The Impact of In-Kind Food Assistance On Pastoralist Livelihoods in Humanitarian Crises

JANUARY 2017
Evidence Synthesis
Humanitarian Evidence Programme
THE IMPACT OF IN-KIND FOOD ASSISTANCE

ON PASTORALIST LIVELIHOODS IN
HUMANITARIAN CRISES
About this evidence synthesis
This is an independent evidence synthesis commissioned by the Humanitarian Evidence
Programme a partnership between Oxfam GB and Feinstein International Center at the
Friedman School of Nutrition Science and Policy, Tufts University. It was funded by aid from
the United Kingdom government through the Humanitarian Innovation and Evidence
Programme at the Department for International Development. The views and opinions
expressed herein are those of the authors and do not necessarily represent those of Oxfam,
the Feinstein International Center or the UK government.
About the research team

This report was written by:
Karol Czuba, PhD Candidate, Department of Political Science, University of Toronto
(Principal Investigator) corresponding author, [email protected]
Tyler J. ONeill, Clinician Scientist, Dalla Lana School of Public Health, University of
Toronto
Ana Patricia Ayala, Instruction and Faculty Liaison Librarian, Gerstein Library, University
of Toronto.
The initial database and website searches took place during June 2016.
Citation
Czuba, K., ONeill, T.J. and Ayala, A.P. (2017). The impact of food assistance on pastoralist
livelihoods in humanitarian crises: An evidence synthesis. Humanitarian Evidence Programme.
Oxford: Oxfam GB.
Acknowledgments
The authors would like to thank the HEP, Oxfam GB, Feinstein International Center and
DFID for the opportunity and funding to conduct this research project. Special thanks go to
Eleanor Ott at Oxfam and Roxanne Krystalli at Feinstein for their support over the course of
the project.
Series editors
These report forms part of a series of humanitarian evidence syntheses and systematic
reviews covering child protection, market support, mental health, nutrition, pastoralist
livelihoods, shelter, urban contexts, and water, sanitation and hygiene. The reports and
corresponding protocols can be found at:
https://www.gov.uk/dfid-research-outputs
http://fic.tufts.edu/research-item/the-humanitarian-evidence-program/
http://policy-practice.oxfam.org.uk/our-work/humanitarian/humanitarian-evidence-
programme.
The series editors are: Roxanne Krystalli, Eleanor Ott and Lisa Walmsley.
Photo credit
Andy Hall/Oxfam 2011.
Copyright Oxfam GB 2017

This publication is subject to copyright but the text may be used free of charge for the
purposes of advocacy, campaigning, education and research, provided that the source is
acknowledged in full. The copyright holder requests that all such use be registered with them
for impact assessment purposes. For copying in any other circumstances, or for re-use in
other publications, or for translation or adaptation, permission must be secured and a fee
may be charged. Email: [email protected]
CONTENTS
0 EXECUTIVE SUMMARY I
1 INTRODUCTION 1
1.1 Background 1
1.2 Rationale and purpose 2
1.3 Research questions 3
2 METHODOLOGY 6
2.1 Protocol development 6
2.2 Implementation of the search strategy 6
2.3 Document screening 7
2.4 Evidence assessment 8
2.5 Data extraction and evidence synthesis 9
3 RESULTS 10
3.1 Summary of reviewed publications 10
3.2 Included publications 11
3.3 Outcomes reported by included publications 12
3.4 Assessment of the strength of evidence and causal claims 12
3.5 Narrative synthesis of results 13
4 CONCLUSION 19
4.1 Limitations 20
APPENDIX 1: DATABASE SEARCH STRATEGY STRINGS 30

APPENDIX 2: INCLUDED PUBLICATIONS AND REPORTED INTERVENTIONS AND
IMPACTS 40
REFERENCES 42
Included publications 42
Other 43
0 EXECUTIVE SUMMARY
This evidence synthesis, commissioned by the Humanitarian Evidence Programme (HEP)
and carried out by a research team from the University of Toronto, represents the first ever
attempt to identify, synthesize and evaluate existing evidence on both the short- and long-
term impacts of in-kind food assistance on pastoralist populations and their livelihoods in
1
humanitarian crises.
The synthesis is based on a rigorous methodology and follows an adapted scoping study
approach. It aims to:
verify the quality of existing evidence relating to this key type of humanitarian intervention
help researchers identify the strengths and weaknesses in such evidence, and thus to
recognize potential improvements and opportunities in future research
assist practitioners and policy makers in evaluating the impact of past choices and
investments.
Definitions
This review uses the term pastoralists to refer to people who depend on livestock for a significant proportion of
their food and income, including those who might also cultivate crops and/or carry out other economic activities.
The term livelihoods is used to refer to the capabilities, assets (including material and social resources) and
activities required for a means of living (Scoones, 1998: 5). The review covers a wide range of livelihood
systems, based on a list derived from Hesse and MacGregor (2006: 5).
For further detail on the definitions used, see the box Definition of common terms included in this
review on pp.1-2 of the full report.
What are the impacts of in-kind food assistance in humanitarian

crises on pastoralists and their livelihoods?
The research team looked at 12 key questions that mapped on to six thematic outcome
areas (see Figure 0.1).
1
The Humanitarian Evidence Programme (HEP) is a partnership between Oxfam GB and the Feinstein International Center at Tufts
University. It is funded by the UK governments Department for International Development (DFID) through the Humanitarian Innovation
and Evidence Programme (HIEP).
The impact of in-kind food assistance on pastoralist livelihoods in humanitarian crises ii
Figure 0.1: The impact of in-kind food assistance on pastoralists livelihoods summary of
findings
Thematic Number of Evidence Research question*** Findings

outcome studies* strength**
Changes in 11 Medium Q1: In what ways have pastoralist livelihood Food assistance can undermine the
livelihood strategies changed since 1967 (and to what livelihood strategies of pastoralists.
strategies and extent have non-pastoralist livelihood strategies
asset and supplanted them)? What is the potential causal
income relationship between these changes and food
dynamics assistance?
Q2: In what ways have pastoralist asset Food assistance can lead to reduced
dynamics changed? What is the potential livestock sales and strengthen herd
causal relationship between these changes growth.
and food assistance?
Q3: In what ways have income dynamics Food assistance may fill gaps in
changed? What is the potential causal pastoralists incomes.
relationship between these changes and food
assistance?
Mobility patterns 9 Limited Q4: In what ways have mobility patterns Food assistance can lead to changes
changed? What is the potential causal in pastoralists mobility patterns, and
relationship between these changes and food especially to sedentarization.
assistance?
Access to in- 7 Limited Q5: What types of in-kind food assistance Food assistance to pastoralists can be
kind food interventions have been offered to pastoralist insufficient and unbalanced.
assistance populations in the context of humanitarian
crises? How have these interventions, the
nature of the strategies, the distribution of food
assistance and its duration changed over time?
Q6: In what ways do pastoralists use the food Four publications claim that food
they receive? Do they consume it, or use it as assistance can lead to dependency,
livestock feed or as a commodity to sell or but there is no relevant empirical
barter? evidence.
Q7: In what ways has access to food and non- According to one publication, food
food items in pastoralist areas changed, assistance can lead to an increase in
including in relation to markets? What is the alcohol production. In some cases
potential causal relationship between these food assistance targeting has been
changes and food assistance? controversial where pastoralists have
perceived issues of unfairness,
unequal distribution and elite capture.
Household- and 7 Limited Q8: What household-level shifts have taken Food assistance can encourage
individual-level place among pastoralist populations? What is pastoralist women to seek alternative
socio- the potential causal relationship between these livelihood strategies.
demographic changes and food assistance?
factors Q9: What individual-level shifts have taken According to most publications, food
place among pastoralist populations? What is assistance leads to decreased
the potential causal relationship between these malnutrition. Others report, however,
changes and food assistance? that food assistance can have negative
impacts on recipients health
outcomes.
Social relations 4 Limited Q10: What shifts have taken place in the social Food assistance can both strengthen
and governance relations of pastoralist populations, including in relations within existing social
relation to social support networks? What is the networks and contribute to the
potential causal relationship between these emergence of new political leaders
changes and food assistance? (and displacement of their
predecessors).
Security 4 Not Q11: In what ways have the security conditions The publications do not make a causal
applicable within which pastoralist livelihoods take place link between food assistance and
changed? What is the potential causal security.
relationship between these changes and food
assistance?
Notes:
* Some publications report more than one outcome and are included in one or more themes.
** Evidence strength: medium = moderate-quality studies, medium-size evidence body, moderate level of consistency; limited =
moderate-to-low-quality studies, medium-size evidence body, low levels of consistency.
*** The evidence synthesis addresses 12 research questions (Qs). Evidence on 11 of them is presented in Figure 0.1. A further question
(Q12) What are the populations and regions affected by food assistance interventions targeting pastoralists? What are the
commonalities and variations between these populations and regions? is considered in the section below on the state of the evidence.
The impact of in-kind food assistance on pastoralist livelihoods in humanitarian crises iii
What evidence was eligible for synthesis?

Of the 23,424 publications identified in an initial search of academic databases, and an
additional 1,442 from grey literature sources, 24 were deemed suitable for synthesis
following screening and quality appraisal. These publications:
are all in English (the search strategy was limited to English and French)
were all published between 1983 and 2015 (and mainly since 2000)
are mainly primary studies (n=20)
include studies that used qualitative methods (n=13), quantitative methods (n=6) and
mixed methods (n=5).
Collectively, the selected 24 publications report food assistance interventions during or after
humanitarian crises that:
took place between 1967 and 2012
lasted between one and 20 years
were predominantly drought-related (n=21) but have other identified causal factors,
including conflict (n=3), famine (n=4), flood (n=2) and disease (n=1)
took place primarily in countries in the East and Horn of Africa, with the remainder in
Algeria, Niger and Mongolia
targeted Mongolian herders in Asia (n=2) and a range of African pastoralist ethnic
groups, including the Beja, Gabbra, Oromo, Pokot, Rashaida, Saharawis, Somalis and
Turkana
targeted populations of between 6,000 and an estimated 20 million people
involved unconditional provision of food in-kind (n=23) or in exchange for assets

(livestock) and work (n=2); nine publications fail to identify the modality of food assistance
provision
were carried out by non-governmental organizations (NGOs) (n=7), United Nations
agencies (n=5), national governments (n=4), religious bodies (n=2) or a combination of
these (n=6).
What is the state of the evidence?

