Land for agricultural development in the era of ‘land grabbing’

ethiopia
barren-lands
spatial-exploration
land-bank
land-grab

#1

Land for agricultural development in the era of ‘land grabbing’
A spatial exploration of the ‘marginal lands’ narrative in contemporary Ethiopia
By: Rachel A. Nalepa

Abstract

In response to concerns over the potential of land leases for new agricultural projects to displace
rural populations and impact food security, the Ethiopian government asserts that only ‘marginal’,
‘barren’ or ‘wasteland’ is being leased to investors. This paper draws on a wide variety of source
material in order to untangle the land classification of ‘marginal’ land as it is used colloquially across
Ethiopian institutional and policy environments and compares this promoted understanding of
marginality to the socio-cultural and biophysical characteristics of actual land areas either already
transferred to investors or currently deposited in the federal ‘land bank’ to be allocated at a future
time. This analysis reveals that ‘marginal’ lands are not unused and/or degraded as often implied but
are potentially productive lands that overlap national park boundaries or are currently supporting
nomadic and semi-nomadic livelihoods. In addition, this paper contends that marginal lands are not
categorized according to any shared criteria, but applied to the lands in weaker regions that are not
being put to highest value use according to the state’s market-oriented developmental strategy.

Introduction

Ethiopia is one of the many developing countries receiving an increased amount of attention in the
ongoing conversation of the ‘global land rush’ or ‘global land grab’. Both domestic and foreign
investors continue to seek leases for new, large-scale agricultural projects on Ethiopia’s perceived
abundant and available fertile lands. The Ethiopian state, represented by the Ethiopian People’s
Revolutionary Democratic Front (EPRDF), legally owns all the land in the country and is complicit:
leases for new projects aim to both achieve food security through securing foreign exchange earnings
to import food with a growing agricultural export economy and also facilitate technology and
knowledge transfer to current smallholders in order to boost domestic yields and modernize the
sector. Land leasing is a politically charged topic given that land is a highly precious asset in Ethiopia;
smallholders comprise 85% of total employment and due to population growth and pervasive land
degradation in many areas, land holdings are already less than 2 ha for nearly 90 percent of rural
households (Gebreselassie 2012, Tamrat 2010). In response to this, the government has asserted
that the only land appropriate to be allocated to investors is land interchangeably characterized as
‘marginal’, ‘unused’ or ‘wasteland’. This response is particularly stressed in the case of biofuel
projects given the states’ sensitivity to the larger ethical and political tensions surrounding the
appropriation or diversion of fertile land for fuel despite a chronically food insecure population (FAO
2010a).

Even though land is state property and these leases executed under legal auspice, the government
has come under scrutiny as field evidence from ethnographically grounded studies have revealed
that some of these allocated ‘marginal’ lands were already being used for a wide variety of purposes
(see Horne 2012; Human Rights Watch 2012; Fisseha 2011; Demeke and Akilu 2008). Recent studies
have also elaborated on the significance of the government’s use of land transfers in the attempted
re-appropriation of Ethiopia’s ‘unused’ lowlands from their current use by pastoralists and shifting
agriculturalists to settled capitalist farmers dedicated to export crops (Lavers 2012, Makki 2012). This
collection of work suggests that ‘marginal’ is a convenient term not necessarily applied to capture
the physical characteristics of the land itself or imply that it is absolutely unfit for food production (as
is argued in the case of biofuels) but rather that it is not purposed to the highest value use according
to the EPRDF’s development plan. This has implications for both ethnic self-determination (a key
tenet of Ethiopia’s federal system) and for the fate of the lowland subsistence sector as the
government creates an archipelago of capital enclosures for large-scale mechanized farming that
‘maps onto older imperial socio-spatial and cultural hierarchies’ (Makki 2012: 81, Lavers 2012).
This paper provides supportive physical evidence for these claims and further investigates the
suggested (yet still unclear) link between ‘marginal’ land and ‘marginal’ populations in Ethiopia. The
presumption is that the organization, labeling and representation of the material and social
environment is an active, value-laden process and a critical, systematic consideration of these
classifications has the potential to reveal more about the intentions of the state when it comes to
land management and development policy.

This paper also makes a contribution to the broader literature by stressing the importance of
integrating spatial information into studies on transferred lands. Investigations into the impacts of
‘land grab’ on domestic political economy often center on the quantity of land changing hands, even
though land quality is just as important while considering the potential impacts of land deals on
ecosystems and rural livelihoods. Investor interest is often directed to the best land in terms of water
availability and irrigation potential, soil fertility, proximity to markets or availability of infrastructure
(Cotula 2012) but without specific spatial information on these deals, we are not able to project this
assumption beyond specific cases. Spatial information helps us to link transferred lands to the
physical and socio-cultural information that characterizes them and to determine how they are
currently being used and by whom. Spatial specificity also helps us move beyond the limitations posed by the comparison of singular case studies in order to elucidate and empirically extrapolate
broader trends through space and time. By situating land transfers and additional land marked to be
allocated in the future, we are able to establish baselines from which to observe how increased
investor interest may alter agrarian landscapes using geo-spatial technologies and tools such as
remote sensing imagery and Geographic Information Systems (GIS).

This paper draws on a wide variety of source material in order to untangle the land classification of
‘marginal’ land as it is used colloquially across Ethiopian institutional and policy environments and
compares this promoted understanding of marginality to the socio-cultural and biophysical
characteristics of actual land areas either already transferred to investors or currently deposited in
the federal ‘land bank’ to be allocated at a future time. Data was gathered between May 2011 and
January 2013 and includes semi-structured interviews with government officials and employees, land
management professionals, and academics, as well as unique datasets and spatial metadata
gathered from government, open access and NGO sources. The first part of this paper discusses the
evolving role of the central government in the land allocation process as well as how leased land or
land intended for investors is being described colloquially. The remainder of this paper explores
various narratives of marginality such as: ‘marginal’ as undervalued land user regimes, ‘marginal’ as
unsuitable for food crops due to biophysical or socioeconomic constraints, and ‘marginal’ as
degraded land.

To Read Complete Report Follow: LDPI_WP_40.pdf (2.7 MB)