experts and local authorities. Unfortunately, these efforts were unable to save the contested forestland from appropriation, although the investor had not yet started clearing the bamboo vegetation at the time of the fieldwork. Another example – widely rumoured among both experts and woreda and zonal authorities – of the transfer of an area known for its rich woodland, water and wildlife resources referred to large tracts of land in the Dangur and Guba woredas. This area is bordered by Alatish Park in the Quara woreda of the Amhara regional state, the boundary being marked by the Ayima River.

As most of the land allocations have been carried out without appropriate feasibility studies, many forested areas not suitable for cultivation – particularly via mechanised farming – have been transferred to investors. As a result, some of the investors who acquired the unsuitable lands ceased cultivation after clearing large areas covered by natural forest. For example, all along the road between Mankush and Almahal in the Guba woreda, six to eight investors abandoned the land they had been allotted after clearing away its thick forest cover and cultivating it for about one agricultural season. Strikingly these investors have already acquired replacement lands in other kebeles within the same woreda. In the Guba woreda areas covered by naturally grown incense and gum trees have also been allocated to investment projects, disregarding their economic value to the local communities and the country.⁴⁹ Such uninformed land allocation practices, which indiscriminately allocate large areas of forestlands to investors, are more likely to have adverse impacts on the natural environment – and thus on local livelihoods. The clear negligence documented so far with regard to protecting forested areas is particularly distressing, being also a manifestation of the hegemonic character of the current land allocations.

The ways in which the allocated land has been used and managed by investors also raise sustainability concerns. According to the experts I spoke to, most of the investors have been utilising their land without preparing sound and efficient land-use plans to ensure that the land is worked in an environmentally sustainable way (interviews, April–June 2012). Some of the elements that need to be carefully considered while developing the land include distance from water bodies and gullies, number of trees left per hectare and the slope of the land. However, these criteria have rarely been met. For example, after its field visits to the area the Ethiopian Parliament’s standing committee on environmental matters recently criticised the way in which the domestic investor, Balzaf Alcohol and Liquor Plc, has been utilising its leased land in the Homosha woreda. Moreover, some investors have been blamed for failing to use the land for its intended objectives. Informants in the Dangur, Guba, and Homosha woredas indicated that some investors have been engaged in the production of timber and charcoal, even though such activities are illegal. In this regard a regional official mentioned that, at times, some investors have asked permissions to transport the charcoal they have produced to sell it in nearby towns (interview, Assosa, April 2012). A representative of one of the NGOs working in the region also confirmed that some investors were producing charcoal, even as the local inhabitants were being told to refrain from such activities.

Conclusions

In Ethiopia the leasing of land to investors for export-oriented agricultural production – especially in lowland regions – has been enacted on a grand scale over the past few years. As this study has shown, land acquisitions have occurred on a significant scale throughout the Benishangul-Gumuz region, particularly in the Dangur and Guba woredas. These acquisitions took place while subverting the land rights and natural resource-based livelihoods of the indigenous local communities. This study has uncovered that indigenous local communities face the threat of losing their land and their natural resource-based livelihood. The ongoing land allocation process – largely predicated on the state’s and other elite groups’ perception of abundant ‘underutilised’ or ‘unoccupied’ land in the region – overlooks the local communities’ traditional land-use practices and social relations. Under this misguided notion, the land from which traditional cultivators derive their livelihoods is being treated as ‘underutilised’ or ‘unoccupied’, and then transferred to investors for permanent forms of land use. That said, there are inherent differences between various actors, who assign different meanings and values to existing local land uses. Thus, the concern should not so much focus on the claim that abundant land resources exist, but rather on the ways through which the ‘available’ land has been identified, delineated and transferred. For instance, despite the substantial land transfers, genuine community consultation and participation have not been part of the land acquisition process. Instead, the state’s hegemonic representations prevailed, as expressed through the power exercised in favour of commercial land investments through the manipulation of existing institutional and administrative frameworks.

As a result, local communities have not only been dispossessed of their cultivated lands, but have also been displaced from their homes, often under the guise of villagisation. Most significantly, forestlands have haphazardly been allocated for commercial agricultural development, even when providing ecological services and traditional sources of livelihoods. Additionally, it seems that those lands already transferred have not been utilised in an environmentally sustainable manner, as the land is generally being used without the necessary land-use plans. Undoubtedly such acquisitions will generate gains for some local economic and political elites, as well as for the state; the region’s indigenous local communities, however, have noted little gain and possibly great loss.

What has now become more apparent is that the impact of land acquisitions is seldom granted much attention, relying instead on dewy-eyed optimism. The government has already started reconsidering some of the already enacted land deals – as they have, so far, failed to deliver on their commitments. Two major measures are needed, however. First, the government should fundamentally rethink the policies that promote the development of large-scale mechanised monoculture. Second, it should carefully scrutinise large land investments, in order to bring local concerns aboard before the current trend, in which the short- and long-term impacts have not yet been critically considered, leads to disastrous long-term effects for the local communities and the natural environment.