Reported evidence indicates that:
Food assistance can, in some cases, achieve its primary goal of addressing food insecurity.
Evidence from Kenya and Somalia indicates that some food assistance interventions have
led to a decrease in rates of malnutrition. On the other hand, provision of food assistance in
Ethiopia and Sudan is claimed to have had negative impacts on health outcomes. The
strength of evidence relating to this theme is limited. Provision of food assistance can
contribute and, as evidence from Kenya and Sudan indicates, has contributed to the
erosion of pastoralist livelihoods. However, interventions in Kenya seem to have enabled
some pastoralists to hold on to their assets, including livestock, and to have supported their
incomes. The strength of evidence relating to this theme is medium.
There is fairly uniform but, because of limited strength of evidence, not necessarily
reliable evidence from Kenya and Sudan that the provision of food assistance leads to
changes in pastoralists mobility patterns, especially sedentarization. The strength of
evidence relating to this theme is limited. Claims that food assistance can lead to
dependency are relatively widespread, but there is no identified empirical evidence of a
causal relationship. On the other hand, food assistance can have other detrimental
effects: according to one publication, provision of food assistance in Kenya has led to an
increase in alcoholism. The strength of evidence relating to this theme is limited.
In some reported cases in Kenya, Mongolia and Sudan the modes of targeting of food
assistance have led to internal controversy (related to unequal distribution, perceived
unfairness and elite capture) within pastoralist communities. Elsewhere, they have
effectively encouraged the emergence of new political leaders who have sought to
channel assistance to their clients and, by extension, have restricted some intended
The impact of in-kind food assistance on pastoralist livelihoods in humanitarian crises iv
beneficiaries access to food; evidence from Kenya and Sudan indicates the existence of
an association between food assistance and the creation of new power structures.
Conversely, one publication reports that an intervention in Kenya encouraged sharing of
food and thereby strengthened existing social networks. The strength of evidence relating
to these themes is limited. There is no evidence in the included publications of the
existence of a causal link between the provision of food assistance and the security
situation in areas inhabited by its beneficiaries.
This review uses guidelines developed by the UK Department for International Development
(DFID) and published in the Assessing the Strength of Evidence note (DFID, 2014) to
evaluate the strength of evidence reported in the included publications. Based on their
conformity with the criteria defined by the note, this review does not consider the strength of
evidence reported in any of the publications to be high. The strength of evidence in 13
publications is classified as moderate, while the remaining 11 publications report evidence of
low strength. Causal inferences made in included publications are especially problematic. In
most publications, the method of analysis is not reported; causal claims are not supported by
evidence and are reported as the authors impressions of observed events or outcomes.
While the research team has chosen not to exclude publications on the grounds of the
strength of reported evidence, claims made in most of the included publications are not
reliable.
Because of the lack of high-strength evidence, the assessment of the impacts of food
assistance on pastoralists contained in this review is necessarily inconclusive. While the
difficulty of conducting research in the context of humanitarian crises should be recognized,
the inadequate quality of the included publications makes it impossible to reliably evaluate
the impacts of common types of humanitarian intervention targeting some of the worlds
most vulnerable populations. As such, it can prevent identification of potential harmful
impacts and other deficiencies of the provision of food assistance to pastoralist populations.
The findings of this review highlight, therefore, the necessity of future multidisciplinary
research and evaluation that can make valid inferences about the causal relationships
between food assistance and various aspects of pastoralist livelihoods. Such high-quality
research could be both qualitative and quantitative, but it should include experimental and
prospective cohort studies, as well as retrospective cohort designs which rely on validated
methods. New research should also disaggregate outcomes by age, gender, and mobility
patterns.
1 INTRODUCTION2
1.1 BACKGROUND
Pastoralists rely on coping and adaptation strategies that have historically allowed them to
achieve high levels of productivity, manage the hazards and unpredictability of life in the
marginal areas that they occupy and moderate the impacts of shocks (Butt et al., 2009;
Hesse and Pattison, 2013; Morton, 2006). Despite the unique suitability of these strategies
to their livelihoods, however, the food security of many pastoralist populations especially in
Africa has increasingly been threatened. For example, the 2006 drought in East Africa and
the Horn of Africa (affecting Djibouti, Eritrea, Ethiopia, Kenya and Somalia) caused severe
shortages of food across the pastoralist corridor that contains the largest grouping of
pastoralists in the world (Mkutu, 2001). The drought affected a reported 11 million people,
including many pastoralists (FAO, n.d.; ODI, 2006). Famine returned to the region in 2011
and 2012; in Somalia, where 80 percent of the population is pastoralist (Central Bank of
Somalia, n.d.) and where the food crisis was most acute, an estimated 260,000 people died,
another 750,000 were affected and 3.3 million people were in need of immediate life-saving
assistance. Altogether, 13 million people were affected by the food security crisis in the Horn
of Africa (Maxwell et al., 2014: 5). Similarly, 800,000 people including many pastoralists
faced severe food insecurity and another 800,000 were moderately insecure during the 2005
3
and 2006 Niger food crisis (Aker, 2008: 7). While such crises have primarily affected African
pastoralists, they have also been reported elsewhere. For example, 8,000 Mongolian
pastoralists affected by dzud (harsh winter conditions) received food assistance in 2010
(Action Against Hunger, 2011).
Crises faced by pastoralists have increased in frequency and intensity in recent decades.
Because of climate change, political marginalization, loss of grazing land, restrictions on
mobility and other detrimental policies pursued by national governments, some pastoralists
are no longer able to overcome them without outside assistance (Markakis, 2004: 4). Such
assistance has taken many forms, but nutrition and food security have been the priorities of
most humanitarian interventions, which have usually involved direct provision of food in-kind
to affected populations.
Definition of common terms included in this review

Pastoralism
Pastoralists are people whose livelihoods rely primarily on livestock keeping and who occupy marginal areas
incapable of supporting an exclusively or primarily agriculturalist population. By extension, pastoralism is the
mode of production practised by such people. Such a basic definition should not be controversial; many specific
aspects of pastoralism are, however, contested.
First, pastoralist societies have developed complex modes of social organization and cultural patterns that are
intrinsically integrated with the pastoralist mode of production. In this respect, therefore, pastoralism is not just a
livelihood strategy, but a cultural identity and way of life (Markakis, 2004: 4). The erosion of the basis of
pastoralist livelihoods in recent decades may imply the decline of pastoralist cultures and societies but, even if
they survive, their direct link with the pastoralist mode of production may be weakened. If this is the case, the
decline of pastoralism as a set of livelihood strategies does not necessarily mean a corresponding decay of
pastoralist culture and society or the disappearance of people who identify themselves as pastoralists.
Therefore, inclusion of these non-economic practices radically expands the definition of pastoralism.
Second, the commonly accepted classification of livelihood systems presents a continuity of practices, from
pure pastoralism to exclusive reliance on agriculture or urban livelihoods. In this typology (as Krtli et al., 2015
note), differences between livelihood systems are effectively a matter of degree: of mobility (nomadic, semi-
nomadic, transhumant, sedentary), of average rainfall (desert, arid, semi-arid, sub-humid), of interaction with the
market economy (subsistence-based, market-driven), of attainment of modernity (traditional, modernizing,
modern) and, of course, of dependence on livestock and involvement in agriculture (Ibid.: 15). The temptation
that this classification offers is to restrict the definition of the term to pure pastoralism, i.e. a form of pastoralist
livelihood that is nomadic, subsistence-based, traditional and dependent solely on livestock keeping, and to
2
This Introduction section is based on the review protocol (Czuba and ONeill, 2016), which considers the issues outlined here in
greater detail.
3
Households or individuals can be classified as facing severe food insecurity if they have large food consumption gaps or have suffered
extreme loss of livelihood assets that will lead to such gaps. They can be classified as facing moderate food insecurity if they have food
consumption gaps or are only able to meet minimum food needs through accelerated depletion of livelihood assets (Rose, 2012).
The impact of in-kind food assistance on pastoralist livelihoods in humanitarian crises 2
develop other labels, such as agropastoralism, to refer to the other degrees. While this results in elegant
categorizations, however, the deep connections between modes of production that rely in some way on livestock
keeping are concealed.
For this reason, a broad definition of pastoralism is likely to offer a better lens through which to consider the
subject of this review. The review follows a definition that incorporates the characteristics common to a wide
range of pastoralist livelihood systems and is derived from a list compiled by Hesse and MacGregor (2006: 5). In
this definition, pastoralists are people who depend on livestock for a significant proportion of their food and
income, although they may also cultivate crops and carry out other economic activities. Their livestock can be
raised for both subsistence and market needs, but animals represent more than economic assets. Instead, they
constitute cultural, social and spiritual assets that define social identity. Livestock depend on natural pastures,
the management of which may require mobility of herds and pastoralists themselves. By extension, pastoralism
is the livelihood system of pastoralists.
This broad definition includes a wide range of livelihood systems and intentionally makes no distinction between
different degrees of pastoralism, as listed above. It is also likely to capture more food assistance interventions and,
therefore, to provide a fuller understanding of the impacts of food assistance on diverse livestock-keeping
populations. It is, however, necessary to be mindful that such a broad understanding of pastoralism is not
universally accepted and that the use of the term in some sources identified during the search process will either
refer exclusively to the pastoralist mode of production or to pure pastoralism. Furthermore, the protocol highlighted
the need for the research team to be aware of the possible presence in the searched publications of commonly held
biases against pastoralists brought about by their economic and political marginalization (Ibid.: 15).
Livelihoods
Livelihoods are defined as the capabilities, assets (including both material and social resources), and activities
required for a means of living (Scoones, 1998: 5). Therefore, pastoralist livelihoods represent:
capabilities such as knowledge, skills and training required to raise and properly care for livestock and
navigate territories which pastoralists occupy. This includes locating appropriate grazing areas and other
sources of food and interacting with other pastoralist and non-pastoralist populations;
assets such as livestock, access to grazing and agricultural land, tools and social organization conducive to
pastoral livelihoods;
the activities that these capabilities and assets make possible, i.e. livestock keeping and, where applicable,
seasonal agricultural production and waged employment.
Food assistance
Food assistance refers to any direct food or food procurement transfer to food-insecure individuals or
households for the purpose of increasing the quality and/or quantity of food consumed or, in some cases, as a
form of income transfer. Food assistance can be provided in the context of low food availability or low food
access. It can be provided by organizations such as national governments, non-government organizations
(NGOs) and international organizations or their specialized agencies. Respondents can receive it on a short-term
basis or, in the context of protracted emergencies, for extended periods of time. The most commonly used type
of humanitarian assistance to pastoralists has been the direct provision of food in-kind, which can be either
short- or long-term and unconditional or tied to assets or work, as in the case of the World Food Programmes
Food Assistance for Assets programmes (WFP, n.d.). In addition, recent innovations in food provision in
humanitarian settings have involved the use of cash transfers and vouchers (ECHO, 2013; Harvey and Bailey,
2011; Manley et al., 2011; Smith et al., 2011; Doocy and Tappis, 2015). Alternatively, the price of food in an
affected area can be subsidized (Levine, 2008: 2). The term food assistance has largely replaced food aid in
the language used by the international community (Clay, 2010; Harvey et al., 2010).
Crisis and humanitarian crisis
A crisis or shock is defined as an event or events that destabilize the livelihoods of a population. A humanitarian
crisis is defined as an event or events that pose an exceptional and generalized threat to the health, safety, well-
being or subsistence of a population. This phenomenon can have either a slow or rapid onset, be acute or
chronic, and be either natural in origin (e.g. drought) or man-made (e.g. conflict) (or its causes can be
attributable to both natural and man-made factors) (Chi et al., 2015).
1.2 RATIONALE AND PURPOSE

