Preparing For The Future? Understanding The Influence of Development Interventions On Adaptive Capacity at Local Level in Ethiopia
Preparing For The Future? Understanding The Influence of Development Interventions On Adaptive Capacity at Local Level in Ethiopia
Preparing For The Future? Understanding The Influence of Development Interventions On Adaptive Capacity at Local Level in Ethiopia
Eva Ludi, Overseas Development Institute; Million Getnet, Haramaya University; Kirsty Wilson, ACCRA/Oxfam GB;
Kindie Tesfaye, Haramaya University; Beneberu Shimelis, Haramaya University;
Simon Levine, Overseas Development Institute and Lindsey Jones,Overseas Development Institute
Cover picture: Densa, an active and innovative woman farmer in Wokin kebele and member of the Oxfam GB/ORDA malt barley marketing project.
ACCRA in Ethiopia is a research and capacity building consortium of Oxfam GB,
Overseas Development Institute (ODI), Care International, Save the Children UK,
the Government of Ethiopia Disaster Risk Management and Food Security Sector and Haramaya University.
It works in Uganda and Mozambique where World Vision International is also a consortium member.
Contents
Acknowledgments 3
Abbreviations 3
Executive summary 4
The authors would like to acknowledge the contribution of the communities in Ander Kello, Kase-hija and Wokin kebeles
as well as staff from Chifra, Gemechis and Dabat wereda bureaus and Care, Oxfam, ORDA and Save the Children UK. The
valuable contributions from Federal and Regional officials who gave their inputs via the validation exercises conducted
in each site are also greatly appreciated and staff of the Federal Environmental Protection Agency and Disaster Risk
Management and Food Security Sector within the Ministry of Agriculture gave helpful feedback which was used to
enhance the quality of this report. Valuable comments were also received from Josephine Lofthouse, Lindsey Jones,
Catherine Pettengell, and ACCRA steering group members. Section 1 was written by Lindsey Jones.
Abbreviations
ACCRA Africa Climate Change Resilience Alliance
AMJ April May June
ANRS Afar National Regional State
asl Above Sea Level
CAHWs Community Animal Health Workers
DFiD UKs Department for International Development
DJF December, January, February
DRR Disaster Risk Reduction
ENSO El Nio Southern Oscillation
GCM Global Climate Model
GDP Gross Domestic Product
HAB Household Asset Building Programme
HIBRET Household Asset Building and Rural Empowerment for Transformation (CARE intervention)
ITCZ Inter-Tropical Convergence Zone
JAS July August September
KA Kebele Association
LAC Local Adaptive Capacity Framework
MDG Millennium Development Goal
NGO Non Governmental Organisation
NMA National Meteorological Agency
ORDA Organisation for Rehabilitation and Development in Amhara
PCDP Pastoral Community Development Programme
PILLAR Preparedness Improves Livelihood Resilience (Save the Children UK intervention)
PSNP Productive Safety Net Programme
SCUK Save the Children UK
SMC Sorghum Maize and Chat Livelihood Zone
SNNPR Southern Nations, Nationalities and Peoples Region
SRES Intergovernmental panel on climate change Special Report on Emissions Scenarios
VSLA Village Saving and Lending Association
4 PREPARING FOR THE FUTURE IN ETHIOPIA
Executive Summary
In the old days, there was plenty of grass in this area. We simply released our
animals into the grass which was more than 3m tall Now we dont move our
animals the way we used to because there are far fewer animals. Some buy feed
for their animals, some have enclosures and Government support has increased.
Compared to the past, we have peace and security, and access to school and health
services. But we are getting poorer because of environmental changes and because
we have fewer animals. AK, 68, Ander Kello
Ethiopia is currently ranked 11th of 233 countries and other political jurisdictions in terms of its vulnerability
to physical climate impacts, and 9th in terms of overall vulnerability, which is physical impacts adjusted
for coping ability (CGD, 2011). Yet little is known about its peoples adaptive capacity at individual and
community level, or how existing interventions influence a communitys ability to adapt.
Recognising the complex relationship between climate and development, research conducted by the Africa
Climate Change Resilience Alliance (ACCRA) seeks to explore how development interventions impact on
adaptive capacity at the local level in Ethiopia, Uganda and Mozambique. It does so using the Local Adaptive
Capacity framework (LAC), depicting adaptive capacity as composed of five interrelated characteristics,
namely: the assets base; knowledge and information; institutions and entitlement; innovation; and flexible
forward-looking governance. Primary and secondary data was gathered from three research sites in three
Ethiopian Regions, namely Ander Kello in Afar Region, Kase-hija in Oromia Region and Wokin in Amhara
Region. This report is a synthesis of the key findings.
ACCRAs research finds that although interventions by governments and development partners are
impacting, and in some cases contributing positively to, the characteristics of adaptive capacity, they often
fall short of their full potential to enhance the capacity of households and local communities to adapt by not
appreciating and maximising their contributions across all five characteristics of adaptive capacity. The LAC
framework shows that the analysis of poverty and vulnerability and resulting development interventions
typically focus on only one or two of the five dimensions, principally broadening the asset base and to a
lesser extent institutional arrangements. This often ignores underlying institutional barriers that prevent
some households from accessing those assets. Institutional barriers and power structures that increase the
vulnerability of some households are insufficiently analysed and understood, and therefore interventions
fail to contribute to improved livelihoods for some households; interventions are often carried out in
isolation, different actors do not consult each other sufficiently, which leads to duplication and inefficiency,
and different actors are not learning sufficiently from experiences of others.
The research concludes that, by using the LAC framework, more focused interventions could be developed
that target both immediate development needs and longer-term adaptation requirements. Interventions can
combine different approaches disaster management, social protection and livelihoods promotion all of
which are necessary. This will only become more important given anticipated climatic and other changes.
We will continue helping each other until the end. If I have only a cow which is
pregnant and my fellow brother comes for help, I will promise to give him the calf
when it gets born. We know we are getting poor together, but we also know we
wont starve to death while our clan members are having something to eat. We
survived the past horrible years because of this culture and we will continue to do
so. Man in Ander Kello
In a changing world, sustainability means being able to adapt continuously and forever. Sustainable
development cannot be achieved by bringing people to a given state, but only by giving them the
PREPARING FOR THE FUTURE IN ETHIOPIA 5
ability to adapt to future change autonomously. Enhancing agency must therefore be at the heart of
any development intervention and particularly of interventions aimed at strengthening adaptation and
adaptive capacity. Development interventions need to focus on the process that is required for enhancing
agency, and avoid undermining peoples and communities own initiatives.
Development partners need to reassess the scale and scope of their interventions
The rain is in the hands of God, but the road is in the hands of the Government. We
cant do anything about the rain, but the Government could help us by building us
road and bringing us telecommunication service. Farmer, Kase-hija
The justification for working primarily at the village level is based on the desire to achieve greatest impact
given the constraints of staffing and finance. However, much of the vulnerability at local level is determined
or affected by structures and processes far beyond the local level. Without addressing the stressors emanating
from higher levels, efforts to improve the asset base and the institutional set-up locally are likely to fail.
Current project interventions are maintaining existing systems, but are not contributing enough to building
adaptive capacity. Part of the reason for this limited focus is that most development efforts concentrate on
outputs (not outcomes), and are dominated by the provision of hardware, while insufficient attention is given
to the dynamic processes of change such as the use of information in decision-making, the development of
equitable institutions, the fostering of local innovation and improved accountability in governance structures.
Development partners need to better use information and knowledge for evidence-based decision
making and project design
I am the first person who tried application of fertiliser on a farm land in Finote
Selam. I am also the first to have good number of beehives, poultry and a eucalyptus
plantation. I have never received food aid, rather I receive advice and information.
Better-off farmer, Dabat
The ACCRA research found that projects are not being designed using the best possible information,
prediction and scenario analysis. This is true both of longer-term shifts (e.g. climate variability and
change) and short-term changes (e.g. specific seasonal weather forecasts). More than once findings were
ambivalent on whether or not an introduced technology is leading to better adaptation to future conditions
or to maladaptation. There is a need for more sophisticated analysis of current vulnerabilities and what
causes and maintains them. This should be expanded to include constraints to adaptive capacity.
Building adaptive capacity requires a continuous process of learning, change and innovation
It was like a wind which blows fast, as a shadow disappearing suddenly and a girl
walking away from my face. We have seen it going away before we realize how
important the technology was Farmer talking about water harvesting technology
in Kase-hija
In order to prepare households and communities better for dealing with the projected impacts of greater climatic
variability and extremes in future, innovation and strengthening innovative capacity must take centre stage
in any development initiative. Innovation is understood as a process of experimentation and exploration of
practices, techniques or new organisational forms. The ACCRA research suggests that local residents have made
a number of innovations, but these have not been replicated, in part because the idea that poor and poorly
educated farmers and pastoralists might be capable of experimenting and learning for themselves was not
found to be well ingrained among development actors. For innovation, a set of requirements was found to be
necessary and includes: An awareness that the current situation needs to change, a sense of being in a position
to change that situation, access to appropriate information about different options that can be used to solve
particular problems, access to resources to invest in testing new things or to act as a safety net in case of failure,
and an enabling policy and institutional environments which encourage and promote innovation. Development
partners need to be better aware of what role they have to play with regards to supporting these requirements.
Section 1:
Exploring the
characteristics
of adaptive
capacity
PREPARING FOR THE FUTURE IN ETHIOPIA 7
This paper syntheses key findings from the Africa Climate Change Resilience Alliances (ACCRA) research
in Ethiopia. Section one explores key concepts and provides background to ACCRA. It goes on to explain
the research methodology and introduces the projects analytical tool, the Local Adaptive Capacity
Framework (LAC). Section two highlights key climate and development challenges for Ethiopia as well
as describing some of the main development interventions in each of three research sites. Section three
draws on research findings to explore how development interventions are impacting on the characteristics
of adaptive capacity. Finally, section four provides an insight into what ACCRAs research means for
development practice.
Discussions within the ACCRA consortium, with the DfID-funded Strengthening Climate Resilience
consortium and experts in the fields of DDR, Livelihoods, Social Protection and Climate Change Adaptation
approaches associated with the study led to two clear challenges. First, that a narrow focus on humanitarian
interventions and a narrow definition of resilience the ability to bounce back after a shock doesnt
deliver what is required by communities faced with climate change.
There are many other development challenges and uncertainties beyond those that are weather-related:
both communities and systems need to learn how to adapt to these uncertainties and changes. However,
the term resilience is hotly debated and many people have taken to using a broader definition, with widely
differing interpretations of what it means. Secondly, one of the biggest challenges within development
programming is how to ensure that individuals and societies can adapt beyond the programme-cycle of
an intervention. This is crucial to climate change adaptation because there is no end-point to adapt to:
people need to acquire the capacity to adapt for generations to come. The challenge to development
practice is how to meet immediate needs while also building the capacity to adapt in the future. The ACCRA
consortium decided, in agreement with DFID, to reflect this need for long term flexibility and focus instead
on the capacity to adapt. ACCRAs research focused on providing insights into the requirements that allow
people and communities to build their own adaptive capacity with a view to supporting this additional
challenge to development practice.
Adaptive capacity refers to the potential of individuals and societies to respond to change; as such, it is not
possible directly to measure adaptive capacity. Instead, ACCRAs research investigated the characteristics
that are considered to contribute to the adaptive capacity of a system in a particular context. These are
the five characteristics that make up the LAC framework (see Section 1.3 for a detailed description) that
was used to frame ACCRAs research. We investigated the impact of development interventions (DRR,
Social Protection, and Livelihoods) on peoples and communities adaptive capacity in order to: understand
how different programming approaches either build or undermine adaptive capacity; understand how
programmes that do not specifically target the impacts of climate variability and change can still improve
8 PREPARING FOR THE FUTURE IN ETHIOPIA
peoples capacity to adapt in the future; and to learn how to improve interventions in all programme
approaches towards building adaptive capacity. The conclusions are intended to support governments and
development actors design and implementation of both humanitarian and development interventions,
and policies that increase poor and vulnerable communities resilience and their wider adaptive capacity.
Communities that are able to anticipate, deal with and respond quickly to climate change are considered
to have high adaptive capacity (Smit and Wandel, 2006). Broadly speaking, adaptive capacity relates to
the capacity of a person or community to respond and adapt to the likely impact of changing shocks and
stresses (Lim and Spanger-Siegfried 2004). More specifically, in the context of climate change, it denotes
the ability of a system to adjust, modify or change its characteristics or actions to moderate potential
damage, take advantage of opportunities or cope with the consequences of a changing climate (IPCC,
2007; Brooks, 2003).
Crucially, adaptive capacity refers to the potential to adapt, as and when needed, and not necessarily the
act of adapting, or its outcome. Adaptive capacity is multi-dimensional and the elements that make up
an individuals adaptive capacity are not entirely agreed. It essentially relates to whether people have
the right tools and the necessary enabling environment to allow them to adapt successfully over the long
term. Also important to bear in mind is that adaptive capacity is context-specific and varies from country
to country, community to community, between social groups and individuals, and over time (Smit and
Wandel, 2006). It is the combination of development choices, adaptation actions and local capacities that
allows for effective action at the local level (Kuriakose et al., 2009).
The concept of adaptation is relatively easy to describe in principle, but hard to depict in practice and in
detail. In the context of this study, adaptation refers to the process by which communities reduce the
adverse effects of shocks and stresses, including climate change, on their livelihoods and well-being, and
take advantage of new opportunities provided by a changing environment (TERI, 2007). More specifically,
it can be described as adjustments in natural or human systems in response to actual or expected climatic
stimuli or their effects, which moderate harm or exploit beneficial opportunities (IPCC, 2001b, p.72).
Adaptation does not occur instantaneously; a person or community requires agency, ability and willingness
to realise their adaptive capacity and adapt successfully (Adger et al., 2004).
PREPARING FOR THE FUTURE IN ETHIOPIA 9
This report has looked at the changes that people have made to what they do or how they do it,
without distinction between the stimuli or process for the change, and without reference to whether
or not the change has been successful. Therefore the terms, adapt, adapting and adaptation have
been used throughout the report in this way.
It is important to highlight that this is distinct from the goal of climate change adaptation, which
is that people are better adapted to a changed or changing climate. Climate change adaptation in
practice refers to actions that people and institutions make in anticipation of, or in response to, a
changing climate, and refers only to those changes that are successful, appropriate and sustain-
able. Changes that are not successful, appropriate and sustainable in the light of climate change are
considered maladaptation (see Figure 2).