As a whole, this study complements recent research that highlights the important roles played by host states in driving contemporary large-scale land acquisitions. In the context of many African countries, in which different authorities overlap and compete over control of land resources, the recent land acquisitions may provoke contestations and changes in power relationships within and between state and society that have important implications for the land rights and livelihoods of indigenous local communities.

Notes on Contributor

Tsegaye Moreda is a PhD candidate at the International Institute of Social Studies (ISS) of Erasmus University Rotterdam. His research interests focus on state–society relations in the context of contemporary development issues in the Eastern Africa region, with a particular focus on land politics, property rights, conflict, migration and resettlement, environment and development, rural youth and farming, agrarian change and development policy, land grabbing and political reactions ‘from below’.

Disclosure statement

No potential conflict of interest was reported by the author.

Acknowledgements

The author is grateful to Jun Borras, Max Spoor and three anonymous reviewers for their helpful comments and suggestions, and to the Netherlands Fellowship Programme (NFP), Land Deal Politics Initiative (LDPI) and Heinrich Boll Foundation (Ecofair Tade Project) for providing financial support to conduct the fieldwork.

Notes

1. Borras et al., “The Politics of Biofuels”; Peluso and Lund, “New Frontiers”; White et al., “The New Enclosures”; Fairhead et al., “Green Grabbing”; Borras et al., “Land Grabbing”; and Edelman et al., “Global Land Grabs.”

2. Borras et al., “The Politics of Agrofuels”; McMichael, “Agrofuels in the Food Regime”; HLPE, Land Tenure; White and Dasgupta, “Agrofuels Capitalism”; Zoomers, “Globalisation and the Foreignisation of Space”; De Schutter, “How not to Think of Land-grabbing”; Visser and Spoor, “Land Grabbing in post-Soviet Eurasia”; Hall, “Land Grabbing in Southern Africa”; and Borras and Franco, “Global Land Grabbing.”

3. Borras and Franco, “Global Land Grabbing.”

4. Alden Wily, “Looking Back,” 752; and Peters, “Land Appropriation.”

5. Salami et al., Smallholder Agriculture in East Africa; Von Braun and Meinzen-Dick, Land Grabbing by Foreign Investors; Vermeulen and Cotula, “Over the Heads of Local People”; De Schutter, “How not to think of Land-grabbing”; and Woodhouse, “New Investment, Old Challenges.”

6. Akram-Lodhi, “(Re)imagining Agrarian Relations?”; Kay, “Development Strategies and Rural Development”; McMichael, “Peasants make their Own History”; and Veltmeyer, “The Dynamics of Land Occupation.”

7. McMichael, “Peasants make their Own History,” 213. See also Veltmeyer, “The Dynamics of Land Occupation”; and Kay, “Latin America’s Agrarian Transformation.”

8. World Bank, Rising Global Interest in Farmland; and Deininger, “Challenges Posed.”

9. Borras and Franco, “From Threat to Opportunity?”; De Schutter, “How not to think of Land-grabbing”; and Li, “Centering Labor.”

10. FAO, Voluntary Guidelines; Borras et al., “The Challenge of Global Land Grabbing”; and Franco et al., “Democratic Land Control.”

11. Peluso and Lund, “New Frontiers”; Wolford et al., “Governing Global Land Deals”; Burnod et al., “Competition over Authority”; and Peters, “Land Appropriation.”

12. Moore, “Contesting Terrain”; and Watts, “The Agrarian Question in Africa.”

13. Watts, “The Agrarian Question in Africa,” 4.

14. Wolford et al., “Governing Global Land Deals,” 189.

15. Wolford et al., “Governing Global Land Deals,” 191.

16. Markakis, Ethiopia; Rahmato, “Land to Investors”; Lavers, “Patterns of Agrarian Transformation”; and Makki, “Development by Dispossession.”

17. HLPE, Land Tenure; Borras et al., “The Politics of Agrofuels”; Rahmato, “Land to Investors”; Lavers, “‘Land Grab’ as Development Strategy?”; Lavers, “Patterns of Agrarian Transformation”; and Shete and Rutten, “Impacts of Large-scale Farming.”

18. See, for example, Von Braun and Meinzen-Dick, Land Grabbing by Foreign Investors; Vermeulen and Cotula, “Over the Heads of Local People”; and Markakis, Ethiopia.

19. There are five tiers of government administration in the country, which include (from the highest to the lowest administrative unit): federal, regional, zone, woreda and kebele. Woreda is roughly equivalent to a district, while kebele, especially in rural areas, corresponds to a group of villages.