Humanitarian crises have affected millions of pastoralists and have led to a range of
interventions that have mobilized considerable resources in an attempt to alleviate suffering
and improve the viability of pastoral livelihoods. However, evidence on the effects that these
interventions have had on pastoralists particularly the extent to which they have achieved
their stated objectives is, at best, fragmentary. Reported assessments of the effectiveness
of interventions that specifically target pastoralists have generally been negative, but the
evidence tends to be limited to selective case studies that should not be used to draw
general inferences (e.g. Fratkin, 2001; Fratkin et al., 2004; Hesse and Thbaud, 2006;
Markakis, 2004). There is a number of more rigorous evidence syntheses that assess
humanitarian interventions (e.g. HEP nutrition review, forthcoming; Harvey and Bailey, 2011;
Manley et al., 2011; Smith et al., 2011; Doocy and Tappis, 2015; Bellos et al., 2010; London
School of Hygiene and Tropical Medicine, 2013; Crumlish and ORourke, 2010; HEP mental
health review, forthcoming; Spangaro et al., 2013; Vu et al., 2014; HEP child protection
review, forthcoming; Parkinson, 2009; Yates et al., forthcoming), but they do not specifically
consider the impacts of these interventions on pastoralists, whose distinctiveness suggests
that the effects of interventions targeting them are likely to differ significantly from other
populations. There is, therefore, a great need for systematic evidence of the impacts of
humanitarian interventions on pastoralists to inform policy and to suggest a future research
or programme monitoring agenda. It is impossible to provide comprehensive evidence of all
impacts in the context of a single review. Because the provision of food assistance to
populations experiencing high levels of food insecurity has long been the dominant form of
relief assistance, it follows that it should be the subject of the first evidence synthesis
addressing the impacts of humanitarian interventions on pastoralists.
The purpose of this review is, therefore, to use evidence synthesis methods (specifically the
scoping study approach, which uses rigorous and transparent methods to comprehensively
identify and analyze all relevant literature pertaining to a research question, but can also
address broader topics and a wider range of study designs than a systematic review), to
identify, synthesize, evaluate and estimate both the short- and long-term effects that the
provision of food assistance in the context of humanitarian crises has had on pastoralists
and their livelihoods. The review considers all potential impacts of in-kind food assistance on
all pastoralist populations that have been affected by humanitarian emergencies in the
4
period since 1967 (when the Food Aid Convention was negotiated). Where possible, this
investigation includes evaluating the intended impacts of the interventions, i.e. their
effectiveness and efficiency in providing food assistance to those in need and the use of
food by their beneficiaries for the purposes intended. Equally important is an examination of
other, often unintended, effects of the provision of food assistance. The literature on
pastoralist societies suggests a number of potential unintended consequences, including
long-term dependency on external provision of food (Abbink et al., 2014; Bassi, 2010;
Boulton, 2012; Helland, 1998; Levine, 2010), exclusion of the poorest (Bishop and Hilhorst,
2010; Maxwell et al., 2014: 28) and undermining of existing social support networks (Sharp,
1999). Other impacts of food assistance might include shifts in livelihood strategies and
mobility patterns (including sedentarization) (Fratkin et al., 2004; Salzman, 1980), new
dynamics concerning assets (including sale, acquisition of assets and asset types) and
income (Abebe et al., 2008; McPeak et al., 2012; ODI, 2006), as well as changes related to
intra-household relations, educational attainment, gender roles (Fratkin, 1992; Krtli, 2009;
Pantuliano, 2002), conflict and insecurity (Oba, 1992) and denudation of the environment
(Ibid.; Blackwell, 2010; McCabe, 1990).
The feasibility of investigating these possible effects is dependent on the availability of

evidence identified during the systematic search process. Where possible, this review aims
to make attributable and, at a minimum, contributable inferences about the provision of
food assistance in the context of humanitarian interventions and various aspects of
pastoralist livelihoods.
1.3 RESEARCH QUESTIONS

The review aims to identify, synthesize, evaluate the strength of evidence and estimate both
the short- and long-term effects that the provision of food assistance in the context of
humanitarian crises has had on pastoralist livelihoods from 1967 onwards. In order to
address this primary objective, a number of research questions were posed to the reported
and available literature:
Q1: In what ways have pastoralist livelihood strategies changed since 1967 (and to what
extent have non-pastoralist livelihood strategies supplanted them)? What is the potential
causal relationship between these changes and food assistance?
Q2: In what ways have pastoralist asset dynamics changed? What is the potential causal
relationship between these changes and food assistance?
Q3: In what ways have income dynamics changed? What is the potential causal
4
The reasons for this being the scope of the review are discussed in detail in the review protocol.
Q4: In what ways have mobility patterns changed? What is the potential causal
Q5: What types of in-kind food assistance interventions have been offered to pastoralist
populations in the context of humanitarian crises? How have these interventions, the
nature of the strategies, the distribution of food assistance and its duration changed over
time?
Q6: In what ways do pastoralists use the food they receive? Do they consume it, or use it
as livestock feed or as a commodity to sell or barter?
Q7: In what ways has access to food and non-food items in pastoralist areas changed,
including in relation to markets? What is the potential causal relationship between these
changes and food assistance?
Q8: What household-level shifts have taken place among pastoralist populations? What
is the potential causal relationship between these changes and food assistance?
Q9: What individual-level shifts have taken place among pastoralist populations? What is
the potential causal relationship between these changes and food assistance?
Q10: What shifts have taken place in the social relations of pastoralist populations,
including in relation to social support networks? What is the potential causal relationship
between these changes and food assistance?
Q11: In what ways have the security conditions within which pastoralist livelihoods take
place changed? What is the potential causal relationship between these changes and
food assistance?
Q12: What are the populations and regions affected by food assistance interventions
targeting pastoralists? What are the commonalities and variations between these
populations and regions?
Population, intervention, control and outcome (PICO) parameters

1. Population
This evidence synthesis considers all pastoralist populations, as previously defined. As specified in the review
protocol, it also aims to investigate the commonalities and variations between the populations (primarily in terms
of livelihood patterns, including migration practices, but also, where relevant, governance systems
and interactions with non-pastoralist populations) in order to facilitate sub-grouping and apt comparison, and
thereby make inferences regarding the relationship between the provision of food assistance and livelihood
changes.
2. Interventions/issues
The review examines in-kind food assistance provision during or after humanitarian crises, as previously defined.
To this end, it aims to identify humanitarian crises that have affected pastoralist populations. It also provides an
overview of food assistance interventions targeting pastoralist populations that have been undertaken in the
wake of these crises. Specifically, it aims to identify the issues, types of intervention and the specific ways in
which they have targeted pastoralist populations.
3. Comparator/context
The review assesses the changes that the livelihoods of pastoralist populations have undergone in the wake of
in-kind food assistance provision, and the changes to the context where they live. Where data exists, it aims to
compare the changes experienced by any relevant population to pastoralist populations (especially ones living in
similar contexts) who have not received in-kind food assistance, pastoralist populations who have received other
types of food assistance (including cash transfers and vouchers), pastoralist populations who have been affected
by different kinds of humanitarian crises (e.g. slow- as opposed to rapid-onset ones), pastoralist populations who
have not experienced comparable humanitarian crises, and to the historical experiences of the same population.
4. Outcomes
The review investigates potential impacts (detailed below) of in-kind food assistance on pastoralist populations.
Whenever possible, it aims to disaggregate the evidence of possible impacts by age, gender, mobility patterns,
populations (such as ethnic groups) and regions of interest, the type of food assistance interventions, the type of
humanitarian crisis in question (slow- or rapid-onset; natural in origin, man-made or a combination) and other
relevant attributes.
The review also considers the extent to which food assistance interventions are potentially responsible for
specific changes in pastoralist livelihoods. These changes may include:
shifts in livelihood strategies, especially changed preferences for specific assets animal species or (among
agropastoralists) access to land for seasonal agricultural production or diversification of strategies to include
non-pastoral production and employment, as well as dependency on food assistance
asset dynamics, including sale and acquisition of assets and asset types
income dynamics, including new sources of income (e.g. from non-pastoral employment and remittances) and
(re)distribution of income within populations
shifts in mobility patterns, including sedentarization and urban migration
shifts in social relations, including in social support networks
household-level shifts, including changes in household size, household income, distribution of assets, the role
of household members in specific aspects of livelihood strategies (determined by age, gender and other
attributes) and household splitting
individual-level shifts, such as acquisition of new skills, educational attainment, gender roles and health or
disease outcomes
conflict and insecurity.
The review seeks to identify the specific ways in which pastoralist livelihoods and related phenomena have
changed and to determine the extent to which they can be associated with food assistance interventions in
response to humanitarian crisis. In particular, it aims to address the possible causal mechanisms that may link
specific food assistance interventions and livelihood changes.
2 METHODOLOGY
Because of the nature of anticipated evidence, the relatively broad character of the research
objective and time constraints, a full systematic review of evidence was not feasible. Instead,
5
the scoping study methodology was selected for the review. While more adaptable and
applicable to broader topics and a wider range of study designs than a systematic review,
the scoping study methodology is still underpinned by the requirement of rigorous and
transparent handling of all stages of the research process, which should be documented to
enable replication (Arksey and OMalley, 2005: 22; DiCenso et al., 2010: 20). This review
adds to traditional scoping study methods by assessing study quality, which is not typically
included in this form of evidence synthesis.
In all, the review seeks to summarize evidence on the impact of in-kind food assistance on
pastoralists livelihoods during and after a humanitarian crisis. It aims to achieve this by:
systematically identifying all available evidence in academic and grey literature sources
comparing and contrasting the effects of assistance delivered (by population, assistance
type, etc.)
synthesizing identified data and concepts to generate summary statements on available
evidence
assessing the strength of evidence, as appropriate, for the body of identified studies
identifying gaps in the current evidence base and further commenting on future research
needs in this space.
The research process involved the following steps:

protocol development
implementation of the search strategy
document screening
assessment of the strength of evidence and causal inferences in included publications
data extraction and evidence synthesis.
2.1 PROTOCOL DEVELOPMENT

The purpose of the protocol (Czuba and ONeill, 2016) was to clearly describe the proposed
research methodology. The protocol set out the theoretical background to the study, its
purpose and objectives, the research questions and PICO parameters, and the review
methodology. The protocol was peer-reviewed and amended before commencement of the
subsequent steps in the research process.
2.2 IMPLEMENTATION OF THE SEARCH STRATEGY

Following the methodologically rigorous scoping review approach proposed by Arksey and
OMalley (2005: 48) to conduct a systematic search, comprehensive literature searches were
conducted in June 2016 in the following databases: Medline, CAB Abstracts, Cochrane
Database of Systematic Reviews, ARTFL-FRANTEXT, International Bibliography of the
Social Sciences, LISA, LILACS, IDEAS, Web of Science Core Collection, PAIS International,
Africa Wide NiPAD and Google Scholar. Searches were designed by a medical librarian
(Ana Patricia Ayala), using a combination of controlled vocabulary (MeSH, Emtree) and free
text terms. The strategy was modified as necessary for individual databases. The search
strategy was limited to English and French, and to sources published between January 1967
and June 2016. Search terms included: pastoral*, domestic livestock, humanitarian
6
intervention, relief, aid, crisis management, in-kind, assist* and animal husbandr*.
5
The reasons for this choice are discussed in the review protocol (Czuba and ONeill, 2016).
6
An asterisk indicates that a word that has been truncated in order to search for variants.
Targeted searches for grey literature (i.e. difficult to locate or unpublished material) by
searching websites that captured the topic were also undertaken and included a large
number of websites, including those of the following organizations and institutions: Addis
Ababa University, African Development Bank, Centre for International Development, Danish
International Development Agency, Deutsche Gesellschaft fr Internationale
Zusammenarbeit, Feinstein International Center, FHI 360, Humanitarian Aid and Civil
Protection Department (ECHO) of the European Commission, Institute for Development
Studies at the University of Nairobi, Institute for Global Health at University College London,
Institute of Development Studies at the University of Sussex, International Institute for
Environment and Development, OpenGrey, Swedish International Development Agency,
World Health Organization and others. The sources and search strategy strings are detailed
in Appendix 1.
2.3 DOCUMENT SCREENING

Inclusion criteria, derived from the PICO parameters, were applied incrementally over the
course of two screening rounds to identify potentially relevant publications. Citations were
imported directly into EndNote (Thomson Reuters, Philadelphia, PA, USA). Subsequently,
citations were imported directly into systematic review software Covidence (Veritas Health
Innovation Ltd., Deerfield, IL, USA) for title and abstract relevance screening and data
7
characterization of full articles.
Round 1 (R1): Title and abstract review for potential relevance

