Patterns of agrarian transformation in Ethiopia: State-mediated commercialisation and the ‘land grab’

Abstract

Much of the literature on the ‘land grab’ has thus far focused on the international drivers of foreign agricultural investment, with far less attention paid to the roles of developing country states and domestic political economy in changing forms of agrarian production. This paper analyses how global and domestic processes combine to produce patterns of agrarian transformation in Ethiopia, one of the main targets of foreign agricultural investment. The paper presents a typology of changes in land use and examines in detail three case studies of investments in Ethiopia drawn from this typology. The paper concludes that the most dramatic changes are taking place in lowland, peripheral regions where large-scale, capital-intensive farms employing wage labour pose a serious risk to pastoralists whose ‘use’ of land is contested by the state. Although the government has been careful to avoid mass displacement of settled smallholders, there are also important changes taking place in highland areas, with the government encouraging investments that combine the resources of investors with the labour and land of smallholders. These investments have resulted in exposure to new forms of market risk.

Keywords:

• Ethiopia
• land grab
• agrarian change
• investment
• displacement

Notes

¹Below the federal government, the hierarchy of government administrations is as follows (from largest to smallest): region, zone, wereda and kebele.

²US$1 was approximately 16.5 birr in September 2010.

³I have been unable to trace the origin of these estimates, which do not refer to specific sources. Similar, claims of vast areas of fertile ‘virgin’ land in the lowlands were used by the Derg to justify resettlement programmes (Pankhurst 1992) and probably contributed to the construction of ‘unused’ land. They also differ considerably from those provided by the World Bank in their recent report on foreign agricultural investment, which suggest that less than 5 million hectares of currently uncultivated land are suitable for cultivation (World Bank 2010, 110).

⁴Respondent C questioned the categorisation of any land as communal land: ‘communal land means it belongs to the community, land in Ethiopia is government land’.

⁵According to the contract, the chairman and deputy chairman should have been granted power of attorney by the kebele members, though this does not appear to have been the case according to respondents O.

⁶This number is the total of (mostly male) landholders. A conservative estimate of family size would suggest that at least 500,000 people were involved in the project.

⁷An additional concern, which is not addressed in the present study, regards the environmental impact of agricultural investment projects. Many of the investments are based on the intensive use of chemical fertilisers and pesticides, which, if not carefully managed, could pollute the local environment. Indeed, the limited research that exists on this topic in Ethiopia suggests that there are significant weaknesses in government environmental legislation and monitoring of investment projects (Getu 2010, The Oakland Institute 2011). Meanwhile studies have concluded that health and safety guidelines relating to the use of pesticides are frequently not adhered to in many flower farms, posing health risks for farmworkers (Haddis 2007), and contributing to pollution of water bodies used for drinking water by local communities and their livestock (Tadele 2009).

⁸This conclusion is supported by past research which suggests that wealthier farmers have been better placed to take advantage of green revolution technologies (Harriss 1987, Bernstein 2010).

Additional information

Notes on contributors

Tom Lavers

This research was carried out as part of a PhD supported by a studentship from the Economic and Social Research Council (ESRC), and this paper was written with financial support from the Land Deal Politics Initiative (LPDI). The support of both organisations is gratefully acknowledged. In addition, I would like to thank Addis Ababa University for facilitating my fieldwork in Ethiopia; Amsalu Darge and Tolasa Mamuye for their interpretation skills during the case studies; and Shea McClanahan, Ruth Hall and two anonymous reviewers for their comments on a draft of this paper.