Actions taken to adapt to climate variability and change can take many forms. In this report we describe
a number of different types of adaptation at the local level (see Figure 2). With this in mind, adaptation
can occur as a result of deliberate planning, known as planned adaptation, or from spontaneous action
as a result of changing shocks and trends, known as autonomous adaptation. In terms of the degree
of change, adaptation can occur as small, incremental changes in livelihoods and practices, or more
transformational actions, requiring fundamental shifts in a systems functions and objectives (Smith
et al., 2010). Important to note is that not all actions taken to adapt will be successful (see Figure 1).
In some cases adaptation actions can increase an individuals vulnerability in the longer term, resulting
in maladaptation (see Figure 2). Though the various types of adaptation are blurred and actions may
constitute more than one type of adaptation, having a thorough understanding of these different types is
useful in helping us to characterise how people respond to change, as well as to better explore the barriers
to successful and sustainable adaptation.
Autonomous Adaptation Adaptation that that occurs naturally by private actors without
intervention of public agencies. Often, autonomous adaptation
does not constitute a conscious response to climatic stimuli, but is
triggered by ecological changes in natural systems and by market
or welfare changes in human systems.
Planned Adaptation Adaptation actions that are result of a deliberate policy decision or
action on the part of public agencies.
a) To understand how existing DRR, Social Protection and Livelihoods projects by ACCRA members build
adaptive capacity to climate change in beneficiaries, and how these approaches can be strengthened.
b) To use the findings to influence donors, development partners and civil society to improve future
planning/action.
c) To work with local and national governments to enhance the capacity to implement interventions
that can build communities adaptive capacity.
Grounding these objectives is done through the use of the LAC framework1. The LAC draws on extensive
consultations with academics, policy-makers and practitioners, and is tested in pilot studies in each
of the three countries. Most assessments of adaptive capacity have focused on assets and capital as
indicators (Dulal et al., 2010). While useful in helping us to understand what resources people need to
adapt, these asset-oriented approaches tend to mask the role of processes and functions (Jones et al.,
2010). Understanding adaptive capacity, therefore, requires that we also recognise the importance of
various intangible processes, such as decision-making and governance, the fostering of innovation and
experimentation and exploiting new opportunities, and the structure of institutions and entitlements, for
example. This means moving away from simply looking at what a system has that enables it to adapt, to
recognising what a system does that enables it to adapt (WRI, 2009). Understanding what development
activities are doing to support this capacity, and what can be done to further enhance it, will be crucial to
strengthening adaptive capacity.
The LAC framework incorporates intangible and dynamic dimensions of adaptive capacity, as well as
more tangible capital and resource-based components, into an analysis of adaptive capacity at the local
level. ACCRAs research recognises that it is not feasible to measure the potential of people and societies
directly. Instead, the LAC proposes that the capacity to adapt at the community level will be broadly similar
in all groups, and separated into five distinct, yet interrelated, characteristics: the asset base; institutions
and entitlements; knowledge and information; innovation; and flexible, forward-looking decision-making
(see Figure 3 and Figure 4).
The underlying assumption behind the framework is that positive impacts on each of these characteristics
should enhance the systems adaptive capacity (Jones et al., 2010). Research was conducted in the three
countries to assess each of the five characteristics. Three sites were selected, drilling down into a range
of livelihoods, vulnerabilities and capacities across each country. Results were then synthesised to draw
common lessons and learning, in order to inform development practice. Key findings from research activities
across the three Ethiopian sites Ander Kello, Kase-hija and Wokin are analysed in the following sections.
PREPARING FOR THE FUTURE IN ETHIOPIA 11
Asset base
Flexible and
Institutions & forward thinking
entitlements decision-making
& governance
Knowledge
& information Innovation
Asset base Availability of key assets that allow the system to respond to
evolving circumstances
Knowledge and The system has the ability to collect, analyse and disseminate
information knowledge and information in support of adaptation activities
Following the development of the programmes conceptual framework and research guidance, the in-
country researched kicked off with an inception workshop bringing together experts from academia,
government, civil society and NGOs to discuss the LAC and adapt it to the Ethiopian context. This was
followed by the development and testing of the research protocol, analysis of available secondary data, an
intensive period of field work in the three research sites, data analysis and the production of site reports
presenting the wealth of collected information, which form the basis for this synthesis report.
Prior to research in the three sites, the LAC research protocol was tested in West Hararghe.
A team of researchers from Haramaya University spent around three weeks in each of the research sites,
with two field work periods at different times of the year to capture seasonal difference. An intensive
process of participatory research and validation, including both beneficiaries of interventions and non-
beneficiaries, was conducted with at least 17 focus group discussions and 32 key informant interviews
in each site at community and district levels. Key informants included government and NGO staff,
representatives of different community groups and representatives of the private sector, where present. A
number of participatory research tools were employed to collect data, including seasonal calendars, wealth
ranking, time lines, community maps, institutional analyses and hazard maps4. Various validation activities
were also conducted with communities and government officials at district, regional and national level.
PREPARING FOR THE FUTURE IN ETHIOPIA 13
Respondents were grouped into different categories according to set criteria (e.g. gender, age, wealth,
livelihood group). Selection of respondents was based on the list of households in the community provided by
local leaders. In Wokin and Kase-hija, a wealth ranking exercise was carried out with the intention to assist in
identifying existing social and economic livelihood groups in the study site in order to represent them adequately
in the different focus group discussions and key informant interviews. Such a wealth ranking, however, was not
possible in the pastoralist site of Ander Kello as village inhabitants refused to classify households based on
their wealth in fear of jeopardising benefits received through the PSNP. In this village, households were only
differentiated based on livelihoods (e.g. pastoralists, agro-pastoralists). A simple sampling frame was followed
to arrive at the numbers of respondents for the different exercises as recommended in the research protocol.
Launching the research in Chifra. In each research site, a training A focus group in Wokin kebele, male farmers discuss the
session on the key concepts being addressed by the research was affecting their lives.
held for district staff prior to starting the research.
1. Local facilitators were used for assisting in research and interpreting in interviews
2. Researchers were introduced to community and local leaders by consortium member operating
in the area
3. Lists of contact people was secured from consortium member prior to field work
4. Compilation of relevant secondary information and meteorological data analysis was done before
field work
5. First phase of field work was carried out between September 2010 and January 2011
6. Collected information was reviewed and preliminary analysis made to guide subsequent data collection
7. Research findings were validated with the participation of Regional and Federal officials in May 2011
8. Second phase of research conducted between March 2011 and April 2011 to fill gaps and capture
information that is sensitive to seasonality
9. Data analysed and site reports produced
10. Site reports analysed and this synthesis report produced
1.6 Limitations
Given the fieldwork took place in only three sites, the ACCRA research did not intend to be a representative
study of the district or a specific agro-ecological zone. Rather ACCRA has used a qualitative approach with
small sample sizes and limited baseline data. We did not therefore aim to be statistically representative
or to conduct rigorous impact assessments of project activities. Rather we intended to generate a rich
discussion and insights about the features of adaptive capacity.
Section 2:
Understanding
the Ethiopian
context
PREPARING FOR THE FUTURE IN ETHIOPIA 15
Reasons for Ethiopias vulnerability are manifold. Its geographical location and topography entail high
vulnerability to the impacts of climate change. The highlands are dominated by sedentary crop farming,
while many lowland areas are characterised by mobile pastoralism, with increasing numbers of agro-
pastoralists in areas between the two. Highlands above 1,500 m asl are the favoured settlement areas,
with around 90 percent of the population living here (Yacob Arsano et al., 2004). Ethiopia has about 16.4
million hectares of arable land (14.6 percent of its total land area), of which about 8 million hectares are
currently used for crop production (Deressa, 2006).
Historically Ethiopia has been prone to extreme weather variability. Rainfall is highly erratic, most rain
falls with high intensity and there is a high degree of variability in both when and where it falls. Since
the early 1980s, the country has suffered seven major droughts five of which have led to severe food
insecurity in addition to dozens of local droughts (World Bank, 2010). One of the reasons for this
pronounced vulnerability is the extremely low level of water resources management, either in the form
of watershed management or investment in multipurpose water infrastructure, which could potentially
include hydropower production, irrigation systems and storage adequate to mitigate both drought and
floods. Less than 5 percent (about 200,000 hectares) of the estimated potential 3.7 million hectares of
irrigable land in Ethiopia is under irrigation (World Bank, 2006), and water storage capacity has been
estimated at only 40m3 per capita5 (Grey and Sadoff, 2006).
Agriculture
Agriculture is dominated by rain-fed small-scale farming, primarily based on traditional technologies.
Modern inputs, in particular fertilisers, are comparably low at an average of 81 kg/ha (Daniel Zerfu & Larson,
2011). Small-scale subsistence farming (about 8 million peasant households) accounts for 95 percent of the
total area under crops and more than 90 percent of total agricultural output. Although the arable area has
expanded slightly in recent years, population growth has outpaced this expansion. Average landholding per
household has dropped to below 1 ha. Average yields remain low, at only 2.1 t/ha for maize, 1.7 t/ha for
wheat, 1.4 t/ha for teff and 1.25 t/ha for barley (Rashid et al., 2010). Although the importance of agriculture,
as measured in terms of its contribution to GDP, has decreased in recent decades, the overwhelming majority
of the population is still mainly rural and depends heavily on agricultural income.
The dominant structure of the agricultural sector in the highlands is household-based, small-scale and
subsistence-oriented. Land is not only the primary means for ensuring a livelihood, but often and particularly
for richer households also a means of accumulating wealth and social status, and transferring them between
generations (Deininger & Binswanger, 1999). Farming practices aim for minimal risks and optimal returns via
a complex mixed system, involving both a variety of crops and livestock. The composition of the household
defines the level of production as well as most of the available labour force. In the majority of households,
most farming tasks are performed by family labour, but employing daily labourers from the community is also
common amongst richer households. Collaborative work and reciprocal exchange are important for specific
time-consuming tasks (e.g. clearing land, ploughing, harvesting).
Livestock
Ethiopia has the largest livestock population in Africa and the tenth largest in the world. Livestock is an
integral part of the farming systems in the country, and the source of social and economic values such as
food, power, fuel, cash income, security and investment. The livestock sector is estimated to contribute
approximately 12 to 15 percent to total GDP and about 25 to 30 percent to agricultural GDP (Deressa,
2006). The livestock sector is the second largest foreign exchange earner after coffee, and in 2006 the
16 PREPARING FOR THE FUTURE IN ETHIOPIA
country earned $121 million from livestock and related products6 (IIED and SOS Sahel, 2010). In 2008,
the direct value of livestock and livestock products from pastoralism was estimated to be $1.22 billion per
annum, with an estimated additional $458 million per annum from other activities (SOS Sahel Ethiopia 2008).
Much of the income generated by livestock comes from pastoral production, the dominant land use in
Ethiopias lowlands (below 1,500m asl). Of the estimated 1.4 million people in Afar region, 90 percent are
pastoralists and 10 percent agro-pastoralists, while of Somali regions 4.4 million people, 85 percent are
pastoralists, and the remainder agro-pastoralists, farmers and urban dwellers. Pastoralists also represent a
significant proportion of the population in Oromia and SNNPRs arid lowlands (Nassef, 2010). Crop farming
is limited in Ethiopias pastoral regions, and constitutes less than 1 percent of Afars land area and only
slightly more than 5 percent of Somali Regions land area (Ibid).
Soil loss rates in Ethiopia are highest on cultivated land, and are estimated at 42 t/ha per year on average.
Total soil loss from all land is estimated at almost 1.5 billion tons per year, of which 45 percent originates
from cropland alone (Hurni, 1993). Of this soil loss, it is estimated that about 90 percent is deposited
downslope, while the remaining 10 percent of sediments are leaving Ethiopia (ibid.). Overall, the impact of
PREPARING FOR THE FUTURE IN ETHIOPIA 17
soil erosion in reducing soil depth, reducing water storage and removing nutrients, and hence reducing soil
productivity locally, are significant. As population pressure increases, more marginal areas and areas that
were previously used for grazing are converted to arable land, leading to further soil erosion. Animals have
to be fed for a longer period of the year from crop residues, which are thus lacking as a soil conditioner
(Ludi, 2004a).
With increasing population and diminishing forest resources, fuelwood requirements exceeded the natural
regeneration capacities of remaining forests. Today, forest resources account for only 4.6 percent of the
total land area. Because indigenous forest resources are very few and decreasing, people are forced to use
animal dung as fuel. Organic matter is thus not brought back to the soil, leading to further soil degradation.
To overcome shortages of timber and fuelwood, Eucalyptus was introduced in the 1890s, and is increasingly
planted around homesteads, both for consumption and as an important cash crop (Ludi, 2004b).
Overgrazing is a serious problem in the densely populated areas of the northern and central highlands and
in dry lowland areas. Livestock are vulnerable because feed resources are extremely limited, rangelands
are under increasing pressure and rangeland degradation is widespread as a result of overgrazing, the
encroachment of crop farming and the spread of invasive plant species (Ludi, 2004b).
Demographics
With over 80 million inhabitants7, Ethiopia is the second most populous country and the fifth-largest
economy in sub-Saharan Africa. The majority of the Ethiopian population is still rural (83 percent). Disparities
between urban and rural areas are extremely pronounced: while 90 percent of the urban population comes
from the richest wealth quintile, this is only 10 percent in rural areas (Demographic and Health Survey,
quoted in Evans & Steven, 2011). Population growth peaked in the early 1990s, when it reached a rate of
3.3 percent a year. It is currently around 2.5 percent per year and is not expected to fall below 2 percent
until after 2025. Health standards are improving, though from a low base. Life expectancy has increased
from 40 years at birth in 1950 to 55 years today (global average 68.9 years). Infant mortality has more
than halved over the last fifty years; fertility levels were around seven live births per woman until 1990,
then started to fall sharply to 4.8 births today. Rates are projected to reach replacement level around mid-
century (ibid.).
Economy
Ethiopia has experienced strong economic growth in recent years. With real GDP growth at or near double
digit levels since 2003/04, the country has consistently outperformed most other countries in Africa.