In the first round of reviews, titles and abstracts of potentially relevant publications identified
during the searches were screened by two researchers (Karol Czuba and Tyler J. ONeill)
independently for further consideration if they satisfied the following inclusion criteria:
Does the study report on a pastoralist population (Appendix 2)? (Yes/No/Unclear)
Has the reported population been affected by a humanitarian crisis in the period since
1967 or can it be compared with a population that has been affected by a humanitarian
crisis or emergency (a comparative population is desired but not necessary for inclusion,
given the suspected limited availability of data)? (Yes/No/Unclear)
Did the response to the crisis or emergency include the provision of in-kind food
assistance? (Yes/No/Unclear)
Round 2 (R2): Full text review for potential relevance

Studies that did not report no in any of the inclusion criteria from R1 were included in R2. In
R2, the researchers re-applied inclusion criteria to potentially relevant publications from R1.
Studies included at this stage were also required to satisfy the following criterion:
Does the study report evidence of potential impact of food assistance on pastoralist
populations? (Yes/No/Unclear)
Studies that failed to meet the inclusion criteria during R1 and R2 (i.e. failed to respond yes
to all inclusion criteria) were excluded from further consideration. All inclusions and
exclusions were recorded in Covidence.
7
Software used in the review differs from software proposed in the review protocol, which envisioned the use of RefWorks and
DistillerSR, instead of EndNote and Covidence respectively. After exploration of ease of use and integration of identified
references across platforms, the research team selected EndNote and Covidence. Both software alternatives are comparable in
function, but EndNote integrates better with Covidence, whereas RefWorks takes increased time and effort, with greater risk of
error when importing in to DistillerSR.
2.4 EVIDENCE ASSESSMENT

The strength of evidence in the individual included publications and the overall strength of
the body of evidence were appraised using the DFID Assessing the Strength of Evidence
8
note (DFID, 2014).
The assessment criteria for individual publications identified in the DFID note are specified in
Figure 2.1. The note specifies the following categories to which publications can be
assigned, depending on the strength of evidence that they report:
high the publication comprehensively addresses multiple principles of quality (as
recommended in the note, publications which satisfy this criterion are identified with the
symbol ).
moderate there are some deficiencies in attention to principles of quality (identified with
).
low there are major deficiencies in attention to principles of quality (identified with ).
The note recommends that strength of evidence should be clearly indicated when the
findings reported in a publication are discussed. This review follows this recommendation; all
in-text citations of included publications are accompanied by an arrow symbol which
indicates the strength of evidence.
The review also follows the criteria for assessment of bodies of evidence developed in the
note:
the (technical) quality of the studies constituting the body of evidence (or the degree to
which risk of bias has been addressed):
high the large majority of reviewed publications report evidence of high strength,
demonstrating adherence to the principles of research quality
moderate approximately equal numbers of reviewed publications are of a high,
moderate and low quality, as assessed according to the principles of research quality
low the large majority of reviewed publications report evidence of low strength,
demonstrating significant deficiencies in adherence to the principles of quality;
the size of the body of evidence:

large
medium
small;
the context in which the evidence is set:

global
context-specific;
the consistency of the findings produced by studies constituting the body of evidence:
consistent a range of publications points to identical or similar conclusions
inconsistent (contested) at least one publication directly refutes or contests the
findings of other publications carried out in the same context or under the same
conditions
mixed publications based on a variety of different designs or methods, applied in a
range of contexts, have produced results that contrast with those of other publications;
8
The DFID guidelines were applied instead of GRADE, for two reasons. First, DFID which has provided funding for this review
requires that the guidelines should be comprehensively applied in the production of DFID evidence papers. Second, GRADE is
designed primarily for controlled trials. Nearly all the included publications are case studies, and it is therefore not appropriate to apply
the GRADE tool to them (they are descriptive studies, not observational). During development of the protocol, the research team
anticipated finding cohort or cross-sectional studies, which would have made the use of GRADE appropriate. Following document
screening, however, we made a post hoc decision to apply a more appropriate tool given the body of evidence we identified.
the overall strength of the body of evidence:

very strong high-quality body of evidence, large in size, consistent and contextually
relevant
strong high-quality body of evidence, large or medium in size, highly or moderately
consistent and contextually relevant
medium moderate-quality studies, medium size evidence body, moderate level of
consistency
limited moderate to low-quality studies, medium-size evidence body, low levels of
consistency
no evidence.
Figure 2.1: DFID strength of evidence assessment criteria for individual

publications
Principles of Associated questions

quality
Conceptual Does the study acknowledge existing research?
framing
Does the study construct a conceptual framework?
Does the study pose a research question or outline a hypothesis?
Transparency Does the study present or link to the raw data it analyses?
Does the study declare sources of support/funding?
Does the study demonstrate why the chosen design and method are well suited to the
research question?
Cultural Does the study explicitly consider any context specific cultural factors that may bias the
sensitivity analysis/findings?
Validity To what extent does the study demonstrate measurement validity?
To what extent is the study internally valid?
To what extent is the study externally valid?
To what extent is the study ecologically valid?
Reliability To what extent are the measures used in the study stable?
To what extent are the measures used in the study internally reliable?
To what extent are the findings likely to be sensitive/changeable depending on the
analytical technique used?
Cogency Does the author signpost the reader throughout?
To what extent does the author consider the studys limitations and/or alternative
interpretations of the analysis?
Are the conclusions clearly based on the studys results?
Source: DFID (2014: 14)
For a summary of the findings, see Figure 4.8.
2.5 DATA EXTRACTION AND EVIDENCE SYNTHESIS

Data abstracted from the included publications is presented in Figure 4.1. The subsequent
sections describe the publications in greater detail and synthesize the evidence about the
impacts of food assistance provided to pastoralists in the context of humanitarian
emergencies.
3 RESULTS
3.1 SUMMARY OF REVIEWED PUBLICATIONS

The search initially identified a total of 23,424 titles from academic databases and an
additional 1,442 titles from grey literature sources (Figure 3.1; Appendix 1). After automatic
de-duplication, 17,108 titles (68.8 percent of the total) were moved forward for title and
abstract screening. Upon further assessment of documents, an additional 1,713 were
manually identified as duplicates and removed, resulting in a total of 15,395 titles and
abstracts (65.7 percent) included in R1. In all, 72 publications (0.3 percent) satisfied R1
criteria and qualified for further assessment in R2. Upon retrieval of full documents, a further
48 (0.2 percent) were deemed to be non-relevant as they did not report the provision of food
assistance to pastoralists during a humanitarian crisis. In total, 24 documents (0.1 percent)
were further assessed for data abstraction and assessment of the strength of evidence.
Figure 3.1: Flowchart demonstrating the process of identification through to

selection of documents included in the review
Records identified Additional records identified

through academic database through grey literature
searching (n=23,424) sources (n=1,442)
Identification
Duplicate records
removed (n=7,758)
Records after
manual de-duplication
(n=17,108)
Further duplicates
removed (n=1,713)
Screening for eligibility
Records screened (R1):

Titles and abstracts
(n=15,395)
Non-relevant
documents removed
(n=15,323)
Records screened (R2):

Full text documents
(n=72)
Non-relevant
documents removed
Included
(n=48)
Documents included in review

(n=24)
3.2 INCLUDED PUBLICATIONS

Records identified through academic databases and grey literature sources were restricted
to English and French; only those written in English met study inclusion criteria. The majority
of included publications (54 percent) have been published since 2000 (range: 1983 to 2015).
They report provision of food assistance that took place between 1920 and 2012 (Figure
4.1). Most publications do not explicitly identify their research design; instead, they were
classified by the research team based on the information they contained. Of reported
research types, the majority of publications are primary studies (n=20); the remaining
publications are secondary studies (n=2) or do not contain information that would allow
identification of the research type (classified as unclear; n=2). The research design of the
two secondary studies is classified as other review (n=2). The research design is
impossible to identify in one publication (unclear; n=1). The remaining publications are
observational (n=21). The majority of publications are case studies (n=19); the case studies
rely on qualitative (n=10), quantitative (n=4) and mixed (n=5) methods. Additionally, one
study has a quasi-experimental design akin to a natural experiment, although it employs
qualitative methods (n=1). There is also a quantitative case control study (n=1) among the
included publications. Three studies are classified as other (n=3). One compares the
experiences of two countries (Ethiopia and Sudan; quantitative). Another summarizes case
studies published elsewhere (qualitative). The third is a news story from the popular press
(qualitative). With the exception of the news story, the included publications are academic
and/or peer-reviewed (n=23).
As recommended by DFID (2014), the study characteristics of each included publication are
stated in-text (together with assessed strength of evidence indicated by arrow symbols)
when the evidence reported in the publication is discussed. The review uses the following
abbreviations to identify the study characteristics:
Design:
P: primary
S: secondary
O: observational
QE: quasi-experimental
OR: other review
UN: unclear
CS: case study
CC: case control
NE: natural experiment.
Methodological approach:
Qual: qualitative
Quant: quantitative
M: mixed.
Most of the included publications report food assistance provision in East Africa and the Horn
of Africa (Sudan, Eritrea, Djibouti, Ethiopia, Somalia, Uganda and Kenya) (n=20). Studies from
Algeria (n=1), Niger (n=1) and Mongolia (n=2) are also included. The following African
pastoralist ethnic groups are reported: Beja (n=2; in one publication, the Amarar/Atmaan and
Bishariyyn groups of the Beja are specifically identified), Gabbra (n=1), Oromo (n=2; the
Borana sub-group is specifically identified in one publication), Pokot (n=2), Rashaida (n=1),
Saharawis (n=1), Somalis (n=3) and Turkana (n=5). Additionally, studies report unidentified
pastoralist populations in Djibouti (n=1), Eritrea (n=1), Ethiopia (n=2), Kenya (n=5), Niger
(n=1), Somalia (although these are presumably Somalis; in accordance with evidence
synthesis conventions, because the population is not explicitly identified it is not reported here;
n=3) and Sudan (n=2). Mongolian herders are identified as recipients of food assistance in
both publications from Asia. The most frequently reported mobility pattern is nomadism (n=14);
agropastoralism is reported twice (n=2). Ten publications fail to report the mobility patterns of
reported populations. Only one publication presents an outcome disaggregated by gender and
only five explicitly note that both men and women are included in the observed populations.
Finally, studies report a large range of individuals potentially receiving food aid in response to a
humanitarian crisis (range: 6,000 to an estimated 20 million).
The majority of study populations have experienced drought (n=21) as the reported
humanitarian crisis. Less frequently, but often in conjunction with drought, the publications
report conflict (n=3), famine (n=4), flood (n=2) and disease (n=1). Both studies from
Mongolia report dzud (n=2) as the humanitarian crisis affecting the population of interest.
One publication fails to report sufficient information regarding the reported humanitarian
crisis. Unconditional provision of food in-kind is the most frequently reported modality of food
assistance (n=13); food-for work (n=2) and food-for-livestock exchange (n=1) are also
documented. Nine studies fail to detail the modality of provision of food assistance. Half of
included studies (n=12) report the provider of food aid. The categories of providers include
national governments (n=4), the United Nations and its specialized agencies (n=5), NGOs
(n=7) and religious bodies (n=2). Several publications report more than one provider. The
duration of aid provision varies, when reported (n=8), from one to 20 years.
3.3 OUTCOMES REPORTED BY INCLUDED PUBLICATIONS