20. BGRS, Benishangul-Gumuz Regional State Food Security Strategy.

21. CSA, Summary and Statistical Report.

22. See Moreda, Postponed Local Concerns?, for details.

23. For the intensive fieldwork two woredas from Metekel zone were selected, namely Dangur and Guba. In addition, a short visit was made to the Homosha woreda in the Assosa zone, in order to strengthen the findings. These woredas, particularly Dangur and Guba, were selected because they are the main foci of recent large-scale land acquisitions in the region. Within these woredas some villages were carefully selected based on investment concentration and expert opinion, particularly regarding accessibility and representativeness. Through this process three kebeles from the Dangur woreda (Gimtiya, Dachigeri and Qotta) and two kebeles from the Guba woreda (Ayicid and Mankush) were selected. In addition, the Berta ethnic group – which dominates the Tsori-al-metema kebele in the Homosha woreda –was included in the study. As part of the research strategy of direct observation a brief field trip was taken between the towns of Mankush and Almehal in order to capture some important aspects that had not been fully represented in the selected kebeles.

24. The Gumuz ethnic group largely inhabit the Metekel and Kemashi administrative zones, whereas the Berta ethnic groups inhabit the Assosa administrative zone. These two groups respectively constitute the most numerous ethnic groups of the two zones (see Figure 1).

25. FDRE, “Re-enactment’; and FDRE, “Investment Incentives.”

26. FDRE, “Council of Ministers Regulation.”

27. MoFED, Growth and Transformation Plan, 48–49.

28. Deininger and Byerlee, Rising Global Interest.

29. Oakland Institute, Understanding Land Investment Deals, 18.

30. Rahmato, “Land to Investors,” 37.

31. MoFED, Annual Progress Report for 2012/13, 38.

32. Ibid.

33. MoFED, Annual Progress Report for 2011/12; and MoFED, Annual Progress Report for 2012/13.

34. It is estimated that nearly 55% of smallholders cultivate holdings measuring one hectare or less. MoARD, Ethiopia’s Agricultural Sector Policy.

35. Ibid., 3.

36. In terms of irrigation potential, the country has roughly 4.3 million hectares of irrigable land. Nevertheless, only an estimated 6% of this potentially irrigable land is currently being utilised. MoARD, Ethiopia’s Agricultural Sector Policy, 4.

37. MoARD, Agricultural Investment Potential, 3.

38. Markakis, Ethiopia; Rahmato, “Land to Investors”; Lavers, “Patterns of Agrarian Transformation”; and Makki, “Power and Property.”

39. MoFED, Growth and Transformation Plan; and Rahmato, “Land to Investors.”

40. Rahmato, “Land to Investors”; Fisseha, Commercial Pressures; Lavers, “Patterns of Agrarian Transformation”; Moreda, Postponed Local Concerns?; Shete and Rutten, “Impacts of Large-scale Farming”; Moreda and Spoor, “The Politics”; and Moreda, “Listening to their Silence?”

41. Jayne et al., “Smallholder Income and Land Distribution.”

42. Abbute, “Gumuz and Highland Resettlers”; and Gebre, “Resettlement.”

43. Abbute, “Gumuz and Highland Resettlers”; and Gebre, “Resettlement.”

44. BGRS, Benishangul-Gumuz Regional State Villagization Plan.

45. Unlike in other regions – such as Amhara region – land registration and certification in the Benishangul-Gumuz regional state have not been undertaken. Thus traditional land-tenure systems are still widely practised. That said, the difference in the de facto tenure system between highland and lowland regions (including the Benishangul-Gumuz region) cannot be attributed to the absence of land registration in the latter, as that is a relatively recent phenomenon in both areas. The Derg’s 1975 land reform abolished land tenure differences across the country by only distributing usufruct rights; despite this, tenure in the highlands has long been based on individual holdings, while customary forms of tenure have largely been practised in the lowland regions.

46. Among the Gumuz women perform numerous livelihood activities, more so than the men of the group. In addition to fully participating in farming activities, they also shoulder the key responsibility of feeding the family on a daily basis.

47. Guanja is the local name of one of the animals commonly hunted by the Gumuz for food. Its equivalent name in Amharic is Midaqua and in English, Duiker.

48. The steam cover of bamboo is used in roofing and the construction of traditional beehives. The hollow bamboo is also used for making traditional musical instruments (widely used among the indigenous local communities). Cups used for drinking coffee and as water jars are made out of the bamboo’s stem (culms), which is cut into small pieces.

49. Despite this, about 37,000 hectares covered by natural plantations of incense and gum trees in Guba woreda have been granted to three domestic investors that will harvest the plantations for domestic and export markets. The companies are: Ethiopian Gums and Incense Enterprise (12,000 ha), Beles Plc (15,000 ha) and Meskerem Natural Gum and Incense Enterprise (10,000 ha). In general several woredas in the region – e.g. Guba, Kurmuk, Sherkole and Sirba Abay – are endowed with naturally growing incense and gum trees.