Official figures show real GDP growth averaged 11.2 percent p.a. during the 2003/04 to 2008/09 period,
putting Ethiopia among the fastest growing economies in sub-Saharan Africa. Although IMF (IMF, 2011)
estimated growth rates around 7.5 percent in 201011, lower than government estimates of 11.4 percent,
this growth performance is still in excess of the population growth rate and the 7 percent rate required for
attaining the MDG goal of halving poverty by 2015 (AFDB, 2010). GDP per capita has steadily increased
over the last decade, from $117 in 2000 to $319 in 2008 (UN Data). Although agriculture is still extremely
important, the service sector has, for the first time in the countrys history, overtaken agriculture as the
largest sector, accounting for 45.1 percent of GDP in 2008/09, followed by agriculture at 43.2 percent, and
industry at just 13.0 percent (Access Capitals, 2010).
The link between climate and GDP in Ethiopia is shown in Figure 5. However, the association of economic
declines following poor rainfall does not imply causality. As Conway & Schipper (2010) point out, food
insecurity is a chronic and complex problem that stems from political, social, environmental and cultural
sources and a number of socio-economic droughts are recorded even when there is no meteorological
drought. Although they conclude that some relationship between rainfall and GDP can be detected, they
also note that the relationship becomes weaker after 2000, primarily because of a greater diversity of
the Ethiopian economy and a concentration of drought in southern and south-eastern Ethiopia, areas with
relatively low agricultural productivity and contribution to GDP.
18 PREPARING FOR THE FUTURE IN ETHIOPIA
30 15
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
-10
-10
-20 -15
GDP Rainfall
-30 -20
(correlation = 0.10, 1 yr lag = 0.24)
Figure 5: Ethiopia rainfall and GDP growth. Source; Conway & Schipper, 2010
Seasonal rainfall is driven by the migration of the Inter-Tropical Convergence Zone (ITCZ), reaching its
northern-most position over northern Ethiopia around JulyAugust, and its southern-most position over
southern Kenya in January and February. This oscillation of the ITCZ characterises the rainy seasons over
Ethiopia, with a main wet season (Kiremt) from mid-June to end-September. Northern and north-eastern
regions, including the eastern Escarpment and parts of the south-eastern Highlands, have a secondary
wet season with sporadic, and much less abundant, rainfall from mid-February to May (Belg). Southern
parts of Ethiopia have two distinct wet seasons in April to June (Gu/Genna/Sugum season) and October
December/January (Deyr/Sapie/Dadaa season).
Jan 2010 Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Jan 2011
Figure 6: Seasonal Calendar and Critical Events. Source: USAID/Famine Early Warning Systems Network (FEWSNET).
The ITCZs movement is influenced by variations in Indian Ocean surface temperatures and varies from year
to year, leading to variations in the onset and duration of the rainfall season from year to year, including
the occurrence of drought. Warming sea surface temperatures particularly in the south-western Indian
Ocean may be linked to decreasing rains in equatorial and subtropical eastern Africa, including Ethiopia
(Funk et al., 2005). The other major driver of rainfall is related to the El Nio Southern Oscillation (ENSO),
with warm phases associated with reduced rainfall during the main rainy season from July to September
(JAS) in northern and central Ethiopia, causing severe drought (Mc Sweeney et al., 2007).
In terms of precipitation, strong inter-annual and inter-decadal variability makes it difficult to detect long-
term trends. There is no statistically significant trend in observed mean rainfall in any season in Ethiopia
between 1960 and 2006. Decreases in JAS rainfall observed in the 1980s recovered in the 1990 and 2000s.
One key problem in analysing rainfall over Ethiopia is the paucity of data (McSweeney et al., 2007).
Conway et al. (2007) find a slight negative trend for the February to April season (Belg) for 19812000 of
0.321.31mm per month. For the June to August season (Kiremt) and for October and November, monthly
precipitation shows a slight positive trend of 0.521.81mm for 19812000. Both authors note that these
average national figures obscure significant regional differences in rainfall trends.
20 PREPARING FOR THE FUTURE IN ETHIOPIA
There are significant differences in different regions of the country, with some parts having relatively
stable rainfall and others showing strong declines. A study of seasonal rainfall trends by Funk et al. (2005)
identified some notable features (see Figure 79):
On a national scale, the March-to-September rainfall totals (upper-right) exhibit a 20-year variation (wet
in the mid-1970s and mid-1990s, dry in the mid-1980s and mid-2000s).
Regional time series suggest variations from this pattern. In north-west (green) Ethiopia rainfall has been
fairly constant over the period of record, with a decline in the mid-1980s followed by gradual recovery to
the present. In the south-west (red) an overall decline since the 1960s is found, with a steep drop after
1996.
Of most concern from a food security perspective is the post-1980 decline in the MarchSeptember south-
east/eastern (orange) rainfall and the post-1996 decline in the north-east (yellow).
Figure 7: Time series of MarchSeptember rainfall at national level (right column) and for four regions (left and centre). Black bars show
seasonal rainfall. Heavy coloured lines show running seven-year means. Orange lines denote rainfall tendencies likely to increase food
insecurity (Funk et al., 2005).
A warming in all four seasons in all regions of Ethiopia (central, east, north, south-west).
Annual warming over the whole of Ethiopia by the 2020s of 1.2C, with a range of 0.7C2.3C.
Annual warming over the whole of Ethiopia by the 2050s of 2.2C, with a range of 1.4C2.9C.
Relatively modest regional differences.
PREPARING FOR THE FUTURE IN ETHIOPIA 21
They further conclude that this warming is associated with more frequent heat-wave events and higher
temperatures, likely leading to higher rates of evaporation and, other influences unchanged, higher rates
of surface water evaporation and higher soil moisture deficits.
Models show differing responses in the mean annual rainfall change over Ethiopia, with some models
predicting more, others less, rainfall.
Modest changes in mean annual rainfall are shown in all models for the 2020s.
By the 2050s, average changes are generally larger, but still modest: for the whole of Ethiopia, average
rainfall may change by +1 to +6 percent depending on emission level.
There is a tendency for models to suggest wetter conditions over Ethiopia by the 2080s, with annual
changes in the order of +20 percent.
Temperature Precipitation
Mean annual temperatures are projected to Projections from different models in the
increase by 1.1C to 3.1C by the 2060s and ensemble are broadly consistent in indicating
1.5C to 5.1C by the 2090s. increases in annual rainfall in Ethiopia, mainly
due to increases in the short rainy season (PND)
All projections indicate substantial increases
in southern Ethiopia.
in the frequency of days and nights that are
considered hot. Projections of change in the rainy seasons AMJ
and JAS, which affect large parts of Ethiopia
All projections indicate decreases in the
and main cropping areas are more mixed, but
frequency of days and nights that are
tend towards slight increases in the south-west
considered cold.
and decreases in the north-east.
Cold nights decrease in frequency more rapidly
The models are broadly consistent in indicating
than cold days, not occurring at all in most
increases in the proportion of total rainfall that
model projections by the 2090s under the
falls in heavy events, with annual changes
highest emissions scenario (A2).
ranging from -1 to +8 percent. The largest
increases are seen in JAS and OND rainfall.
The models in the ensemble are broadly
consistent in indicating increases in the
magnitude of 1- and 5-day rainfall maxima.
Table 3: Summarised outputs of GCM model projections of the future climate for Ethiopia. (Source: McSweeney et al., 2007)
While climate change is certainly a key challenge for Ethiopia, its impacts need to be seen in light of a
great variety of other drivers of uncertainty and pressures facing Ethiopias development and individuals, as
summarised in Section 2.1. Internal drivers such as population growth resulting in decreasing availability of
arable land per household, degradation and even depletion of natural resources undermining the prospect
of agricultural growth, changing land use patterns, competition over the allocation of strategic resources
such as water, conflicts or humanitarian disasters, and externally-induced drivers such as food and fuel
price hikes, worsening terms of trade or conflict in neighbouring countries spilling over the border, all
contribute to Ethiopias vulnerability. The ability of Ethiopians to adapt to climate variability and change will
not only depend on the impacts of these other developments and pressures, but they themselves will be
influenced by, or result from, climate variability and change. In the end, climate change will add another
layer of complexity to already complex development challenges.
22 PREPARING FOR THE FUTURE IN ETHIOPIA
NGO studied Save the Children UK CARE Ethiopia Oxfam GB and ORDA
Partnership SCUK is working in CARE employs its own ORDA has own
arrangement partnership with community facilitator agricultural expert
government staff, uses at KA level to cover 3 at district level,
experts from district. KAs. Otherwise close but otherwise
At local level, SCUK is collaboration with works closely with
also directly engaged Development Agents government experts.
in implementation, and other government Use government
service delivery and staff in implementation Development
engagement with of PSNP. Agents in
community interactions with
communities at local
level
Wokin,
Dabat District
Ande Kello,
Chifra District
Adis Ababa
Kase-hija,
Gemechis District
The climate of the district is arid to semi-arid with a temperature range of 2840C, two rain seasons,
namely Karma (long rain season) from mid-July to mid October and Sugum (short rain season) from March
to end-April. Dadaa is another very short period of rain in January. The annual rainfall in the District ranges
from 200 to 600mm (SCUK, 2007). The bi-modal rainfall pattern is illustrated in Figure 9 using data from
Mille meteorological station, which is around 98km from the study site.
500
37 38
40
26 400
17 13
mm
20 12
6
5 5 4 300
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 200
Source: NMA
100
Figure 9: Average monthly rainfall (mm) at Mille Meteorological Station 1963-2009
0
1963
1965
1967
1969
1971
1973
153
24 PREPARING FOR THE FUTURE IN ETHIOPIA
Chifra District has a total population of 91,078 (CSA, 2008), making it, with 60 persons/km2, one of
the most densely populated Districts in Afar Region (average 25 persons/km2) (Kene, 2008). Around 61
percent of the total population are pastoralists and the rest agro-pastoralist. In terms of wealth distribution,
38 percent of the population is classified as poor, 40 percent middle and 22 percent better-off (ibid). Poor
households purchase 53 percent of their annual food needs, better-off households 47 percent. Food aid
contributes one-third to all food requirements of poor households. Livestock sales are the main source
of income for all households. Poor households additionally are engaged in self-employment such as
the collection and sale of firewood, which contributes approximately 15 percent to their income (SCUK,
DPPBFB, DPPA, 2008).
The district is part of the Araamis ke Adaar pastoral livelihood zone, which is predominantly livestock
based, including cattle, goat, sheep and camels (SCUK, DPPBFB, DDPA, 2008). The number of livestock in
the District is estimated to be 352,316 cattle, 342,286 sheep, 306,720 goats, 126,349 camels and 24,977
donkeys (ARBoFED, 2009). Crop cultivation is increasingly important in three of the 19 Kebeles of Chifra
District, but still small-scale. The total land area under crop production during the 2009 cropping season
was 656 ha, of which 536 ha was irrigated. Major crops grown are maize and sesame (PAPDB, 2010).
PREPARING FOR THE FUTURE IN ETHIOPIA 25
The European Commission Humanitarian Aid and Civil Protection (ECHO)-funded PILLAR project was
established under ECHOs Drought Preparedness Program. Its specific objective was to improve
drought preparedness through protecting and diversifying the livelihood assets of pastoralists in the
drought-prone areas of southern and eastern Ethiopia. There have been three phases of the project
so far, PILLAR I started in 2007, PILLAR II, which was implemented for 15 months (April 2009 June
2010), and PILLAR Plus, launched in September 2010, covering selected Districts in Afar and Somali
Region. The research focused specifically on PILLAR II project activities in Ander Kello kebele. PILLAR
II was organised around four main result areas. These were:
Result 1: Enhanced community capacity and resilience to cope with drought and vulnerability.
This included support to animal health services, though training 40 community based animal
health workers and de-worming and vaccination campaigns for 317,203 livestock. As well as work
on Natural resource management through the creation of area enclosures and forage production
groups (53 women in Ander Kello). Income diversification was also supported through the creation
of livestock marketing (27 women in Ander Kello) and cereal trading groups (not in Ander Kello),
irrigated agriculture was also promoted through the construction of a pond and irrigation canals and
establishment of women farming group.
Result 2: Increased capacity of state and non-state actors in livelihoods-based food security
and early warning information management systems.
Training for CAHWs to collect and compile early warning data at community level was carried out.
SCUK staff entered the data into a simple software programme used to analyse and produce reports.
However, data timeliness has been a problem; the system was not functional at the time of the
research and the information was not used by the District experts.
Result 3: Harmonised and improved coordination between government and NGOs in drought
preparedness response.
A common approach was developed for livestock health care during drought among District staff
and NGOs working in Chifra and staff were trained on the national guidelines for relief interventions
in pastoralist areas. District-level emergency disease response teams were formed.
mm
500
37 38
40
26 400
17 13
mm
20
2.3.2 Kase-hija Kebele, Gemechis
6 District 12 5 5 4 300
Kase-hija Kebele, in Gemechis
0 District, is found in West Hararghe Zone in Oromia Region. It is characterised as
lowland (Kolla) (GDFEDO, 2008).
Jan It is 48km
Feb Mar AprfromMay
theJun
District
Jul capital,
Aug SepKuni,OctandNov
comprises
Dec 1,999 ha. Gemechis is 200
a relatively new District, established only in 2004. It is located 17km from the Zonal capital, Chiro, and 343km
from Addis Ababa. Its topography is varied, and includes three different agro-ecological zones. The District has 100
two rainy seasons, Belg, which extends from March to May, and Kiremt, lasting from July to September. Mean
0
annual rainfall in Mieso, the nearest meteorological station, which is 77km from the study site, is 745mm.
1963
1965
Average Monthly Rainfall (mm), Mieso (1962-2009)
200
153
150
127
101
100 86
68
54
43 39
50 30
20
13 9
450
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 400
Source: NMA
350
300
Figure 10: Average monthly rainfall (mm) at Meiso Meteorological Station 1962-2009
250
mm
The natural forest area inAverageGemechis DistrictRainfall
Monthly has been(mm),reduced
Debarkto(1980-2009)
5,350 ha (6.7 percent), all of which is 200
concentrated in highland areas (WHZFEDO, 2008). Inhabitants of Kase-hija report significant deforestation over 150
350 321 311
the last 3540 years. Potable water coverage in rural areas was 28.1 percent in 2007/08, which is very low
300 100
(GDWO, 2010). In Kase-hija, the main source of water is Kase Spring. Motorised pumps distribute the water to
250 At the time of the study, two of the four water points were not working due to poor
four public tap stands. 50
maintenance. Some200people have their own pipes and sell water to others. Others, especially the poor, fetch
mm
160 0
water from the spring itself, as they are unable to afford water from132
private suppliers. The community has high
1963
1965
150
irrigation potential 100
from three sources: Kase, Burka 78 Jini and Dagaga rivers. While Kase River is permanent and
has a diversion irrigation system built by the 40 government in the mid-1980s, 47 Burka Jini and Degaga rivers are
50 17 12
seasonal and flow only for 2a few months during the wet season. Around half
4 of the
3 households in Kase-hija
0
Kebele and 18 of the 116 households
Jan Feb Marin Hajedin
Apr Mayvillage
Jun haveAug
Jul access
Sep to Oct
irrigation water. Gemechis District has
Nov Dec
significantly increased investment in rural road construction over the last three years. Nevertheless, Kase-hija
has a very poor road connection, with roads impassable during the rainy season and difficult to use during the
dry season. There is no mobile or landline telephone connection in the Kebele.