The impacts of the provision of in-kind food assistance during a humanitarian crisis are
reported in all included publications. Some publications report more than one outcome
measure.
Based on abstracted data, the review team has organized outcomes into six thematic areas: (i)
changes in livelihood strategies and asset and income dynamics (n=11; research questions 1,
2 and 3); (ii) access, use and dependency on food assistance (n=7; research questions 5, 6
and 7); (iii) changes in social relations (n=4; research question 10); (iv) security and
governance (n=4; research question 11); (v) mobility patterns (n=9; research question 4); and
(vi) household- and individual-level socio-demographic shifts (n=7; research questions 8 and
9). Some publications report more than one outcome and are included in multiple themes.
Given the limited number of relevant publications included in the review, the research team
was unable to disaggregate evidence beyond geographical region (i.e. Horn of Africa, Africa
other, Mongolia). The findings on livelihood changes in Kenya, which are reported by multiple
sources (but report mixed evidence), offer a partial exception. Furthermore, for the same
reason and also because publications do not contextualize outcomes by age, gender or
mobility patterns, we were unable to disaggregate these characteristics.
3.4 ASSESSMENT OF THE STRENGTH OF EVIDENCE AND

CAUSAL CLAIMS
The publications have been evaluated with respect to standards outlined in the DFID
Assessing the Strength of Evidence note (DFID, 2014). These standards correspond to the
questions presented in Figure 2.1. Figure 4.8 reports the extent to which the publications
conform to the standards.
The publications generally satisfy the conceptual framing standards. Seventeen publications
acknowledge existing research and 18 pose a research question (or questions) or outline a
hypothesis (or hypotheses). The research team has interpreted the term conceptual
framework broadly to include publications (n=20) that make theoretical claims of any kind.
The transparency standards are followed less consistently. Seventeen publications present
or link to the raw data that they analyze (this includes both formal reporting of results
obtained from quantitative analysis of datasets and more offhand references to findings
obtained from e.g. interviews and focus groups in qualitative sources); however, only one
declares its sources of funding.
The majority of publications do not satisfy the appropriateness, cultural sensitivity, validity or
reliability standards. No publication justifies the choice of the research design and methods or
their appropriateness to the research questions. Cultural sensitivity of the research undertaken
is also not considered by any publication. The six publications that rely on observational
quantitative methods appear to be internally valid and use valid measures, but these issues
are not explicitly addressed. The remaining publications do not demonstrate measurement
validity in any way; their internal validity is also unclear. Based on information contained in
these publications, it is impossible to determine whether the validity and reliability standards
are met. Given that these criteria are not satisfied, none of the included publications provides a
high strength of evidence for the criteria of internal measurement validity.
More publications satisfy at least some of the cogency standards. All publications other than
the three very short ones (n=21) signpost the reader throughout the text. In most cases
(n=18), conclusions appear to be based on the results reported in the publications. Only one
publication, however, explicitly addresses the limitations of the reported research project.
Overall, based on the publications conformity with the DFID standards, this review does not
consider the strength of evidence reported in any of the publications to be high. The strength
of evidence of 13 publications can be classified as moderate, while there are 11 publications
of low strength of evidence.
Because of the large number of low-strength-of-evidence studies among the included

publications and the lack of any high-strength studies, the quality of the body of evidence is
classified as low, based on the DFID assessment criteria. The size of the body of evidence
(n=24) can be classified as medium. The distribution of the publications geographic focus
roughly corresponds to the geographic distribution of pastoralists in the world: most of them
report findings from East Africa and the Horn; the remaining ones address the provision of
food assistance in other parts of Africa and in Mongolia. For this reason, the context of the
body of evidence can be classified as global, although the possibility of reporting bias should
be noted because all the included sources are in English. As the subsequent section
demonstrates, the publications report different and, at times, seemingly contradictory
impacts of the provision of food assistance on specific aspects of pastoralist livelihoods.
Because they report findings from multiple contexts, however, the consistency of the findings
is classified as mixed, rather than inconsistent (or contested).
The characteristics of the body of evidence indicate that its overall strength should be
classified as limited: the publications have low to moderate strength of evidence; the size of
the body of evidence is medium; the levels of consistency are relatively low. The
assessment of the overall strength of the body of evidence is summarized in Figure 3.2.
Figure 3.2: Overall strength of the body of evidence
Quality of the body of evidence Low

Size of the body of evidence Medium
Context of the body of evidence Global
Consistency of the findings Mixed
Overall strength of the body of evidence Limited
3.5 NARRATIVE SYNTHESIS OF RESULTS

This section is organized according to the six themes introduced earlier in the Results
section.
Changes in livelihood strategies and asset and income dynamics

This theme corresponds to three of the research questions:
Q1: In what ways have pastoralist livelihood strategies changed since 1967 (and to what
extent have non-pastoralist livelihood strategies supplanted them)? What is the potential
causal relationship between these changes and food assistance?
Q2: In what ways have pastoralist asset dynamics changed? What is the potential causal
Q3: In what ways have income dynamics changed? What is the potential causal
Changes in livelihood strategies
These questions address some of the most important aspects of the potential impacts of
food assistance on pastoralist livelihoods. The evidence on these impacts obtained from the
included publications is, however, very limited.
The best available evidence about the impact of food assistance on pastoralist livelihoods is
from McCabe (1990; P, QE, Qual, NE, ), who reports that the Ngilukumong, Ghiyapakuno,
Ngikamatak and Ngibocheros sections of the Turkana in Kenya that received food
assistance during the 197981 drought witnessed a breakdown of indigenous coping
strategies for drought. No similar development occurred among the Ngisonyoka of southern
Turkana, who did not receive food assistance. McCabe outlines a causal chain in which the
attraction of free food led to the settling of pastoralist households near relief distribution
centres; the presence of large numbers of pastoralists and their herds around the relief
centres in turn contributed to denudation of the environment, which forced livestock to forage
on sparser and less nutritious plants than they would have away from the settlements. (The
negative impact of food assistance on the environment is also mentioned, albeit without
providing any details, by Pantuliano, 2007 [P, O, Qual, CS, ].) Inevitably, livestock losses
through starvation or disease followed, undermining the basis of pastoralist livelihoods. The
relief centres, McCabe observes, were, to some extent, creating their own clients. He notes
differences between the southern (Ngisonyoka) and northern (Ngilukumong, Ghiyapakuno,
Ngikamatak and Ngibocheros) Turkana that could account for some of the variation;
nonetheless, the stark contrast between the two populations strongly indicates that the
provision of food assistance had negative consequences for the livelihoods of the Turkana
who received it between 1979 and 1981.
The long-term consequences and external validity of developments reported by McCabe are
not clear. According to Bersaglio et al. (2015 [P, O, Qual, CS, ]), a small-scale food
assistance intervention in Turkana contributed to movement away from pastoralist
livelihoods among the beneficiaries while simultaneously improving their quality of living.
Snow (1984 [UN, Qual, CS, ]) finds that food-for-work projects in Turkana encouraged a
shift towards cash crop agricultural production. Nangulu (2009 [P, O, Qual, CS, ]) claims
that livelihood strategies of the Kenyan Pokot, the southern neighbours of the Turkana, have
been undermined by food assistance. Evidence from the multiple publications which report
on Kenya is, therefore, mixed.
Outside Kenya, and similarly to Nangulu, Pantuliano (2002 [S, OR, Qual, CS, ]) reports
negative impacts of food assistance among the Beja in Sudan. The Overseas Development
Institute (2006 [P, O, Qual, CS, ]) claims that the humanitarian response to the crisis in the
Greater Horn of Africa has done very little, to date, to protect livelihoods. The assessment
that these publications offer suggests detrimental impacts of food assistance on pastoralist
livelihoods in a large number of settings.
Effective comparisons with pastoralist populations that have not received food assistance
with the exception of the Ngisonyoka reported by McCabe (1990) do not appear to be
possible given the lack of reported evidence. Abbink (1993) observes that the Suri of Lower
Omo Valley in Ethiopia were able to successfully weather the severe famine of 1985 and
preserve their livelihoods without access to food assistance, but their traditional coping
strategies were complemented by exploitation of gold and investment in automatic weapons.
Given the unique circumstances, the experience of the Suri does not offer generalizable
insights into the viability of pastoralist livelihood strategies in the context of humanitarian
crises in the absence of food assistance.
Changes in asset dynamics and income
Evidence on asset dynamics is even more limited and inconclusive. All of it is also drawn
from Kenya. Buchanan-Smith and Barton (1999 [P, O, M, CS, ]) report that the provision of
food assistance in Wajir in Kenya resulted in a reduction of livestock sales, which were
monitored by the World Banks Arid Lands Resource Management Project. Bush (1995 [P,
O, M, CS, ]) reports that provision of food assistance in Turkana led to a strengthening of
herd growth and simultaneously filled income gaps among recipients. The effects that these
interventions have had on pastoralist economies suggest that food assistance does not
necessarily always undermine pastoralist livelihoods.
The quality of the publications which report on this theme is moderate. The body of evidence
is small and context-specific. The findings reported by the publications are mixed. The
overall strength of evidence relating to this theme is medium.
Given the inadequate evidence (and its limited scope), the three research questions cannot
be answered conclusively. It appears, however, that the provision of food assistance has, at
least in some cases, contributed to the erosion of pastoralist livelihoods, although it might
also have enabled some pastoralists to hold on to their assets including livestock and
supported redistribution of income.
The results reported in this section are summarized in Figure 4.2.
Mobility patterns
This theme corresponds to the following research question:
Q4: In what ways have mobility patterns changed? What is the potential causal
The included publications present more evidence on changes in mobility patterns than any
other theme. Although a number of these publications have a low strength of evidence, the
consistency of the information that they provide suggests the veracity of their findings. Adow
(2008 [UN, O, Qual, CS, ]), Bersaglio et al. (2015 [P, O, Qual, CS, ]), McCabe (1990 [P,
QE, Qual, NE, ]) and Ngunjiri (1989 [P, O, Qual, CS, ]) in Kenya and Kilby (1993 [S, OR,
Qual, CS, ]) and Pantuliano (2002 [S, OR, Qual, CS, ] and 2007 [P, O, Qual, CS, ]) in
Sudan all observe increased rates of sedentarization among the beneficiary populations and
attribute it to the provision of food assistance. The most plausible causal connection
between food assistance and sedentarization is made by McCabe (and presented in the
section above on Changes in livelihood strategies). While it is clear that sedentarization
(and in some settings urbanization) has taken place among a large number of pastoralist
populations, the relationship between this process and food assistance is not always clear.
Other humanitarian interventions, development projects and broader economic changes may
also influence the mobility patterns of pastoralists.
The quality of the publications which report on this theme is low. The body of evidence is
small and context-specific. The findings reported by the publications are consistent. The
strength of evidence relating to this theme is limited.
Access to food assistance, use of food assistance and