Gemechis District has a total population of 180,170. Kase-hija Kebele has 1,545 households and a
population of 10,455 (GDFEDO, 2008). Hajedin, a village within Kase-hija, which was the focus of data
400
Mean
collection for this study, has Min
126and Max Temperature
households (C) Mille
and a population (1972-2009)
of 793. The District falls into two livelihood
zones13: the sorghum, maize & chat14 (SMC) livelihood zone and Chercher/Gololcha coffee, chat & maize 350
45
(CGC) livelihood zones. Kase-hija falls within the SMC Livelihood Zone. Households with access to irrigation 300
have chat and also40 plant vegetables (carrots, onions, red beet) and fruit (bananas, sugarcane, coffee).
250
Sweet potato is widely cultivated and is considered an important survival crop during droughts. Richer
T Max
mm
35
Degrees Celcius
households also keep livestock, with cattle, sheep and goats the most popular. Petty T Min
trading, sale of 200
firewood and labouring
30 are important sources of additional income (DPPA/LIU 2009). Linear (T MAx)
150
As part of the field 25
research, the population of Hajedin village were ranked by wealth status. The definitions
100
were generated by the community and reflect local perceptions of wealth. Of the 126 households in
20
Hajedin village, 17 were classified as rich, 30 as medium and 79 households were considered poor. 50
15
0
1970 1980 1990 2000 2010
1963
1965
PREPARING FOR THE FUTURE IN ETHIOPIA 27
CARE Ethiopia has been operational in West Hararghe Zone since 1984 and is one of the NGOs involved
in the implementation of the governments PSNP. The PSNP sits within the wider HIBRET programme,
which has been operational since December 2004 and aims to assist 167,602 people.
Project goals are to reduce food insecurity and increase community resilience. The objectives are:
1. to protect household assets and community resources; and
2. to increase agricultural output through integrated natural resource management and strengthened
civil society.
Under the first objective, the programme provides monthly food transfers for chronically food insecure
households. These are provided in return for labour on public works (to 22,390 people about 12 percent
of the total population of Gemechis District) or as direct support for those unable to work (2,000 people
about 1 percent of the total population). The public works activities aim to enhance community assets,
and take the form of soil and water conservation activities and the construction of basic infrastructure
(feeder roads, health centres, classrooms etc.). Support is also provided for effective and timely community
preparation, mitigation and response to shocks and emergencies. Early warning systems have been
established in 13 Kebeles with emergency preparedness plans developed in 24 Kebeles.
Under the second objective, in Gemechis District as a whole, 1,073 farmers have been trained in
improved agricultural practices, such as making compost and improved traditional storage, home
horticulture production and drip irrigation practices. New seed varieties have been distributed and links
with seed suppliers improved. Efforts to strengthen community-based natural resource management
have focused on area closure and watershed management, accompanied by efforts to promote honey,
fruit and fodder production as livelihood options for 29 landless groups using the conserved land.
In Kase-hija Kebele, two hectares of mountainous land have been enclosed and soil and water
conservation activities carried out with 3.9km of bunds constructed, and over 30,000 tree seedlings
planted. The project has also constructed one new school and maintained ten class rooms. An early
warning committee with eight members has been formed and training was given for 20 other
community members on early warning.
Under Objective 2, in Kase-hija, the project has distributed new maize and haricot bean varieties,
vegetable seeds and fruit seedlings, and promoted compost making, backyard gardening and the use
of improved stoves. Village Saving and Lending Associations (VSLA) were established and a number of
groups, including Landless, Vegetable, Handicraft and Seed Production, were organised and strengthened.
28 PREPARING FOR THE FUTURE IN ETHIOPIA
500
37 38
40
26 400
17 13
mm
20 12
6
5 5 4 300
0
Soil and water conservation activities
Jan
inFeb
Kase-hija kebele May
Mar Apr Jun Jul Aug Sep Oct Nov Dec 200
Dabat District is in North Gonder Zone, Amhara Regional State. It has 26 Kebeles and covers an area of
0
122,328 ha. This study was conducted in Wokin Kebele, which is 13km from Dabat Town on the main road
1963
1965
to Debark (DWARDO, 2010). Some activities focused on Finote Selam village within the Kebele.
250
mm
200
Average Monthly Rainfall (mm), Debark (1980-2009)
350 150
321 311
300 100
250 50
200
mm
160 0
132
1963
1965
150
100 78
40 47
50 17
4 12 3
2
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Source: NMA
Figure 11: Average monthly rainfall (mm) at Debark Meteorological Station 1980-2009
400
Mean Min and Max Temperature (C) Mille (1972-2009)
350
45
300
PREPARING FOR THE FUTURE IN ETHIOPIA 29
harvest extends from October to December. Figure 11 shows average monthly rainfall, taken from Debark
Meteorological station, 15km north of the study site.
Dabat District has a total population of 144,652, with an average household size of 4.5. The population
of Wokin Kebele is 10,304. Average land holding per household is 0.87 ha (DWFEDO, 2009). Finote Selam
village is located in the highland part of Wokin Kebele, with a population of 420 households. Some
130 households (31 percent) in Finote Selam are dependent on PSNP to supplement their annual food
requirements.
The wealth ranking conducted with the community in the study area showed that there are five categories.
Out of the 420 households, the majority (67 percent) were found to be poor and very poor, 25 percent
were classified as medium and 8 percent were identified as rich; 29 percent of all households were
classified as landless and 13 percent (54) were women-headed households.
Livelihoods
Rich (8%) Diversified livelihood options because own broad range of assets, including
access to irrigable land.
Good contacts with local development agents and ability to buy inputs, use
improved technologies such as improved seeds and fertiliser.
Ability to expand access to land through renting and sharecropping.
Able to produce surplus for sale.
Produce additional fodder for their animals from rented crop land.
Additional income from sale of eucalyptus tree, renting town house and
sale of livestock.
Poor and very poor Livelihood depends mainly on farming; poor access to modern farm
(67%) technologies because of limited capacity to buy necessary inputs and
limited connection to extension agents.
Rent out land as they may not have the resource such as oxen, labour or
seed to farm on their own.
Depend on alternative livelihoods outside farming such as daily labour,
borrowing money from local lenders at high interest, etc. to complement
own production.
A significant number depend on PSNP.
In 2007, Oxfam began an Agricultural Scale Up Programme. This 12-year programme aims to
empower and improve the lives of one million smallholder farmers through policy influencing,
enhancing market access and improving gender equity in three regions of Ethiopia. In Dabat district
Oxfam began a partnership with ORDA (Organisation for Rehabilitation and Development in Amhara)
in October 2007 with a focus on two value chains: malt barley and honey.
Malt Barley Value Chain: Implementation strategies included organising farmers into clusters and
providing basic training and technical support on malt barley basic seed production and linking malt
barely producers with markets via malt factories. In the 2010 cropping season, 143 households in
Dabat District participated in barley seed multiplication and 836 in malt grain production, of which
71 and 415 households were women-headed, respectively. In Finote Selam village, 66 households
participated in seed multiplication and 148 in malt barley gain production.
Apiculture: The objective was to improve the income of smallholder beekeepers through modernising
beekeeping practices, organising beekeepers into cooperatives and linking them with private sector
and cooperative unions. The intervention was implemented from March 2008 to April 2009. The
main activities undertaken were training of technical staff and beneficiaries, supply of inputs (e.g.
modern honey production accessories and forage seeds), establishment and strengthening of honey
producers cooperatives and follow-up and technical support. 220 farmers from Dabat District, who
had traditional beehives and who were willing to join the cooperative, were selected.
Rainfall failure and food price rises in 2007/08 threatened to undermine the Agricultural Scale Up
Programme. Alongside efforts to increase cash incomes through market-focused interventions, Oxfam
and ORDA ran an Emergency and Livelihood Response Project between 15 October 2008 and 30
October 2009. The objectives of the project were: duction and purchasing power, and
to enhance smallholder farmers resilience to soaring food prices through increased agricultural pro
to enhance the productive capacities and governance of the agriculture sector to ensure
sustainability.
The project carried out cash for work activities to strengthen the agricultural and marketing
infrastructure, including the construction of irrigation schemes, diffused light stores (for seed
potatoes), grain stores and rural roads. In Wokin, it focused on the construction of a honey collection
centre. The project also created savings and credit groups for women and provided DRM training.
mm
PREPARING FOR THE FUTURE IN ETHIOPIA 500 31
37 38
40
26 400
17 13
mm
20 12
6
5 5 4 300
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec 200
100
1963
1965
1967
1969
1971
1973
1975
Average Monthly Rainfall (mm), Mieso (1962-2009)
200
153
150
127
101
Malt barley and honey100
participants in Wokin kebele 86
68
54
43 39
2.4 Changing climates 50 in the30three research sites Rainfal
20
Although the three research sites are in quite different geographical and climatic areas
13 9 of Ethiopia, ranging from the
450
semi-arid and hot lowlands
0 in Ander Kello to the cool and relatively rainfall-secure highland in Wokin, respondents
Jan Feb Mar Apr
in all sites identified distinct perceived changesMay Jun
to the Jul
local Aug Sep Oct
climate Nov Dec
in the recent past. Patterns of change differ, 400
although in all sites respondents identified increasing temperatures, particularly at night, and changes in seasonality, 350
i.e. changes in the onset and duration of rainfall as well as greater variability and uncertainty in rainfall patterns. All
300
respondents agreed that such changes had profound implications for their livelihoods.
250
mm
These reported changes in the local climate can also be detected in meteorological data from nearby
200
Average
weather stations. Caution, Monthly
however, Rainfallin(mm),
is required Debark (1980-2009)
the interpretation of the following figures, as accurate
long-term meteorological
350 data is not readily available; even
321 311 where formal weather stations exist, data 150
gaps are significant and do not allow any further statistical analysis. As weather data was available for 100
300
none of the research sites, data from meteorological stations some 15km to 100km away had to be used as
250 50
a proxy. The trends reported in the data, therefore, cannot be interpreted as directly reflecting the climate
in the research sites,
200 but may only serve as useful indicators of temperature and rainfall trends.
mm
160 0
132
1963
1965
1967
1969
1971
1973
1975
150
2.4.1 Mille Meteorological
100 Station16 78
Long-term mean maximum temperature40ranged from 35.939.5C, with 47 a mean value of 37.5C. No trend
50 17
in mean maximum temperature 2 4 can be detected, though it seems that 12 positive
3 and negative anomalies
0
have decreased in the last Jan decade.
Feb Unlike
Mar Apr maximum
May Jun temperature,
Jul the
Aug Sep Oct mean minimum
Nov Dec temperature shows
an increasing trend for the period 19792009. Minimum temperatures appear to be lower before the mid-
1980s, after which they increase to slightly below 25C. Why this happened cannot be explained as no further
information is available to assess whether this is a reflection of temperature increases or changing equipment.
Rainfall d
400
Mean Min and Max Temperature (C) Mille (1972-2009)
350
45
300
40
250
T Max
mm
35
Degrees Celcius
T Min 200
30 Linear (T MAx)
150
25
100
20
50
15
0
1970 1980 1990 2000 2010
1963
1965
1967
1969
1971
1973
1975
Source: NMA
Mean annual rainfall in Mille has been recorded at 306.5 mm (112 mm), of which 34 percent is received
from February to May and 55 percent from July to September. The 42-year mean rainfall ranged from
131mm to 662mm, with a coefficient of variation of 37 percent. No trend can be detected over the 42
years considered in this analysis (Figure 13).
Annualrainfall
Annual rainfall(mm)
(mm)recorded
recordedininMille
Mille1963-2009
1963-2009
700
700
600
600
500
500
400
400
13
mm
13
mm
5 4 300
5 4 300
100
100
0
0
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
1962-2009)
962-2009) Source: NMA
mm
39
39
20
Rainfallduring
Rainfall duringmain
mainrainy
rainyseason
season(June-October)
(June-October)
20 9
9
450
in Mille 1963-2009
in Mille 1963-2009
450
Oct Nov Dec 400
Oct Nov Dec 400
350
350
300
300
250
mm
250
mm
200
980-2009)
80-2009) 200
150
150
100
100
50
50
0
0
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
47
47
12 Source: NMA
12 3
3
Rainfallduring
Rainfall duringsmall
smallrainy
rainyseason
season(February-May)
(February-May)
400 ininMille
Mille1963-2009
1963-2009
400
972-2009)
72-2009)
350
100
50
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
20
15
10
1974
in Mille 1963-2009
400
350
300
250
T Max
mm
T Min 200
Linear (T MAx)
150 2000
100
2000
50
0 2000
1963
1965
1967
1969
1971
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
2000
Source: NMA
2000
Figure 15: Rainfall during small rainy season (FebruaryMay) in Mille 1963-2009
While there is a slight increase in rainfall during the June to October rainy season and a slight decrease
during the February to May rainy period, none of the trends was statistically significant. No trend could
be detected in the number of rainy days per year, though a slight increase of maximum rainfall per rainy
day was found.
30
Degrees Celcius
25
20
15
10
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
Source: NMA
1400
1200
34 PREPARING FOR THE FUTURE IN ETHIOPIA
Long-term mean maximum temperatures in Mieso ranged from 29.1C to 31.6C, with an average of
30.4C. The majority of positive temperature anomalies occurred after 1993, indicating warmer days after
1993 in comparison to the period between 1973 and 1993. Long-term minimum temperatures ranged
from 14.1C to 15.9C, with a mean value of 14.8C. A slight increase in mean minimum temperatures
could be detected, with increasing anomaly after 1990, indicating a slight warming of nights. Extreme
Mean Min and Max Temperature (C)
maximum temperatures declined overMiesothe 37-year period.
(1973-2009)
The long-term35
rainfall in Mieso shows a bi-modal pattern typical for the Eastern Highlands of Ethiopia. The
small rainy season in spring (Belg) extends from March to May and the main rainy season (Kiremt) lasts
from July to 30
September. Mean annual rainfall in Mieso was recorded at 745mm, with 30 percent of rainfall
Degrees Celcius
occurring during
25
Belg and 38 percent during Kiremt. Data for the period 1962 to 2009 indicated below-
average rainfall in 1969, 1973, 1984, 1986, 1992, 1995 and 2002, while years with above-average rainfall
20
were recorded in 1967, 1968, 1996 and 1997. Overall, variability is high with a coefficient of variation
of 27 percent.
15
Monthly rainfall has slightly increased for the months of April and September and slightly
decreased for February, July and August, but none of these trends is significant. Although no trend could
be detected10of increasing rainfall amounts per day, a shift of days with maximum rainfall from July/August
1973
1975
1977
1979
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
to September/October could be identified.
2003
2005
2007
2009
1400
1200
1000
mm
800
600
400
200
1962 1967 1972 1977 1982 1987 1992 1997 2002 2007
Source: NMA
2005
2007
2009
Mean maximum temperatures recorded in Debark ranged from 17.9C to 20.9C with a mean value of
19.8C. Mean20minimum temperatures ranged from 7.5C to 9.0C with a mean value of 8.4C. A slight
increase in both mean minimum and mean maximum temperatures can be detected. In terms of
magnitude, the
15 1980s had higher maximum temperatures than the years before or after. An increase in
minimum temperatures was observed after 1994, which corresponds with respondents perceptions of a
decreased incidence
10 of frost.
5
1974
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
1200
200
1000
0
mm
800 1962 1967 1972 1977 1982 1987 1992 1997 2002 2007
600
400
200
Mean Min and Max Temperature (C)
0 Debark (1974-2009)
1962 1967 1972 1977 1982 1987 1992 1997 2002 2007
25
20
15
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
20
15 Source: NMA
5
Total Annual
In terms of rainfall, long-term records indicate Rainfall,Debark
average annual rainfall1980-2009
of 1,126mm and an annual variability
1974
1981
1983
1985
1987
1989
1991
1993
1995
1997
1999
2001
2003
2005
2007
2009
of 24 percent for Debark Meteorological Station. There is only one rainy season in Debark, lasting from May
to September. During this period, 89 percent of annual rainfall is recorded.
2000
2000
2000
Total Annual Rainfall,Debark 1980-2009
2000
2000
2000
2000
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
2000
2000
2000
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
2006
2008
Source: NMA
Whilst there was no change in rainfall amounts in the main rainy months (June, July and August), there was
a slight decreasing trend in the amount of rain falling at the beginning (May) and at the end (September)
of the rainy season. A slight decrease in rainfall can also be detected for March, which is an important
period for land preparation, while there seems to be a small increase in rainfall in November, which might
have negative impacts as it coincides with the harvest of major crops.
Section 3:
Exploring
adaptive capacity
at the local level
PREPARING FOR THE FUTURE IN ETHIOPIA 37
Table 7: The five top-ranked hazards in the three ACCRA research sites. Source: FGD in the three research sites
Kase-hija, Gemechis
Rainfall amount Decreasing
Rainfall distribution Rainfall season shortening
Temperature Increasing
Forest cover Deforestation increased and species disappeared
Farmland size Decreasing
Irrigation Traditional Modern with increasing demand
Invasive species Partinium invading the area
Wokin, Dabat
Rainfall amount Feb-May rainfall becomes irregular
Frost Frost decreasing over time
Temperature Increasing since 1985
Figure 20: Some of the key climate and development stresses identified in each site through the community timeline exercise.
Source: FGD in the three research sites
In all three ACCRA research sites, weather-related hazards droughts or floods were ranked highest by all
focus groups. In all sites, respondents agreed that the impact of stresses on their livelihoods has increased
over time. But they also agreed that their ability to react and adjust had changed. Respondents in all sites
observed that the impacts of climatic hazards, stresses and shocks are felt indirectly. For example, more
uncertain rains and below-average rainfall in the past few years in Ander Kello have been felt in the form
of decreasing rangeland productivity, which has led to reduced numbers of livestock, the spread of invasive
plant species, decreasing water availability, increasing incidence of livestock diseases and worsening terms
of trade. In Kase-hija, seasonal change and pressure on land are felt in terms of decreased productivity, the
increased need for seasonal migration with livestock and increasing food insecurity.
In the old days, there was plenty of grass in this area. We simply released our
animals into the grass which was more than 3m tall Now we dont move our
animals the way we used to because there are far fewer animals. Some buy feed
for their animals, some have enclosures and Government support has increased.
Compared to the past, we have peace and security, and access to school and health
services. But we are getting poorer because of environmental changes and because
we have fewer animals. A K, 68, Ander Kello
Whilst there was broad agreement between women and men in this ranking, particularly in relation to
covariate weather-related hazards that affect all community members, there were also some important
PREPARING FOR THE FUTURE IN ETHIOPIA 39
differences across wealth and livelihood groups and gender in how hazards were ranked, and how their
impacts were described. For example, women and agro-pastoralists in Ander Kello ranked shortage of
water higher than men. Women were also more concerned by human disease, while pastoralist men were
more concerned with animal diseases. Both men and women, however, highlighted that traditional re-
stocking mechanisms had weakened due to declines in herd sizes, making it more difficult for households
to give animals to other households affected by livestock loss. In Kase-hija, poor women ranked shortage
of grazing land quite high, reflecting the increasing pressure on them to migrate with livestock for several
months during the dry season to reach better pasture land. Married women in Kase-hija also noted the
decline in wealth-sharing to help households cope with shocks, whereas men felt that this had not
changed. In Wokin, both poor men and women ranked hail storms as the second major hazard; as they
only have a single plot of land, its destruction would make them extremely vulnerable. Richer households
are more likely to have more than one plot, which means hail damage in one area is less significant. In
all three sites, women noted the psychological stress they faced due to the difficulty of providing food and
water for the household during in times of climate hazards.
Children reflected the concerns of their parents, but also spoke of specific impacts affecting them. For
example, in Ander Kello drought often meant being forced to leave school to search for pasture. In Kase-
hija, children highlighted increasing temperatures wilting crops and intense rain and flooding threatening
them when they were looking after livestock. In Wokin, children said strong winds, hail and intense rain
made live difficult when away from the village with grazing animals.
People reported a wide range of responses, ranging from small incremental changes to existing structures
and systems to deep and major changes to systems, including livelihoods (transformative adaptation).
However, some of these adaptation strategies are being undermined by developments in other areas,
such as increasing pressure on natural resources, the erosion of local institutions and government policies.
Coping with immediate hazards often meant calling on the support of relatives, neighbours and traditional
institutions. In all three sites traditional support institutions were set up to deal with particular, individual
shocks, such as fire, theft or the death of a family member, rather than weather-related shocks which
impact upon the entire community. Perspectives on social support networks obviously depended on who
was speaking and where they were from. In Wokin, the 1991 land redistribution changed the way poverty
was perceived and may have weakened social support mechanisms:
In Ander Kello, the Afar community is organised into clans who feel a strong social obligation to assist each other.
We will continue helping each other until the end. If I have only a cow which is
pregnant and my fellow brother comes for help, I will promise to give him the calf
when it gets born. We know we are getting poor together, but we also know we
wont starve to death while our clan members are having something to eat. We
survived the past horrible years because of this culture and we will continue to do
so. Man in Ander Kello
In all three sites, communities have developed livelihood strategies with in-built adaptive capacity that
has historically allowed them to thrive in a highly variable climate. For example, in the pastoral/agro-
pastoral Ander Kello, people respond to drought by changing their temporary migration routes. If such
adjustments are no longer available, people start selling livestock. However some long-term development
pressures make these traditional strategies inadequate and so communities shift resources within the
40 PREPARING FOR THE FUTURE IN ETHIOPIA
Pastoralists in Ander Kello report preferences for browsers given shortages of pasture.
The CARE community facilitator discusses preferences for short maturing maize with Haramaya researcher
PREPARING FOR THE FUTURE IN ETHIOPIA 41
system. For example, as a result of ongoing rangeland degradation coupled with more frequent droughts,
households report changing their herd composition, reducing the number of grazers (sheep and cattle)
in favour of more resistant browsers (goats, camel). However ultimately these pressures are such that
pastoral households look for more radical solutions and in some cases have changed their livelihoods more
fundamentally and become increasingly engaged in crop cultivation. However, all these changes also lead
to changing institutional arrangements. Crop cultivation in such semi-arid environments relies on irrigation,
which prevents livestock from accessing water at critical times. Smaller herds make longer migration less
attractive, and private enclosures for fodder production more attractive. In Ander Kello this trend is still
limited, though increasing, but in other pastoral areas the spread of private enclosures to produce fodder
has inhibited the movement of animals and increasingly leads to conflict with livestock keepers (Flintan
2011). This highlights the importance of strengthening capacity to evaluate adaptation options and avoid
maladaptive choices which will ultimately limit peoples livelihoods.
In Kase-hija, perceived changes in rainfall patterns have led people to replace long-maturing sorghum
varieties that have increasingly failed due to water stress at critical times, and replaced them with shorter-
maturing maize varieties, promoted and made available by the extension system, but probably also made
more attractive because of buoyant markets for maize. Inhabitants of Kase-hija are increasingly investing
in small-scale irrigation and have expanded irrigation infrastructure. New institutions have developed that
allow households without their own irrigable land to plant sweet potato on the irrigable land of other
households during the dry season or in times of stress.
In Wokin, inhabitants are adapting to increased flooding, which, they say, is a result of changing rainfall
but also of changing land use and grazing patterns by investing in soil conservation, building check dams
around farmland to give protection from runoff from adjacent land, constructing better drainage systems
and stabilising and rehabilitating gullies. As pressure on arable land increases and agricultural production
stagnates or declines, households are adopting supplementary livelihood activities such as planting
Eucalyptus as a cash crop or growing horticulture crops for sale. Communities now realise that whilst
Eucalyptus is a valuable source of additional income, it cannot be planted near to cropping areas or water
points, therefore they highlight that without proper management it could be considered maladaptation.
Access to roads and markets can make a significant difference to how people deal with climate variability
and stress. In Wokin, households can diversify their livelihood activities and therefore spread risks, whereas
the very poor road connection in Kase-hija prevents people from engaging in high-value cash-crop
production, with the exception of chat20.
The rain is in the hands of God, but the road is in the hands of the Government. We
cant do anything about the rain, but the Government could help us by building us
road and bringing us telecommunication service. Farmer, Kase-hija
Having access to roads and markets has also meant that people in Wokin can engage in daily labour or self-
employment, which, they feel, helps them to deal more effectively with stress and shocks. In Ander Kello,
pastoralists noted that they had not seen significant livelihood benefits as a result of better infrastructure and
services and better market prices, mainly because they had fewer animals to sell.
Access to a range of assets helps to diversify livelihoods and so to deal with climatic variability. In the two
agricultural sites, respondents mentioned that having different plots in different agro-ecological niches
helps in dealing with climatic hazards, such as flooding or hailstorm, whilst having access to communal
grazing land, albeit increasingly degraded, helps with keeping livestock, which is not only a productive
asset but an important means of savings. In Kase-hija in particular, being able to access irrigation water
was described repeatedly as a key determinant of wealth accumulation in a changing climate as it allows
two to three cropping seasons and decreased dependence on rainfall. In Ander Kello, livestock are still
the major productive asset and rainfed agriculture has largely failed to produce any significant return
on investment. However, those who access irrigated crop land have a more diverse range of livelihood
42 PREPARING FOR THE FUTURE IN ETHIOPIA
options. Currently these are small areas that are under irrigation and which may help some households to
adapt, but as the amount of irrigated land increases, the analysis of the role irrigation plays in adaptation
and resilience may change, because irrigated agriculture is at the expense of livestock production on
communal grazing land. In many contexts it became clear that the availability of assets like grazing land is
in itself not enough without the institutions that guarantee access to them (see Box 1 ).
People reported that the impacts of climate stress and shocks are becoming worse, in large part due
to factors independent of the climate: increasing population, decreasing availability of and access to
natural resources, decreasing landholdings and increasing costs of inputs and basic consumer goods. Local
adaptations documented in all three sites are a reaction to multiple pressures. Such insights will help
in designing development interventions that reflect the reality that adaptation decisions are usually a
response to a variety of social, economic and environmental factors.
Respondents pointed out that development project activities have contributed to strengthening and
diversifying the asset base on which they rely. Examples include interventions targeted at individual
households, specific groups of households within a community21 and communities as a whole. Interventions
can also be grouped into those that are aimed at protecting key assets, and those that are aimed at
diversifying the asset base upon which households depend. The former include the establishment of
village savings and credit associations (Kase-hija) or activities carried out under the Productive Safety
Net Programme (PSNP), such as investments in soil and water conservation using food or cash-for-work
approaches (all sites). The latter includes activities which introduce new agricultural technologies or
practices and which try to improve incomes through marketing. Examples include the development of a
small-scale irrigation scheme and the formation of a womens group to engage in horticulture production,
including skills development and training (Ander Kello) or the establishment of malt barley production
clusters (Wokin). Some interventions accompanied work on assets with support for the establishment of
related institutions for example in Wokin besides providing barley seeds and training on how to produce
high-quality malt barley, links with the beer brewery have also been established. However, the research
findings were ambivalent as to whether the mechanisms for scaling up or achieving sustainability of some
of the strategies being promoted was well-understood by the implementers. For example the promotion
of irrigated agriculture and rangeland enclosure in pastoral areas could potentially be considered a
maladaptation, if, in the long term, it restricts mobility
PREPARING FOR THE FUTURE IN ETHIOPIA 43
The role of the government in protecting key assets and strengthening livelihoods was stressed in all
sites. These activities are rarely related to climate variability and change but focus on the provision of
infrastructure such as roads, and services such as education, health, adult training and skills development,
agricultural extension and business development support. These assets and services are considered crucial
in enabling people to respond to a changing climate or other hazards. However, respondents also observed
that local government provision of these services is often inadequate, mainly because of shortage of
human and financial capacity. For example, although agricultural extension agents are posted in the
villages, inputs are not always timely or provided in sufficient quantities. Significant investment is still
needed to meet the basic needs of communities in these areas. So far limited time has been spent on
thinking about how such services can take risks associated with the changing climate into consideration,
but ultimately this is needed to ensure that these services contribute towards building adaptive capacity
and avoid potential maladaptation.