dependency
This theme corresponds to three research questions:
Q5: What types of in-kind food assistance interventions have been offered to pastoralist
populations in the context of humanitarian crises? How have these interventions, the
nature of the strategies, the distribution of food assistance and its duration changed over
time?
Q6: In what ways do pastoralists use the food they receive? Do they consume it, or use it
as livestock feed or as a commodity to sell or barter?
Q7: In what ways has access to food and non-food items in pastoralist areas changed,
including in relation to markets? What is the potential causal relationship between these
changes and food assistance?
The included publications report the impacts of three types of food assistance: food provided
unconditionally (n=12), in return for work (n=2) and in return for livestock (n=1). The last
modality involved the exchange of livestock which, it was believed, would not survive a
drought, in return for food (Ngunjiri, 1989 [P, O, Qual, CS, ]). Because only three
interventions in which food assistance was distributed in return for assets or work are
reported, is not clear from the publications in what ways the provision of food assistance has
changed over time. (The types of food assistance provided are reported for each publication
in Figure 4.1.)
Problems related to the targeting of food assistance are noted by a number of publications.
Pantuliano (2002 [S, OR, Qual, CS, ]) reports that distribution of food assistance to the
Beja was captured by elites, who excluded some groups within the population. Similarly,
unequal distribution of food assistance caused controversy in Marsabit in Kenya (Lekapana,
2013 [P, O, M, CS, ]). Fernandez-Gimenez et al. (2012 [P, O, M, CS, ]) relate that in
Mongolia targeting of poor households was considered by many members of the local
population as unfair to those herders who worked hard to prepare for winter. Such
controversies appear to be widespread and are also noted in excluded publications (e.g.
Merten and Haller, 2009; Moris, 1988). In some cases (e.g. Pantuliano, 2002 [S, OR, Qual,
CS, ]), they are evidently the result of sub-standard targeting that excludes potentially
eligible beneficiaries; in others (e.g. Fernandez-Gimenez et al., 2012 [P, O, M, CS, ];
Lekapana, 2013 [P, O, M, CS, ]), they may reflect the inherent difficulties of effective and
equitable targeting in settings where universal distribution of food assistance is considered
unnecessary. The latter modality of distribution, in which all pastoralists in an area affected
by humanitarian crisis receive food assistance, appears to be rare and is reported only by
Bush (1995 [P, O, M, CS, ]). The amount of food assistance provided to pastoralists is
also not always sufficient. Cicalese et al. (2009 [P, O, Quant, CS, ]) report that food
distributed to Saharawis in Algerian refugee camps was insufficient and nutritionally
unbalanced, leading to widespread malnutrition. This deficiency of food assistance is not
reported by other publications.
Only one publication reports unintended use of food assistance: according to Snow (1984
[UN, Qual, CS, ]), provision of food assistance in Turkana led to an increase in alcoholism
as beneficiaries with adequate food supplies distilled corn that they received and sold it in
their communities.
A number of publications claim that the provision of food assistance has led to dependency
among the recipient population (Adow, 2008 [UN, O, Qual, CS, ]; Farzin, 1991 [P, O,
Quant, CS, ]; Lekapana, 2013 [P, O, M, CS, ]; Nangulu, 2009 [P, O, Qual, CS, ]). Most
of them are classified as reporting evidence of low strength and beyond noting their
authors impressions they do not provide any evidence for this claim. While it is
conceivable that populations who receive food assistance for extended periods of time may
come to depend on them, the dependency thesis is not supported by evidence.
The included publications do not report the ways in which access to food and non-food items
has changed in pastoralist areas.
Household- and individual-level socio-demographic shifts

This theme corresponds to two research questions:
Q8: What household-level shifts have taken place among pastoralist populations? What
is the potential causal relationship between these changes and food assistance?
Q9: What individual-level shifts have taken place among pastoralist populations? What is
the potential causal relationship between these changes and food assistance?
There is effectively no evidence in the publications on household- and individual-level shifts

such as changes in household size, distribution of assets, the role of household members in
specific aspects of livelihood strategies, acquisition of skills or educational attainment. (The
issue of household income is addressed in the section on livelihood changes and asset and
income dynamics.) The exceptions are the issue of gender roles and health outcomes.
Gender is addressed in one publication. Snow (1984 [UN, Qual, CS, ]) notes that there was
a growing tendency, encouraged (but not caused) by food assistance, for young Turkana
women to seek alternatives to pastoralist livelihoods. Given the lack of corroborating
evidence from other publications, the veracity of this finding cannot be confirmed.
The included publications include some evidence of a causal relationship between food
assistance and the health of its recipients. Two publications offer negative assessments of
the health impacts of food assistance. No author (2001 [P, O, Quant, CS, ]) finds that
because during famine in Ethiopia in 2000 food was distributed in feeding centres large
concentrations of susceptible people in one location could contribute to an increase in the
transmission of infectious diseases such as measles and diarrhoea. According to Kilby (1993
[S, OR, Qual, CS, ]), food assistance had little positive effect on human mortality levels in
Sudan, yet no clear reason for mortality was offered (i.e. due to food insecurity, disease or
otherwise). It is not clear if either of these claims is based on empirical evidence. On the
other hand, four publications report that the provision of food assistance has had positive
impacts on health outcomes of its beneficiaries. Bush (1995 [P, O, M, CS, ]), Buchanan-
Smith and Barton (1999 [P, O, M, CS, ]) and Taylor (1983 [P, O, Quant, CS, ]) all argue
that rates of malnutrition decreased because of access to food assistance. Bersaglio et al.
(2015 [P, O, Qual, CS, ]) observe a reduction in anxiety and enhanced spirituality (and,
thereby, increased well-being) among recipients of food assistance. While evidence is
limited and offers varying assessments of the association between food assistance and
health, the findings reported in Bush (1995 [P, O, M, CS, ]), Buchanan-Smith and Barton
(1999 [P, O, M, CS, ]), and Taylor (1983 [P, O, Quant, CS, ]) suggest that at least in
some settings and in the short term food assistance can achieve its primary goal of
addressing food insecurity.
is small and context-specific. The findings reported by the publications are mixed. The
Changes in social relations

Q10: What shifts have taken place in the social relations of pastoralist populations,
including in relation to social support networks? What is the potential causal relationship
between these changes and food assistance?
Evidence on shifts in social relations is limited. Snow (1984 [UN, Qual, CS, ]) reports that food
assistance may have led to the creation of new power structures and the emergence of power
brokers who challenge the authority of elders and the respect given to them. Pantuliano (2002
[S, OR, Qual, CS, ]) reports similar developments among the Beja in Sudan, where
problems related to the targeting of food assistance created a new class of food shaikhs,
which brought about a modification of the existing social structure and the traditional inter-
diwab (lineage) dynamics founded on the exchange of resources and reciprocity. On the other
hand, Bush (1995 [P, O, M, CS, ]) finds that the Turkana who received food shared it with
relatives and were thus able to maintain the fabric of their society.
It is difficult to draw conclusions from the scant evidence available. It appears that food
assistance can contribute to changes in social relations, but the nature of these changes is
likely to depend on the mode of its provision (for example, relying on local leaders can
encourage the emergence of new power structures and, conceivably, strengthen the role of
existing ones) and the local setting.
is small and context-specific. The findings reported by the publications are consistent. The
Security and governance

Q11: In what ways have the security conditions within which pastoralist livelihoods take
place changed? What is the potential causal relationship between these changes and
food assistance?
While insecurity in pastoralist areas is reported by a large number of publications (e.g.

Adams and Hawksley, 1989 [P, O, Qual, CS, ]; Bersaglio et al., 2015 [P, O, Qual, CS, ];
Cicalese et al., 2009 [P, O, Quant, CS, ]; Kilby, 1993 [S, OR, Qual, CS, ]; Overseas
Development Institute, 2006 [P, O, Qual, CS, ]), none of them makes a causal connection
between the provision of food assistance and the security situation in areas inhabited by its
beneficiaries. Adams and Hawksley (1989 [P, O, Qual, CS, ]) note that during the crisis in
Darfur during the 1980s, food assistance was believed to have saved a large number of
lives. This finding is not surprising given the apparent impacts of food assistance on health
outcomes (reported in the section above on Household- and individual-level socio-
demographic shifts), but is not corroborated by other publications.
A related issue that was not addressed in the research questions developed for the review
protocol relates to the impact that food assistance may have on governance relations in
pastoralist areas. As the findings of Pantuliano (2002 [S, OR, Qual, CS, ]) and Snow (1984
[UN, Qual, CS, ]) indicate, food assistance may contribute to altering local power structures
through the creation of power brokers used by providers to support its distribution. In a later
publication, Pantuliano (2007 [P, O, Qual, CS, ]) appears to confirm that this can have
long-lasting effects on pastoralist communities. Furthermore, Snow (1984 [UN, Qual, CS, ])
hypothesizes that widespread provision of food assistance in Turkana was likely to lead to
an increase in the level of control exercised over the region by the Kenyan government. This
claim, however, is based on speculation, rather than empirical evidence.
strength of evidence relating to this theme is limited (where governance is concerned) or
non-existent (in relation to security).

4 CONCLUSION
This review has identified 24 publications that draw associations between the provision of in-
kind food assistance to pastoralists and changes in their health outcomes, livelihood
strategies, asset and income dynamics, mobility patterns and political and social relations.
Assessment of these publications permits some tentative conclusions to be drawn about the
impacts of food assistance on pastoralists.
Changes in livelihood strategies and asset and income dynamics:

Asset dynamics:
Provision of food assistance can lead to a reduction of livestock sales and
strengthen herd growth.
Income dynamics:
Provision of food assistance may fill gaps in pastoralists incomes.
Livelihood strategies:
Provision of food assistance can undermine the livelihood strategies of pastoralists.
The overall strength of the evidence relating to this theme is medium.
Mobility patterns:
Provision of food assistance can lead to changes in pastoralists mobility patterns, and
especially to sedentarization. This claim is made uniformly in a number of
publications.
The strength of evidence relating to this theme is limited.
Access to food assistance:

Food provided to pastoralists through food assistance interventions can be insufficient
and unbalanced.
Multiple publications claim that food assistance can lead to dependency, but there is
no empirical evidence of such a causal relationship.
According to one publication, provision of food assistance can lead to an increase in
alcohol production.
In some cases the modes of targeting of food assistance have led to internal
controversy within pastoralist communities. Elsewhere, targeting has effectively
encouraged the emergence of new political leaders who have sought to channel
assistance to their clients and, by extension, restricted some intended beneficiaries
access to food.
Household- and individual-level socio-demographic shifts:

Provision of food assistance can encourage pastoralist women to seek alternative
livelihood strategies.
According to most publications, access to food assistance leads to a decrease in
malnutrition. Others report, however, that food assistance can have negative impacts
on recipients health outcomes.
Social relations and governance:

Provision of food assistance can both strengthen relations within existing social
networks and contribute to the emergence of new political leaders (and displacement
of their predecessors).
The strength of evidence relating to these issues is limited.
Security:
The publications do not make a causal link between the provision of food assistance
and security.
4.1 LIMITATIONS
As the findings reported above suggest, the conclusions that this review can make are only
tentative, not only because of the small number of included publications but also because of
the limited strength of evidence that they provide. No included publication is based on
experimental data. Only one takes advantage of a quasi-experimental research design that
enables the author to effectively compare the impacts of food assistance on a pastoralist
population with a control group that did not receive food assistance. Some publications make
plausible but unavoidably far from conclusive causal claims based on credible analysis of
survey or qualitative data. However, in most publications a number of which have been
published in peer-reviewed journals causal claims are made offhand and without reference
to analysis or data on which they are based.
Due to the limited strength of evidence assessed in the included body of publications, the
conclusions of this review are necessarily tentative. In addition to laying the foundation for
further research examining the causal pathways contributing to impacts of food assistance
provided to pastoralist populations in the context of humanitarian crises, the included
publications make two additional important contributions to the future of aid.
First, the validation of current findings on impacts on pastoralist livelihoods highlights the
necessity for further population-based research. While the research team acknowledges that
the challenges involved in planning and conducting controlled trials may be deemed
logistically difficult and, in some cases, unethical, cohort designs (prospective cohort studies
as well as retrospective cohort designs which rely on validated methods) may prove to be
especially useful. Establishing routine and standardized measures of the provision of food
assistance to these populations can aid in estimating their qualitative and quantitative
impacts, while illustrating areas for modifications and cost savings in certain scenarios.
Furthermore, the establishment of cohort designs and linkages to measurable interventions
and outcomes will make it possible to ask a myriad of questions regarding the
appropriateness, cost and patterns of provision of food assistance among heterogeneous
pastoralist populations. Collaboration and guidance from humanitarian providers can create
novel opportunities to understand the effects that interventions may or may not have on the
intended population.
The second important contribution made by this review is the advancement in knowledge
regarding the appropriateness of the provision of food assistance to often marginalized and
vulnerable pastoralist populations. Although the findings themselves are limited in the quality
of their causal claims, they are significant in that they fill a conspicuous gap in the
humanitarian evidence literature and represent the first extensive attempt to date to identify,
collect and evaluate the provision of food assistance to pastoralist populations. Thus, the
illumination of the limited evidence and substantial opportunities for future evaluation and
measurement may prove to be the most important finding. The research team encourages
leaders in this space to work together to improve the overall quality of data collection and
reporting. To this end, we encourage the provision of funds for evaluating programmes
retrospectively, using mixed methods and a multidisciplinary approach, the goal of which will
be to develop a deeper, more dynamic and more valid understanding of the impacts of food
aid during different types of humanitarian crisis across representative populations.
Figure 4.1: Study characteristics of included publications by author, year, language, study design
and methods, population details, humanitarian crisis type and response with food aid among
included publications (n=24)
Publication Language Research Research Design and Study Study aims Location(s) Description of Number Crisis type Details of food
type design method period population(s) affected aid
HORN OF AFRICA
Taylor English Primary Observational Case study, 198081 To evaluate Somalia Population: Not 40,492 Drought Aid:
(1983) quantitative supplementary reported Unconditional
Conflict
feeding programmes food provision
Mobility: Nomadic border
(SFPs) functioning in
dispute Provider:
Somali refugee camps Gender: Not
between Somali Ministry
reported
Somalia of Health
and
Duration of aid:
Ethiopia
1 year
Snow English Unclear Unclear Case study, Not To investigate the Kenya Population: Not Drought Aid:
(1984) qualitative reported best ways to provide Turkana reported Unconditional
relief to pastoralist food provision,
Mobility: Nomadic
nomads such as the food-for-work
Turkana and its likely Gender: Men and
Provider: Not
long-term effects women
reported
Duration of aid:
Not reported
Adams and English Primary Observational Case study, 195689 To examine Darfurs Sudan Population: Not 34 million Drought Aid:
Hawksley qualitative experience with relief reported Unconditional
Famine
(1989) and development food provision
Mobility:
institutions in the
Agriculturalist, Provider:
process of post-
agropastoralist, United Nations,
drought recovery
nomadic government,
NGOs
Gender: Not
(unnamed)
reported
Duration of aid:
2 years
Ngunjiri English Primary Observational Case study, 198389 To describe two Kenya Population: Not Not Drought Aid: Food-for-
(1989) qualitative innovative projects reported reported livestock
undertaken by Oxfam exchange
Mobility: Not
in the pastoral areas
reported Provider:
of Kenya
Oxfam
Gender: Men and
women Duration of aid:
Not reported
Asmarom English Primary Observational Case study, Not To examine the Kenya Population: Not Drought Aid:
(1989) qualitative reported patterns of pastoral Borana, Gabbra reported Unconditional
Famine
adaptation to arid food provision
Mobility: Nomadic
environments, how
Provider:
the adaptive Gender: Not
Catholic
processes are reported
mission (further
affected by extended
details not
periods of drought,
provided)
and what types of
development might be Duration of aid:
appropriate under Not reported
those conditions
McCabe English Primary Quasi- Natural 197981 To compare the Kenya Population: Not Drought Aid:
(1990) experimental experiment, pastoral economies of Ngilukumong, reported Unconditional
qualitative two groups of Turkana Ngiyapakuno, food provision
Ngikamatak,
Provider: Not
Ngibocheros of
reported
Turkana
Duration of aid:
Mobility: Nomadic
3 years
Gender: Not
reported
Farzin English Primary Observational Case study, 197084 To assess whether Somalia Population: Not Not Not Aid:
(1991) quantitative the impact of food aid reported reported reported Unconditional
on Somalia has been food provision
Mobility: Not
positive or negative
reported Provider:
Multiple (details
Gender: Not
not provided)
reported
Duration of aid:
Not provided
Cutler English Primary Observational Other, 198485 To compare provision Ethiopia, Population: Not Not Drought Aid: Not
(1991) quantitative of food assistance in Sudan reported reported reported
Ethiopia and Sudan in
the mid-1980s Mobility: Not Provider:
reported Multiple (details
not provided)
Gender: Not
reported Duration of aid:
Not reported
Kilby (1993) English Secondary Other review Other, 198390 To assess the impacts Sudan Population: Not Not Drought Aid:
qualitative of food assistance reported reported Unconditional
programmes and an food aid
integrated pastoral Mobility: Not
development reported Provider:
programme Oxfam
Gender: Not
reported Duration of aid:
24 years
Bush (1995) English Primary Observational Case study, 199294 To evaluate the role of Kenya Population: 224,000 Drought Aid:
mixed food assistance in the Turkana Unconditional
context of long-term Flood food aid
drought Mobility: Nomadic
Provider:
Gender: Not Oxfam, World
reported Vision
Duration of aid:
2 years
Buchanan- English Primary Observational Case study, 199698 To evaluate the Kenya Population: Not 190,000 Drought Aid:
Smith and mixed Oxfam Wajir Relief reported Unconditional
Barton Programme 199698 Flood food aid
(1999) Mobility: Nomadic
Provider:
Gender: Men and Oxfam
women
Duration of aid:
2 years
No author English Primary Observational Case study, 2000 Ethiopia Population: Not Drought Aid: Not
(2001) quantitative Ethiopian Somalis reported reported
Famine
Mobility: Not Provider:
reported Multiple NGOs
Gender: Not Duration of aid:
reported Not reported
Pantuliano English Secondary Other review Case study, 1890 To examine the Sudan Population: Not Drought Aid: Food-for-
(2002) qualitative 2002 transformation of the Amarar/ Atmaan, reported work
Beja livelihood system Bishariyyn groups
in recent years in of the Beja Provider: World
response to changing Food
external Mobility: Nomadic Programme
circumstances, and Gender: Men and Duration of aid:
chart how coping women Not reported
mechanisms have
evolved to become
adaptive strategies
Overseas English Primary Observational Case study, 2006 To identify critical Djibouti, Population: 11 million Drought Aid: Not
Developme qualitative gaps in the response Eritrea, Pastoralists of reported
nt Institute to the drought in the Ethiopia, Djibouti, Eritrea,
(2006) Greater Horn of Africa Kenya, Ethiopia, Kenya Provider: Not
Somalia and Somalia reported
Mobility: Not Duration of aid:

Gender: Not
reported
Pantuliano English Primary Observational Case study, Mid- To consider the Sudan Population: Beja, 3.75 million Drought Aid:
(2007) qualitative 1980s alternatives to food Rashaida, others Unconditional
2007 assistance in eastern (unidentified) food aid
Sudan
Mobility: Nomadic, Provider: WFP
agropastoralist,
Duration of aid:
agriculturalist
20 years
Gender: Not
reported
Adow English Unclear Observational Case study, Not Unclear (perhaps to Kenya Population: Not Drought Aid: Not
(2008) qualitative reported observe the changes Pastoralists of reported reported
in pastoralist northern Kenya Flood
livelihoods in Northern Provider: Not
Kenya as a result of Mobility: Not reported
drought and floods) reported
Duration of aid:
Gender: Not Not reported
reported
Nangulu English Primary Observational Case study, 192095 To analyze food Kenya Population: Pokot Not Drought Aid: Not
(2009) qualitative security and coping reported reported
mechanisms in West Mobility: Nomadic Disease
Pokot, Kenya Provider:
Gender: Men and Livestock National
women raiding government,
WFP,
unidentified
NGOs
Duration of aid:
Not reported
Stockton English Primary Observational Case study, Not To discuss some of Horn of Africa Population: 20 million Drought Aid: Not
(2012) mixed reported the realities faced by Oromo, Pokot, reported
the 20 million Somali, Turkana, Conflict
pastoralists and the other (unidentified) Provider: Not
international agencies reported
that deliver Mobility: Nomadic
Duration of aid:
emergency and Gender: Not Not reported
development reported
assistance in the
pastoral arc of the
Horn of Africa
Lekapana English Primary Observational Case study, Not To study the Kenya Population: Not Not Drought Aid: Not
(2013) mixed reported socioeconomic effects reported reported reported
of drought on
pastoralists, their Mobility: Not Provider:
coping and adaptation reported Government,
strategies, and the development
Gender: Not partners
government reported
interventions in (details not
Loiyangalani Division reported)
of Marsabit County Duration of aid:
Not reported
Bersaglio et English Primary Observational Case study, 201112 To contextualize Kenya Population: 6,000 Drought Aid:
al. (2015) qualitative emergency responses Turkana Unconditional
to famine among food aid
Turkana pastoralists Mobility: Nomadic
Provider:
Gender: Men and Christian
women mission (details
not reported)
Duration of aid:
1 year
AFRICA, OTHER
News24 English Primary Observational Other, 2005 Niger Population: 12 million Drought Aid:
(2005) qualitative Nigerien Unconditional
pastoralists Famine food aid
Mobility: Not Provider:
reported United Nations
Gender: Not Duration of aid:
Cicalese et English Primary Observational Case study, 200607 To establish the Algeria Population: 250,000 Conflict Aid: Not
al. (2009) quantitative nutritional impact of Saharawis expulsion reported
the food aids from
distributed to the Mobility: Nomadic Western Provider: Not
Saharawi population Sahara reported
Gender: Not
in view of possible reported region Duration of aid:
clinical correlations Not reported
with celiac disease
ASIA
No author English Primary Observational Case 1999 To assess reported Mongolia Population: 810,000 Dzud Aid: Not
(2002) control, 2001 severe psychological Mongolian herders reported
quantitative stress, increased
school drop-out rates Mobility: Not Provider: Not
and increased reported reported
migration of rural Gender: Not Duration of aid:
herders into urban reported Not reported
centres
Fernandez- English Primary Observational Case study, 200910 To document Mongolia Population: 769,000 Dzud Aid:
Gimenez et mixed individual and Mongolian herders Unconditional
al. (2012) community food aid
experiences with dzud Mobility: Nomadic
and identify the Provider: Not
Gender: Not reported
factors that make reported
some households and Duration of aid:
communities more Not reported
vulnerable to dzud
and others less so
Figure 4.2: Changes in livelihood strategies and asset and income dynamics
Publication Type of Outcome(s) Description of impact(s)