Although the findings do not suggest that wealth differences directly correlate with innovation or peoples
willingness to experiment and change, there is evidence from all research sites that those with more
assets are more likely to adopt externally promoted innovation packages, which tend to be input-intensive,
and depend on improved seeds or fertiliser. Explanations given for this include the broader asset base of
the better-off households and thus their greater ability to take risks, but respondents also stressed that the
better social status of wealthier households gives them better access to information and new knowledge
provided by agricultural extension agents at community level, and agricultural experts at District level.
Better-off households could also more easily access the credit necessary for agricultural intensification
programmes based on costly inputs such as artificial fertilisers and improved seeds.
44 PREPARING FOR THE FUTURE IN ETHIOPIA
I am the first person who tried application of fertiliser on a farm land in Finote
Selam. I am also the first to have good number of beehives, poultry and a eucalyptus
plantation. I have never received food aid, rather I receive advice and information.
Better-off farmer, Dabat
Asset-poor and vulnerable households in all three sites have benefited from the Productive Safety Net
Programme, which has prevented them from having to migrate or sell key productive assets to meet
basic household needs. However, this support is rarely sufficient to lift recipients out of poverty. Valuable
community assets have also been built, for instance through hillside enclosures, afforestation and soil and
water conservation schemes. However, from the research in Kase-hija, it is clear that there are limitations
in the scale of this work, considering that PSNP has been in operation in the area for five years. In Wokin,
the lack of an overall watershed plan or suitable biological materials for soil bunds has undermined the
sustainability of public works activities.
It is not part of the PSNP design to directly support households to engage in new activities to transform
their livelihoods, for example by moving out of agriculture into rural non-farm employment. Whilst the
Household Asset Building programme could in theory do this through the provision of credit, in reality this
credit appears to be mostly used to supplement traditional livelihoods with the purchase of additional
livestock. This raises the same question which the use of irrigation in pastoral areas raised: when is support
to adaptation positive? Just as it can be asked whether it is better in the long term to support pastoralism
as a resilient livelihood or to promote irrigation for crop farming, which inherently undermines pastoralist
livestock rearing systems, so it can be asked whether it is positive to support people in poverty in their
existing livelihoods, as the PSNP has done, or whether this is maladaptation. The support provided might
have prevented households from engaging in more radical transformative adaptation. Whether the PSNP
has thus help to trap households in increasingly unviable livelihoods or has helped them to avoid complete
destitution is a matter of debate that cannot be further explored in this report. Worryingly, it is not clear
that a sufficiently in-depth analysis of this question underlies actual development interventions.
The institutional environment in Ethiopia is highly complex and characterised by a pluralism of formal
and informal institutions, sometimes overlapping and competing with each other. Examples are gender
roles and responsibilities, including norms that prevent women from ploughing land, religious rules and
regulations, for example the number of religious holidays in Orthodox Christian areas, and the rules
governing access to natural resources such as arable land, grazing land and irrigation water. Most of these
rules have their own enforcement mechanisms. These kinds of rules are often labelled traditional, though
this is misleading since they are far from unchanging. Gender roles, for example, change as more girls
receive education, some religious rules are questioned by the government, insisting that farmers should
observe fewer holidays, and local rules over access to natural resources are overlaid by formal laws that
govern how land is allocated.
In all research sites, respondents highlighted the importance of local institutions in influencing their responses
to shocks and stresses, particularly though their role in governing entitlements to assets. Although land is
owned by the state and people only have usufruct user rights, local institutions are decisive in organising
the use of the land. For example, in sites dominated by crop cultivation, local institutions are responsible for
dealing with sharecropping arrangements (e.g. defining which party contributes how much labour and which
PREPARING FOR THE FUTURE IN ETHIOPIA 45
share of the crop they receive), rental arrangements and inheritance arrangements. In pastoral areas, use of
rangeland, livestock migration patterns and private enclosures are governed by local institutions dominated
by the clan system. Access to irrigation water from locally developed schemes such as in Kase-hija and
Wokin is also usually governed through local institutions (water committees). How these institutions function
determines which households get more water, and which ones lose out (see Box 1).
What happens when institutions are neglected: the case of Kase-hijas irrigation scheme
Although communities in Kase-hija had a functioning irrigation system, they still expressed regret
that CAREs HIBRET project did not work on irrigation because of the institutional dimension.
Access to irrigation water was the main determinant for improved livelihoods and food security.
The physical availability of water is not the same as having access to it, and it is the institutional
management of any irrigation scheme which determines who actually benefits. In Kase-hija, as
usually happens in Ethiopia, communities were organised into user committees. These committees
are supposed to ensure that user in each village gets water every month, each farmer getting
water for 23 hours.
Farmers complained that bribes are paid to the leaders of the water committees, so those who
cannot afford to pay bribes (i.e. poor people) and people of low social standing (such as female-
headed households) do not get irrigation water at all or only in limited quantities. The incentives for
excluding the poor increase as demand for water increases for cash crop production. Communities
believe that available water is declining due to deforestation around the source and irregular
rainfall. This only increases the need for an institution that can manage water in an equitable and
sustainable way.
Box 1 What happens when institutions are neglected: the case of Kase-hijas irrigation scheme
46 PREPARING FOR THE FUTURE IN ETHIOPIA
Local institutions and the way rules are enforced are crucial for understanding differences in adaptive
capacity within a community. In the worst case they may even prevent certain groups from adapting and
thus make them more vulnerable to stresses and shocks. Rules such as the right to free-graze livestock
after harvest on crop residues increases the capital of some (livestock-owning) households, but may
undermine efforts to increase natural capital through soil and water conservation (of benefit to farmers).
To analyse the contribution of local institutions to adaptive capacity, a much stronger focus will have to
be placed in future on understanding underlying power structures that shape institutions. The tendency of
organisations (NGOs and government) to discuss problems and interventions in terms of the community
as if this were a homogenous unit is an illustration of the lack of attention to power relations and the way
in which institutions mediate access to assets and knowledge.
Marginalised groups in particular women, but also the very poor and landless youth face considerable
discrimination, which reduces their ability to adapt to stresses and shocks. In all research sites, community
politics and decision-making is heavily male-dominated. Although in all three sites very resolute and
outspoken women were found, they seldom influenced decisions or debates, and access to crucial
resources, such as irrigation water in Kase-hija, is skewed towards male-headed households (see Box 1).
Although formal government structures are often dominated by better-educated young or middle-aged
inhabitants, informal structures and decision-making are solidly in the hands of elder male inhabitants. In
the case of formal government services such as agricultural extension, women reported receiving hardly
any specific information or technology advice targeted at activities which are typically in the womens
domain, and that they need to have their husbands present in discussions with agricultural extension
agents. The biggest restriction women face in terms of productive livelihood activities, however, is that they
are discouraged from ploughing by persistent social norms, which means that female-headed households
have to depend on sons or male relatives for land preparation. This puts them in a very disadvantaged
position as the timing of land preparation is late, which reduces productivity. Female-headed households
are often forced to enter into unfavourable sharecropping arrangements, further undermining their ability
to cope and adapt. Women reported that, in times of hardship, institutional barriers are increased and they
are subject to considerable work and livelihood pressures. Not only do they have to carry out their normal
household chores, such as caring for children and the elderly or cooking, but labour-intensive tasks such
as collecting fuel wood and water often become more demanding. Without support from male household
members, women are often forced to remove children from school, which in the long term will undermine
their ability to adopt new livelihood activities and adapt to changes.
A mixed picture emerges in the way development interventions addressed marginalisation by local
institutions. While there is evidence of some support for people who are marginalised (for example through
the establishment of women-only savings and credit associations in Kase-hija, by targeting the poor and
vulnerable in the PSNP in Kase-hija and Wokin22 or through womens groups in Ander Kello), there is much
less evidence of attempts to address institutional marginalisation itself. The formation of groups womens
savings and credit groups, landless groups engaged in natural resource conservation and its productive use
or womens farming and livestock marketing groups has provided a valued space for discussion, collective
action and risk sharing. There is less evidence, however, that these groups influence local decision-making
or are sufficiently strong and unified to challenge restrictive institutions. That said, they have enhanced
agency among group members mainly through the provision of valuable assets to help members respond
to stress and shocks. However, just establishing a group of marginalised people without addressing the
institutional barriers and power structures that determine their marginal status will not be sufficient to
enhance their agency or adaptive capacity.
Institutions can also support or constrain innovation. As discussed below (see 3.2.4) interventions which
introduced innovations sometimes ignored the institutional dimension, leading the project to fail once the
direct support of the NGO had ended. For example, the introduction of enclosures in pastoral Ander Kello
successfully demonstrated how quickly rangeland can regenerate when not disturbed by grazing animals,
but it did not successfully work at the institutional level, necessary to support the management
PREPARING FOR THE FUTURE IN ETHIOPIA 47
of the enclosure. Thus, while respondents admired the quick regeneration of the vegetation during the dry
season, they reverted back to old practices of free grazing as there were no institutions in place that would
have allowed the continued exclusion of livestock or the sharing of the created assets (grass). Cultural
values have also rewarded conformity rather than innovation (i.e. deviation from established norms), as
discussed below. The research did not find examples where outside interventions had made any attempt
to address such institutional constraints.
Communities noted the importance of traditional knowledge in decision-making, but also the increasing
challenges to these traditional methods in the face of perceived changes in seasonality and natural resource
degradation. Traditional weather forecasting methods were not seen as particularly reliable anymore.
However, the information provided by the National Meteorological Agency via radio was also not trusted
or used. Although seasonal forecasts are available, information is not passed down to districts and end-
48 PREPARING FOR THE FUTURE IN ETHIOPIA
users. Farmers and pastoralists thus have no reliable source of long-term weather forecasting to aid their
planning. This problem was not being addressed by development interventions in any of the research sites.
Communities and households can adapt much better if they have reliable information both about what
they need to adapt to (trends, predictions of prices, climate, etc.) and about their current options (available
technologies, prevailing prices, etc). Agencies are not focussing on this dimension of development, and there
are signs that they themselves are not using information adequately in designing their own interventions.
For example, support for crop cultivation in Ander Kello is based on the assumption that irrigation water
will continue to be available in sufficient quantity. Current projections of climate variability and change
suggest that this might not be the case. The findings thus indicate two issues. First, agencies do not see
their role as providing information to their beneficiaries or helping them to analyse its implications, e.g.
to help the population of Ander Kello to make its own mind up about the advantages and disadvantages
of relying on irrigation. This could not be done without engaging with the ways in which such decisions
get taken, and the power relations involved the institutional framework. In other words, dealing with one
dimension of adaptive capacity almost always entails engaging with others. Second, if agencies (NGOs or
governmental) are basing interventions on inadequate or insufficient information, they might undermine
adaptive capacity by encouraging the adoption of livelihood strategies that in the longer run prove to be
examples of maladaptation.
Most project interventions focused on the poorest and the vulnerable. In Wokin, however, project interventions
were deliberately targeted at those who had land in a particular area or who were already engaged in
bee-keeping. Ultimately this meant better-off households were targeted who were able to engage in new
enterprises and practices such as malt and seed barley production. (This still equates innovation with the
early adoption of an externally provided technology.) Research further suggests that those most likely to
adopt new practices not only have above-average assets, but have social networks that enable them to
access information and knowledge from sources beyond the village (for example, by having relatives in
larger towns), have travelled to areas beyond their village of residence (for example because they were
serving in the army) or are able to access information from experts at the district level. These factors which
foster the innovation and adoption of new practices were not assessed by projects and did not inform the
planning of development interventions. Research from all three sites suggests that it is not the introduction
of new technologies and practices that is lacking, but that development interventions fail to support an
enabling environment that allows autonomous innovation and experimentation, which ultimately would
mean new technologies were more appropriate and sustainable. The institutional framework that supports
innovation was not considered by projects, and the previous example of establishing pasture enclosures in
Ander Kello (see 3.1.2) is an example of how this tends to lead to unsustainable change. There were also
examples where agricultural research was producing useful ideas, but the focus on getting set technology
packages adopted was overriding local testing and adaptation.
PREPARING FOR THE FUTURE IN ETHIOPIA 49
Even during this research, agencies argued that they cannot be expected to deal with everything in
what are sometimes small projects. This reveals a continued adherence to a model of interventions that
depicts them as linear, mirroring a simplistic log-frame where inputs lead to set outputs which bring
about predetermined outcomes. Such a model has long been critiqued but still remains prevalent in
donor and NGO monitoring and reporting requirements and is therefore prioritised by the managers of
projects. Whether or not agencies are aware of it, the outcomes of their interventions will always be
mediated by institutions and power relations and there are always unintended impacts both positive
and negative. Agencies can choose to ignore this complex reality but they cannot make it disappear.
However small the project, if an agency wants to achieve its objectives it has to be aware of the myriad
dimensions of its intervention. If it takes the need for sustainability seriously, it is also bound to consider
adaptive capacity, since all project outputs will necessarily need to be adapted and changed over time
as contexts inevitably change. Taking adaptive capacity seriously means, for example, using the five
lenses described in section 1 to analyse the possible impact of an intervention and ensuring that the
project design maximises its contribution23. All the characteristics of adaptive capacity will always be
relevant to every project, however small: the choice is merely whether or not to be aware of them.
There are examples of considerable leadership and initiative by residents in improving their situation.
In Kase-hija, for example, farmers have extended an irrigation scheme established some 30 years ago,
building a dam to store water and extending the channels to irrigate additional areas. Extending the
system was not an easy task it has taken the community seven attempts to finally get things right. At
times the majority of residents wanted to give up, but a few individuals championed the work, against
all odds and despite receiving no support from irrigation experts. However, these examples are too rare,
52 PREPARING FOR THE FUTURE IN ETHIOPIA
and are not routinely encouraged. Initiative is undermined by limited participation and a perception by
decision-makers that they know best what farmers and pastoralists need. This attitude leads government
and development partners to supply packages of technologies which they expect farmers to adopt,
whatever their specific needs or the local circumstances and without giving any opportunity for end-
users to say what they think would support them best. This puts farmers and pastoralists at the mercy
of government and development partner staff and stifles their ability to hold them to account. It also
leads people to expect everything to come from outside and solutions to be delivered without their own
contribution. Examples can be found in all three research sites. In Ander Kello, for example, people often
said that they needed the government to provide irrigation water or livestock health services.
First I planted maize, then I started growing grass for the goats. There was no
water nearby and I struggled to keep the plants alive by carrying water from far
water points. When I started cultivation and refused my daughters marriage, the
community called me crazy. I wasnt crazy, but I decided to focus on my farming.
Farmer, Ander Kello
Agency can be supported in various ways. Supporting social organisations that encourage the sharing
of ideas and the spreading of risk is one way, and has already been discussed above. It is important
to identify the constraints to agency, which may come from within or outside the community. In some
communities social pressure was preventing agency, with people being excluded for innovating;
external actors need to find ways to support agency without being seen to be in conflict with local
customs and norms.
PREPARING FOR THE FUTURE IN ETHIOPIA 53
Group formation, however, is not free of difficulties. There are dangers of elite capture by influential
individuals as well as corruption; groups can be inefficient and transaction costs high, all of which
undermine the credibility of the group. Across the different groups studied in this research, success meant
having a clear sense of purpose and effective governance, choosing activities that were economically
important to the groups members and being able to scale up and support people in larger numbers.
Attention also needs to be paid to the local social system so as to avoid social exclusion, as happened with
some groups in Ander Kello, where groups are often dominated by members of influential clans.
Taking on board agency will require a major paradigm shift in how local people are perceived and
treated. Development actors will have to learn to trust people in new ways, to see their role as
supporting peoples own life projects and to stop trying to determine how they should use the assets,
information and opportunities they have. This message is not new, but is yet to be taken seriously,
despite the ubiquitous rhetoric of participation.
Helping residents to protect their assets and improve their livelihoods is laudable, but not sufficient. In
all sites, pressure on available arable land and natural resources is already considerable and is only going
to increase. More focus is needed on interventions beyond agriculture. Current project interventions are
maintaining existing systems, but are not contributing enough to building adaptive capacity. Part of the
reason for this limited focus is that most development efforts concentrate on outputs (not outcomes),
and are dominated by the provision of hardware. Expenditure and monitoring focus on increasing the
number of irrigation channels, not on ensuring that everyone receives their fair share of irrigation water.
At the same time, the role of the private sector as a rural service-provider is rarely considered by NGOs.
Whilst the private sector is in its infancy in all research sites and a reliance on the state or NGOs in input
supply, there is reduced flexibility in supply chains and this leads to less choice and reliability. The late
provision of inputs, after the time they are needed by farmers, was common. In Wokin and Kase-hija,
improved inputs provided by the extension services were in short supply and no alternative sources were
available. Geographical remoteness and limited purchasing power of residents means it has not yet been
profitable for private sector to engage in input supply in the research sites. In the ACCRA research sites,
neither the government nor NGOs have shown an instinct for fostering competition and encouraging
private input suppliers, which would be more responsive to local demands or other circumstances
such as changing climate. Whilst private animal health workers had been trained in Ander Kello, their
drug supplies were rapidly exhausted in offering free treatments and could not be easily re-stocked.
More efforts to train both animal health workers and their clients to perceive the service as a business,
alongside efforts to establish a more effective drug supply chain could have improved the sustainability
of this service.
It is striking how often it is said that the private sector is entirely absent in rural Ethiopia, despite the fact
that almost the entire rural population buys and sells goods at mutually agreed prices for private profit.
Until it is more widely recognised that farmers and pastoralists are the private sector it is unlikely that it
will enjoy greater support.
Reassessing scale and scope of interventions does not mean that NGOs have to do everything
themselves, but rather, they should aim to leverage change through influencing governments or
fostering sustainable businesses through support to the private sector. Unfortunately, development actors
(Government, donors and NGOs) do not give adequate attention to dynamic processes of change such
as the use of information in decision-making, the development of equitable institutions, the fostering of
local innovation and improved accountability in governance structures. Many of the ways in which these
software aspects could be improved have low input costs but require time and highly qualified staff.
Taking this challenge on board requires a change in how projects are budgeted, financed, staffed and
monitored. It will be very hard to make these changes if project interventions continue to have a very
short lifespan, rarely as much as three years, and often with changing priorities during the projects life.
4.3 Enhancing the use of information and knowledge for evidence-based decision-making
and project design
More than once in this report, it has been left open whether or not an introduced technology is
leading to better adaptation to future conditions or to maladaptation. Although all predictions remain
uncertain, projects need to be designed based on a thorough analysis of predicted trends and their
implications trends in population demographics, in world and local price movements, in technology
and communication development and, of course, in climate change. The ACCRA research found that
projects are not being designed using the best possible information, prediction and scenario analysis.
This is true both of longer-term shifts (e.g. climate variability and change) and short-term changes (e.g.
specific seasonal weather forecasts). The latter are not available at local level. Whilst some information
is disseminated to regional offices, it rarely reaches the district or community level. NGOs do not analyse
this information routinely and do not use it to plan project interventions in the sites studied. Weather
forecasts are mainly used for predicting emergency food needs in planning for a humanitarian response.
PREPARING FOR THE FUTURE IN ETHIOPIA 55
There is also a need for more sophisticated analysis of current vulnerabilities and what causes and maintains
them. This should be expanded to include constraints to adaptive capacity. Entry points for interventions can
come from looking at institutional barriers to livelihoods and adaptive capacity, understanding who lacks
knowledge and information and why, barriers to innovation and issues of governance.
Projects need to ensure not only that they use the best possible information, but also that it is more widely
adopted by others, especially state services and policy-makers. This includes using both broader and more
locally specific knowledge, e.g. using better assessment of weather patterns in Kase-hija to select more
appropriate crop varieties, using climate prediction and analysis to develop appropriate livelihood strategies
for pastoralists and agro-pastoralists in Ander Kello and to avoid maladaptation by encouraging reliance on
a technology (e.g. irrigation) that may not be sustainable. This involves information being made available
and used both by those designing policy and interventions (government, civil servants, NGOs, etc.) and
local people themselves. The lack of reliable weather forecasting for farmers and pastoralists has already
been noted (see section 3.2.3). Helping farmers analyse for themselves the implications of a particular
weather forecast (or, for example, predicted price movements) would combine support for agency with
support for knowledge and evidence-based decision-making.
A number of less successful development interventions could have greatly benefited had they been
designed using available information and the experiences of similar interventions elsewhere in Ethiopia.
Examples include the design of soil and water conservation practices or rangeland enclosures, where
technical interventions, without adequate institutions are prone to fail. This could cover many dimensions
relevant to project success: institutions and institutional barriers (e.g. best practices for irrigation water
allocation and governance, management of rangeland enclosures), technical aspects of interventions
(e.g. best practices in the construction of ponds, irrigation schemes), organisational aspects (e.g. the
timeframe of project interventions, the quality and capacity of staff, available time to engage in lengthy
processes, strengths and weaknesses of local organisations).
Local government officials in Chifra, Afar discuss the ACCRA research findings on policy constraints to adaptation and risk reduction
56 PREPARING FOR THE FUTURE IN ETHIOPIA
.4 Building adaptive capacity at individual and community level requires a continuous process of
4
learning, change and innovation and an enabling policy environment
In order to prepare residents and households better for dealing with the projected impacts of greater
climatic variability and extremes in future, innovation and strengthening innovative capacity must
take centre stage in any development initiative. Innovation here is understood as a process of
experimentation and exploration of practices, techniques or new organisational forms; innovation
systems are defined as networks of organisations, enterprises and individuals that focus on bringing new
products, processes and forms of organisation into use. They not only create knowledge, but they also
provide access to knowledge, create demand for new knowledge, share knowledge and foster learning
(Rajalahti et al., 2008).
T he ACCRA research suggests that local residents have made a number of innovations, but these have not
been replicated (as discussed in section 3). For people to be innovative, they need to have:
access to resources to test new things and a safety net to fall back on in case
of failure; and
s discussed above, the research found that projects had not identified or set out to address these
A
conditions, even though some interventions had an indirect impact on them. The idea that poor and
poorly educated farmers and pastoralists might be capable of experimenting and learning for themselves
was not found to be well ingrained among development actors. A substantial body of literature exists on
innovation and farmer experimentation, and on how to support it, but it is ignored by most development
interventions.
Access to appropriate information about different options that can be used to solve particular problems
An awareness of different options is usually required to inspire individuals to experiment. Those village
residents that have been to areas beyond their village have often brought back new ideas and have
started experimenting in their village. The current approach of the agricultural extension system is to
offer farmers and pastoralists fixed packages of inputs and training on specific new technologies, and
PREPARING FOR THE FUTURE IN ETHIOPIA 57
many projects mirror this approach. Whilst some technologies have been widely adopted, others are
unlikely to be taken up unless their value can be better demonstrated and supply and demand problems
are resolved. In general, farmers and pastoralists are not in a position to assess the suitability of
proposed new technologies or practices because they lack adequate and appropriate information.
Access to resources to invest in testing new things or to act as a safety net in case of failure
In all three sites, better-off households were more likely to adopt new technologies than poor
households, as they could invest in resources such as water pumps, tree seedlings, beehives or improved
seeds without endangering their immediate livelihood prospects. In Kase-hija and Ander Kello, project
interventions tried to overcome this by, for example, initiating savings groups for poor households. Work
to remove barriers to adoption, though, still rests on the assumption that what is being promoted is right
and right for everyone. The idea has rarely been considered that, if the richer are the ones adopting what
is being promoted, this may indicate that what is being promoted is relevant to the better-off and not to
the poor. Apart from addressing barriers of investment capital and risk, there is also a need to invest in
supporting the poor with the skills and knowledge to make informed decisions about when and how to
experiment and invest, instead of expecting them only to adopt off-the-shelf packages.
Enabling policy and institutional environments which encourage and promote innovation
An enabling environment to encourage innovation is necessary at local level, as well as at higher levels.
Better links are needed between local communities and the agricultural research and extension system
at regional and federal level. Despite the national extension policy being described as a Participatory
Demonstration and Training Extension System (PADTES), research has shown that extension packages are
in fact formulated at Federal or Regional level and there is limited capacity at District and Community
level for involving communities in decision-making, resulting in disempowering local residents to find
solutions to their problems (EEA/EEPRI 2006, Belay 2003). In Wokin and Kase-hija, there was evidence
that this was still the case, and in Ander Kello, there was limited evidence of agricultural extension
advice reaching pastoral communities.
Two perspectives were heard on how to react to the degraded rangeland: move away from relying
on grazing livestock on the rangeland, or work to improve it. Which strategy will prove more
resilient to climate change?
Our village used to be green and rich with We are experts. We know how to rear animals.
livestock pasture. Now all that is gone. With The Government needs to tap this potential
it, our livestock production deteriorated and develop the livestock economy with us
significantly. We cant continue this way. We Government can assist us by developing our
have to change (and) turn to agriculture. Our grazing land. We have extensive grazing land
land is fertile; we only need Government to which used to be very productive in the past
bring us irrigation water. and highly degraded now.
It is clear that much more attention needs to be paid to government development planning both
central and local. NGOs also need to give greater emphasis to integrating their own activities with
local government processes. However, though it was clear that whilst this area is crucial, it was beyond
the scope of ACCRA research to analyse the functioning of central or local government. This research
therefore identifies this as a key area for further effort, but cannot make any specific recommendations.
4.7 Conclusion
Research in the three sites in Ethiopia provides ample evidence that development interventions can
contribute to adaptive capacity at household and community level. Yet the findings also show that
interventions do not recognise the importance of adaptive capacity, and their analysis of poverty and
vulnerability only considers one or two dimensions, principally broadening the asset base. This usually
occurs without addressing the underlying institutional barriers that prevent some households from
accessing those assets. There are also suggestions that institutional barriers and power structures that
increase the vulnerability of some households are insufficiently analysed and understood, and therefore
interventions fail to contribute to improved livelihoods for some households; interventions are often
carried out in isolation, different actors do not consult each other sufficiently, which leads to duplication
and inefficiency, and different actors are not learning sufficiently from experiences of others.
The research concludes that, by using the LAC framework developed by ACCRA to assess contributions
to adaptive capacity, more focused interventions could be developed that target both immediate
development needs and longer-term adaptation requirements, based on forward-looking anticipation
of changes and threats. Interventions can combine different approaches disaster management, social
protection and livelihoods promotion all of which are necessary to adequately support local residents
and address their development needs. This will only become more important given anticipated climatic
and other changes.
PREPARING FOR THE FUTURE IN ETHIOPIA 59
Endnotes
1. For more information on the LAC framework and its characteristics see Jones et al. 2010 or access the consultation document on the ACCRA website:
http://community.eldis.org/accra/
2. In the Ethiopian context known as Wereda.
3. The study compares analysis of meteorological data with that of community perceptions, the meteorological rainfall and temperature records used for climate
analysis for Kase-hija and Ander Kello Kebeles were obtained from meteorological stations that are far (>50 Km) from the studied sites and the data used for all
three sites had a number of missing years which makes it harder to identify trends. In addition, some data, such as hourly rainfall which could be used to analyse
rainfall intensity, was not available.
4. Ludi et al., 2010, ACCRA research protocol.
5. In comparison, South Africa has a water storage capacity per capita of around 750 m3 (World Bank, 2006)
6. This income relates to officially recorded exports only but is most likely to be much higher if non-official exports were considered as well.
7. The 2007 census gave a population of 74 million (CSA 2008); UNDP (2010) estimated the population to be 85 million in 2010.
8. See www.cgdev.org/section/topics/climate_change/mapping_the_impacts_of_climate_change for details.
9. These general statements should be read with caution, as pooling of data from several meteorological stations (most of them with missing data) masks important
location-specific weather patterns.
10. In the Ethiopian context known as Kebele Associations (KA).
11. See Annex 1 for more information on Ethiopian agro-ecological classifications
12. Whilst the Productive Safety Net Programme (PSNP) was presented and discussed with respondents in all three sites. Only in Kase-hija was it a major focus
of the study,
13. As defined by the Household Economy Analysis methodology (HLA, 2010).
14. A flowering plant native to Ethiopia which contains an amphetamine-like stimulant and is chewed widely across Ethiopia, but particularly in the eastern parts
of Oromia and Somali region. East and West Hararghe Zones are main producing areas of this highly valuable cash crop, which in the 2008-2009 financial year
accounted for $139.2 million in export earnings (Reuters, July 21st 2009)
15. See annex 1 for more information on Ethiopian agro-ecological classifications
16. Temperature data available for 1979 2009, rainfall data for 1963 2009. Years were excluded from the analysis if values are missing for more than three days.
Trend lines shown in the figures are only indicative because of missing values.
17. Temperature data available for 1973 2009, rainfall data for 1962 2009. Years were excluded from the analysis if values are missing for more than three days.
Trend lines shown in the figures are only indicative because of missing values.
18. Temperature data available for 1980 2009, rainfall data for 1980 2009. Years were excluded from the analysis if values are missing for more than three days.
Trend lines shown in the figures are only indicative because of missing values.
19. Respondents did not always distinguish problems from hazards. Both concepts fall under chigger or its equivalent in Afar and Oromifaa
20. A mild stimulant which has an increasing demand and can command a high price for export. Therefore traders are prepared to travel long distances over rough
terrain to access it during the dry season, when it can be grown under irrigation in Kase-hija, but may not be as plentiful in other areas which harvest during the
wet season.
21. When using the term community in this report, it usually refers to a village.
22. The PSNP is operational in Ander Kello, but the community refuses to participate in targeting so transfers are shared equally between all households in the
community.
23. Other frameworks for thinking about adaptive capacity may use slightly different lenses.
Site Reports
Million Getnet, Kindie Tesfaye, Beneberu Shimelis, Hiluf Gebrekidan, Belay Kassa, 2011, A Research Report on Assessment of Local Adaptive Capacity in Chifra
District, Zone 2, Afar Regional State, Ethiopia. Haramaya University.
Beneberu Shimelis, Million Getnet, Kindie Tesfaye, Hiluf Gebrekidan, Belay Kassa, 2011, A Research Report on Assessment of Local Adaptive Capacity in Gemechis
District, West Hararghe Zone, Oromia Regional State, Ethiopia. Haramaya University.
Kindie Tesfaye, Million Getnet, Beneberu Shimelis, Hiluf Gebrekidan, Belay Kassa, 2011, A Research Report on Assessment of Local Adaptive Capacity in Dabat
District, North Gonder Zone, Amhara Regional State, Ethiopia. Haramaya University.
References
Access Capitals, 2010. The Ethiopia Macroeconomic Handbook 2010. Access Capitals, Addis Abeba.
Adger, N. et al., 2003. Adaptation to climate change in the developing world. Progress in Development Studies, 3(3), pp.179-195.
Adger, N., Brooks, N. & Kelly, M., 2004. New Indicators of Adaptive Capacity, Norwich: Tyndall Center for Climate Change Research.
AFDB, 2010, Ethiopias Economic Growth Performance: Current Situation and Challenges. Economic Brief, 1/5.
ARBoFED [Afar Region Bureau of Finance and Economic Development], 2009, Regional Atlas. Semmera.
Belay, K, 2003, Agricultural Extension in Ethiopia: The Case of Participatory Demonstration and Training Extension System. Journal of Social Development in Africa.
Vol 18, No 1.
Boyd, E., Grist, N., et al., 2009. Exploring Development Futures in a Changing Climate: Frontiers for Development Policy and Practice. Development Policy Review,
27(6), pp.659-674.
Brooks, N., 2003. Vulnerability, risk and adaptation: A conceptual framework, Norwich: Tyndall Center for Climate Change Research.
Byerlee, D., Spielman, D.J., Alemu, D., Gautam, M., 2007. Policies to promote Cereal Intensification in Ethiopia: A Review of Evidence and Experience. International
Food Policy Research Institute. Development Strategy and Governance Division, Discussion Paper 00707.
CGD [Centre for Global Development], 2010, Map and vulnerability ranking accessed at http://www.cgdev.org/section/topics/climate_change/mapping_the_
impacts_of_climate_change (June 2011), The map is based on Quantifying Vulnerability to Climate Change: Implications for Adaptation Assistance, D. Wheeler,
Centre for Global Development,
Conway, D., Schipper, L., Mahmud Yesuf, Menale Kassie, Persechino, A., and Bereket Kebede. Reducing Vulnerability in Ethiopia: Addressing the Implications of
Climate Change. Report prepared for DFID and CIDA. University of East Anglia, Norwich.
Conway, D. and Schipper E. L. F. (2011) Adaptation to climate change in Africa: Challenges and opportunities identified from Ethiopia. Global Environmental Change
21 (2011), 227237.
CSA [Central Statistical Agency of Ethiopia], 2008, Summary and statistical report of the 2007 population and housing census: Population size by age and sex. FDRE
Population Census Commission, Addis Abeba
Daniel Zerfu and D. F. Larson, 2011, Incomplete Markets and Fertilizer Use: Evidence from Ethiopia. Paper presented at the CSAE 25th Anniversary Conference 2011:
Economic Development in Africa, March 2011, Oxford.
Daze, A., Amborse, K. & Ehrhart, C., 2009. Climate Vulnerability and Capacity Analysis, Care International.
Deininger K, Binswanger H, 1999, The evolution of World Banks land policy: Principles, experience, and future challenges. The World Bank Research Observer 14
(2): 247-276.
60 PREPARING FOR THE FUTURE IN ETHIOPIA
Deressa, T T, 2006 Measuring the Economic Impact of Climate Change on Ethiopian Agriculture: Ricardian Approach. CEEPA Discussion Paper No.25. Centre for
Environmental Economics and Policy in Africa: University of Pretoria.
DPPA / LIU [Disaster Preparedness and Prevention Authority / Livelihoods Integration Unit], 2009, Livelihood Zone Profiles. Government of Ethiopia.
Dulal H et al., 2010, Capitalising on Assets: vulnerability and adaptation to climate change in Nepal, Washington, DC: The World Bank.
DWARDO [Dabat Woreda Agriculture and Rural Development Office], 2010, Socio-Economic Information of Dabat District. Dabat.
DWFEDO [Dabat Woreda Finance and Economic Development Office], 2009, Dabat District Socio Economic Information. Dabat.
EEA / EEPRI [Ethiopian Economic Association / Ethiopian Economic Policy Research Institute, 2006, Evaluation of the Ethiopian agricultural extension with particular
emphasis on the Participatory Demonstration and Training Extension System (PADETES). Addis Ababa.
Evans, A. and D. Steven, 2011, Ethiopia Scarcity Country Case Study: Background Paper: Understanding the Scarcity Dynamic. Working Paper. Center on International
Cooperation, New York University.
FDRE [Federal Republic of Ethiopia], 2007, Summary and Statistical Report of the 2007 Population and Housing Census. Population Census Commission.
Flintan, F (2011), The causes, processes and impacts of land fragmentation in the rangelands of Ethiopia, Kenya and Uganda, Regional Learning and Advocacy
Programme for Vulnerable Dryland Communities (REGLAP).
Funk, C., Senay, G., Alemu Asfaw, Verdin, J., Rowland, J. Diriba Korecha, Eilerts, G., Michaelsen, J., Saud Amer and Choularton, R., (2005), Recent Drought Tendencies
in Ethiopia and Equatorial Subtropical Eastern Africa. Fews Net Famine Early Warning System Network. Vulnerability to Food Insecurity: Factor Identification and
Characterization Report. Number 01/2005. Fews Net.
GDFEDO [Gemechis District Finance and Economic Development Office], 2008, Recurrent Detail Budget by Source of Finance. Gemechis.
GDWO [Gemechis District Water Office], 2010, Gemechis District Water Office Strategic Plan (2003-2007). Gemechis.
HDR [Human Development Report], 2010, The Real Wealth of Nations: Pathways to Human Development. UNDP, New York.
Hurni H. 1993. Land degradation, famine, and land resource scenarios in Ethiopia. In: Pimentel D., [ed.]. World Soil Erosion and Conservation. Cambridge: Cambridge
Studies in Applied Ecology and Resource Management. pp. 27-61.
IIED and SOS Sahel UK (2010). Modern and Mobile: The future of livestock production in Africas drylands. IIED, London.
IMF, 2011, Mission Statement following Article 4 Consultation with Ethiopia
IPCC, 2007. Climate Change 2007: The Physical Science Basis, Working Group I to the Intergovernmental Panel on Climate Change Fourth Assessment Report,
Cambridge, UK: Climate Change 2007: The Physical Science Basis.
IPCC, 2001a. Climate Change 2001: Impacts, Adaptation, and Vulnerability. Working Group II contribution to the Third Assessment Report of theIntergovernmental
Panel on Climate Change, Cambridge, UK: Cambridge University Press.
IPCC, 2001b. Climate Change 2001: Working Group I - The Scientific Basis, Cambridge University Press.
Issack, A A, Yusuf A, 2010. Save the Children UK experiences on piloting Community Based Early Warning Systems in 3 districts of Somali and Afar Regions of
Ethiopia.
Jayne, T S, Daniel Molla ,1995, Toward a Research Agenda to Promote Household Access to Food in Ethiopia. Food Security Research Project, Working Paper 2. Addis
Abeba: Ministry of Economic Development and Co-operation (MEDaC).
Jones, L, Ludi, E, Levine, S, 2010, Towards a characterisation of adaptive capacity: a framework for analysing adaptive capacity at the local level. ODI Background
Note, December 2010. ODI, London.
Jones, L, Ayorekire, F, Bharihihai, M, Kagoro, A, Ruta, D, 2011, Understanding Adaptive Capacity at the local level across Uganda: Africa Climate Change Resilience
Alliance (ACCRA) Uganda Synthesis Report. ACCRA.
Kene, H.E. 2008, Baseline Survey of 55 Districts of PCDP Phase III. Addis Ababa
Kuriakose, A., Livia, B. & Bachofen, C., 2009. Assessing vulnerability and adaptive capacity to climate risks: Methods for investigation at local and national levels,
World Bank.
Lim, B. & Spanger-Siegfried, E., 2004. Adaptation Policy Frameworks for Climate Change, Cambridge: Cambridge University Press. Available at: http://www.undp.
org/climatechange/adapt/apf.html.
Ludi, E, 2004a, Economic Analysis of Soil Conservation: Case Studies from the Highlands of Amhara Region, Ethiopia. Geographica Bernensia, A18. Centre for
Development and Environment, Bern.
Ludi, E ,2004b, Soil and Water Conservation in the Ethiopian Subsistence Farming System. Unpublished manuscript. Centre for Development and Environment, Bern.
McGray, H., Hammil, A. & Bradley, R., 2007. Weathering the storm: Options for Framing Adaptation and Development, World Resources Institute.
McSweeney, C., New, M., Lixcano, G., (2007) Ethiopia UNDP Climate Change Country Profiles.
Nassef, M., 2010, Synthesis of Existing Knowledge and Experience on the Provision of Water Supplies to Pastoral Communities in Ethiopia. RiPPLE, Care Ethiopia,
Save The Children, US.
NGZWRDO [North Gondar Zone Agriculture and Rural Development Office], 2010, Annual Report of 2009/10 Budget Year. Gondar.
North D C, 1993, The New Institutional Economics and Development. Washington University, St. Louis.
PAPDB [Pastoral and Agro-pastoral Development Bureau], 2010, Annual (2009/2010) Agricultural Basic Information. Afar Regional Sate, Afar.
Rajalahti R, Janssen W, Pehu, E, 2008, Agricultural Innovation Systems: From Diagnostics toward Operational Practices. World Bank, Washington DC.
Rashid, S., Kindie Getnet and Solomon Lemma (2010) Maize Value Chain Potential in Ethiopia. Constraints and opportunities for enhancing the System. IFPRI Working
Paper July 2010. IFPRI, Washington DC.
Samuel Gebreselassie, 2006, Land, Land Policy and Smallholder Agriculture in Ethiopia: Options and Scenarios. Future Agricultures Discussion Paper 008. Future
Agricultures, Brighton.
SCUK (Save the Children, UK), DPPBFB (Disaster Prevention, Preparedness and Food Security Bureau) and DPPA (Disaster Prevention and Preparedness Agency),
2008. Livelihoods and Vulnerabilities: An Understanding of Livelihoods in Afar Regional State, Ethiopia, Addis Ababa.
SCUK (Save the Children, UK), 2007. Nutrition assessment report of Chifra district of Zone 1, Afar Region, Addis Ababa
Smit, B. & Wandel, J., 2006. Adaptation, adaptive capacity and vulnerability. Global Environmental Change, 16, pp.282-292.
Smith, M. et al., 2010. Rethinking adaptation for a 4C world. Philosophical Transitions A, 369, pp.196-216.
SOS Sahel Ethiopia (2008) Pastoralism in Ethiopia: Its Total Economic Values and Development Challenges. For IUCN-WISP, Nairobi.
TERI, 2007. Adaptation to Climate Change in the context of Sustainable Development, The Energy and Resources Institute. Available at: http://www.teriin.org/
events/docs/adapt.pdf.
Tompkins, E. & Adger, N., 2004. Does Adaptive Management of Natural Resources Enhance Resilience to Climate Change? Ecology and Society, 9(2), p.10.
van Aalst, M.K., Cannon, T. & Burton, I., 2008. Community level adaptation to climate change: The potential role of participatory community risk assessment. Global
Environmental Change, 18(1), pp.165-179.
WHZFEDO [West Hararghe Zone Finance and Economic Development Office], 2008, Bio-Physical Data of West Hararghe Zone. Chiro.
World Bank, 2006, Ethiopia: Managing Water Resources to Maximize Sustainable Growth. A World Bank Water Resources Assistance Strategy for Ethiopia. World
Bank, Washington DC.
World Bank, 2010. Ethiopia Economics of Adaptation to Climate Change. Wold Bank, Addis Abeba.
WRI, 2009. The National Adaptive Capacity Framework: Pilot Draft, Washington, D.C.: World Resources Institute.
Yacob Arsano, Berhanu Debele, Eva Ludi, Seyoum Gebre Selasie, 2004, Managing Resources and Disputes in Uncertain Environments. JACS Horn of Africa. In: Hurni,
H., U Wiesmann, R. Schertenleib (eds.), 2004, Research for Mitigating Syndromes of Global Change. A Transdisciplinary Appraisal of Selected Regions of the World
to Prepare Development-Oriented Research Partnerships. Perspectives of the NCCR North-South. Vol 1. Geographica Bernensia, Bern.
PREPARING FOR THE FUTURE IN ETHIOPIA 61