intervention measured
HORN OF AFRICA
McCabe (1990) Unconditional Livelihood Compared with the nomadic Ngisonyoka of Southern Turkana, those in
food aid strategies the Ngilukumong, Ghiyapakuno, Ngikamatak and Ngibocheros sections
of the northern Turkana witnessed a breakdown of indigenous drought
coping strategies.
Cutler (1991) Not reported Income Undermining the domestic food economy: grain prices dropped;
dynamics labourers failed to migrate where needed to assist with the harvest; and
crops were left unharvested because the market for them became too
weak.
Kilby (1993) Unconditional Income The food assistance programmes in Sudan were compared with the
food aid dynamics Oxfam restocking project in Kenya (Moris, 1988): The results suggest
that it is still far more economic to restock than to provide food aid for
four years.
Snow (1984) Unconditional Mobility Established food-for-work projects leading to a life away from
food aid and patterns pastoralism, towards agriculture. A possible shift to a cash crop
food-for-work economy was also observed.
Bush (1995) Unconditional Livelihood Food aid led to a strengthening of herd growth while it filled income gaps
food aid strategies among recipients.
Buchanan-Smith and Unconditional Asset and The provision of food aid resulted in a reduction in livestock sales while
Barton (1999) food aid income reducing overall individual-level and household debt.
dynamics
Pantuliano (2002) Food-for-work Livelihood Changes in livelihood strategy patterns were observed, in particular a
strategies weakening of the Beja livelihood system.
Overseas Not reported Livelihood Although limited details were provided on the impact of food aid, the
Development Institute strategies authors noted: The humanitarian response to the crisis in the Greater
(2006) Horn of Africa has done very little, to date, to protect livelihoods.
Nangulu (2009) Not reported Livelihood Patterns in livelihood strategies have been substantially undermined by
strategies the receipt of food aid.
Bersaglio et al. (2015) Unconditional Livelihood Livelihood patterns modified in response to food aid, including the
food aid strategies creation of new spaces for community collaboration. However, this has
also contributed to the movement away from pastoralism.
Figure 4.3: Mobility patterns
Publication Type of Reported Description of impact(s)

intervention impact(s)
HORN OF AFRICA
Snow (1984) Unconditional Mobility Established food-for-work projects leading to a life away from
food aid and patterns pastoralism, towards agriculture. A possible shift to a cash crop
food-for-work economy was also observed.
Ngunjiri (1989) Unconditional Mobility The provision of food aid contributed to internal distribution of both
food aid patterns assets and responsibilities: Stock owners tended to divide their herds in
two in response to the drought. Women, children, and older members of
the household were often left with the milk herd near centers where relief
was likely to be provided. The men often moved the remainder of the
herd in search of better grazing in areas distant from the center. Though
this meant that the project tended to reach the most vulnerable members
of the household, it also meant that much of the stock put forward to
slaughter was female stock, which would have better been retained for
rebuilding the herd after the drought.
McCabe (1990) Unconditional Mobility The attraction of free food influenced traditional patterns of movement
food aid patterns for many individual herd-owners. If members of the family were receiving
food aid, there was a strong incentive for the herd-owner to try to remain
near the relief centre. Once the new rules requiring whole families to
apply for food aid were instituted, this became a necessity. Since the
environment close to the settlements was rapidly becoming denuded,
livestock which, in many instances, were already stressed were forced to
forage on sparser and less nutritious plants than they would have away
from the settlements. This practice inevitably led to livestock losses
through starvation or disease. Thus the famine relief centres were, to
some extent, creating their own clients.
Kilby (1993) Unconditional Mobility As a result of food aid, those from the Red Sea Hills had increased rates
food aid patterns of sedentarization.
Pantuliano (2002) Food-for-work Mobility Food aid contributed to increased sedentarization and urban drift among
patterns recipients.
Pantuliano (2007) Unconditional Mobility Sedentarization increased as a result of food aid.
food aid patterns
Adow (2008) Food aid Mobility Movement patterns changed in response to food aid: Many move near
patterns urban centres to seek emergency food aid.
Stockton (2012) Not reported Mobility Through the receipt of food aid, populations are reported to be
patterns increasingly sedentary. Although sedentarization is claimed to be the
consequence of food aid, it may be more a result of the development of
water resources, which has taken place alongside the provision of food
aid.
Bersaglio et al. (2015) Unconditional Mobility Livelihood patterns modified in response to food aid have contributed to
food aid patterns the movement away from pastoralism.
Figure 4.4: Access, use and dependency on food aid
Publication Type of Outcome(s) Description of impact(s)

intervention measured
HORN OF AFRICA
Snow (1984) Unconditional Use of food Once food aid was implemented, there was an increase in alcoholism
food aid and assistance due to distillation of corn, intended for relief, by people with adequate
food-for-work food supplies who set up stills.
Farzin (1991) Unconditional Dependency The provision of food aid led to dependence upon its existence while
food aid having a disincentive effect on the domestic food supply.
Nangulu (2009) Not reported Dependency Food aid may lead to dependence and enhance poverty among
recipients.
Adow (2008) Not reported Dependency Pastoralists in Northern Kenya are almost entirely dependent on
emergency food aid.
Lekapana (2013) Not reported Dependency Successive drought episodes have forced pastoral communities to rely
on emergency food aid, which has reinforced the cycle of dependency.
Access to Although food aid was offered to all, unequal distribution of available
food goods was reported.
AFRICA, OTHER
Cicalese et al. (2009) Not reported Access to Food aid provided was reported to be insufficient and unbalanced,
food leading to widespread malnutrition among the recipient population.
ASIA
Fernandez-Gimenez Unconditional Dependency Food aid may have created dependency among recipients: Relief aid
et al. (2012) food aid that helps prevent loss of life, suffering, and impoverishment in the short-
term may contribute to long-term dependence syndromes, social
disparities, and lack of initiatives on the part of both herders and local
government.
Figure 4.5: Gender and health

HORN OF AFRICA
Taylor (1983) Unconditional Health Children enrolled in food programmes gained weight over the course of
food aid access to food relief.
Kilby (1993) Unconditional Health Food aid had little positive effect on human mortality levels.
food aid
Bush (1995) Unconditional Health There was a decline in rates of childhood malnutrition.
food aid
Buchanan-Smith and Unconditional Health Rates of malnutrition decreased during the provision of food aid.
Barton (1999) food aid
No author (2001) Not reported Health Health was substantially impacted through food aid: Even though food
aid and feeding centers are a priority during famine, attracting a large
concentration of susceptible persons to feeding centers may increase
transmission of infectious diseases such as measles and diarrhea.
Gender Gender roles were impacted by the provision and receipt of food aid.
roles There was a growing tendency for young Turkana women to seek
alternatives to nomadic lifestyles.
Bersaglio et al. (2015) Unconditional Well-being A reduction in anxiety, improvement of living standards and enhanced
food aid spirituality were observed with food aid receipt.
ASIA
No author (2002) Not reported Health Compared with those herders in districts only slightly or not affected:
Relief efforts to distribute food to affected areas might have lessened
the nutritional impact of livestock losses among herders and their
families.
Figure 4.6: Changes in social relations

HORN OF AFRICA
Snow (1984) Unconditional Social Food aid may have led to the creation of new power structures and
food aid and relations power brokers who challenged the authority of and respect given to
food-for-work elders. Furthermore, greater concentration of people was observed in
camps.
Asmarom (1989) Unconditional Social Demoralization of the community receiving food assistance was
food aid relations observed: It is the closest thing to an Ik type of non-community that I
ever observed in the Boran country.9
Bush (1995) Unconditional Social Food aid led to a strengthening of social networks.
food aid relations
Pantuliano (2002) Food-for-work Social Aid distribution created a new class of food shaikhs, which brought
relations about a modification of the existing social structure and the traditional
inter-diwab (lineage) dynamics founded on exchange of resources and
reciprocity.
Figure 4.7: Security and governance

HORN OF AFRICA
Snow (1984) Unconditional Government Greater and increasing power of the Kenyan state and its ability to
food aid and control the Turkana were observed through the provision of food aid.
food-for-work
Adams and Hawksley Unconditional Security During the crisis in Darfur, food aid by non-governmental intervention
(1989) food aid conditions was widely lauded to have saved a great many lives during the
emergency.
Pantuliano (2007) Unconditional Mobility Sedentarization increased as a result of food aid, not only leading to a
food aid patterns strengthening of internal political divisions but also establishing power
brokers.
9
Asmarom alludes to Turnbulls (1974) controversial study of the Ik of northern Karamoja in Uganda, in which he details the alleged
disintegration of their society following the supposed destruction of their hunting-based livelihood system.
Figure 4.8: Assessment of the strength of evidence
Abbreviations and symbols: Y yes, N no, UN unclear, moderate strength of evidence, low strength of evidence
Publication Does the study
Pose a research question or outline a
Draw conclusions based on results?

Demonstrate measurement validity?
Construct a conceptual framework?
Consider context-specific cultural
Have results sensitive to analysis

Use internally reliable measures?
Acknowledge existing research?
Acknowledge limitations and/or

factors that may bias results?
Demonstrate the suitability of
Declare sources of funding?
Present or link to raw data it
alternative interpretations?
Appear ecologically valid?
Report stable measures?

Appear internally valid?
Signpost throughout?
Strength of evidence
research design?
hypothesis?
analyses?
used?
Taylor (1983) Y Y Y Y N N N UN UN UN UN UN UN Y N Y
Snow (1984) Y Y Y N N N N N UN UN UN UN UN Y N Y
Adams and Hawksley (1989) Y Y N N N N N N UN UN UN UN UN Y N Y
Ngunjiri (1989) N N N Y N N N N UN UN UN UN UN Y N UN
Asmarom (1989) Y Y Y Y N N N N UN UN UN UN UN Y N Y
McCabe (1990) Y Y Y Y N N N N UN UN UN UN UN Y N Y
Farzin (1991) Y Y Y Y N N N UN UN UN UN UN UN Y N Y
Cutler (1991) Y Y Y N N N N N UN UN UN UN UN Y N Y
Kilby (1993) Y Y Y Y N N N N UN UN UN UN UN Y N Y
Bush (1995) Y Y Y Y N N N UN UN UN UN UN UN Y N Y
Buchanan-Smith and Barton Y Y Y Y Y N N N UN UN UN UN UN Y N Y

(1999)
No author (2001) N Y N Y N N N UN UN UN UN UN UN Y N Y
Pantuliano (2002) Y Y Y Y N N N UN UN UN UN UN Y N Y
No author (2002) N Y N Y N N N UN UN UN UN UN UN Y N Y
News24 (2005) N N N N N N N N UN UN UN UN UN N/A N N/A
Overseas Development Y Y Y N N N N N UN UN UN UN UN Y N UN
Institute (2006)
Pantuliano (2007) Y Y Y Y N N N N UN UN UN UN UN Y N Y
Adow (2008) N N N N N N N N UN UN UN UN UN N/A N N
Nangulu (2009) Y Y Y Y N N N N UN UN UN UN UN Y N Y
Cicalese et al. (2009) N Y Y Y N N N N UN UN UN UN UN N/A N Y
Fernandez-Gimenez et al. Y Y Y Y N N N UN UN UN UN UN UN Y N Y
(2012)
Stockton (2012) N N Y N N N N N UN UN UN UN UN Y N UN
Lekapana (2013) Y Y Y Y N N N N UN UN UN UN UN Y Y UN
Bersaglio et al. (2015) Y Y Y Y N N N N UN UN UN UN UN Y N Y

APPENDIX 1: DATABASE SEARCH
STRATEGY STRINGS
Ovid Medline 1946 to May Week 4 2016, Ovid Medline In-
Process and other non-indexed citations 1 June 2016
Search strategy
# Searches Results
1 exp Disasters/pc, mt [Prevention & Control, Methods] 2,285
2 ((food or nourish* or nutrition* or sustenance or ration* or meal*) adj3 (assist* or relief* or 6,749
relie* or help* or aid* or provision*)).tw,sh,fs,kw,ot.
3 relief work/ or food assistance/ 3,924
4 (in-kind* adj3 (food* or nutrition* or meal* or nourish* or ration*)).tw,sh,fs,kw,ot. 23
5 (in-kind adj3 (assist* or aid*)).tw,sh,fs,kw,ot. 13
6 (food adj3 (distribut* or work* or asset*)).tw,sh,fs,kw,ot. 2,592
7 (humanitarian* adj3 (relie* or aid* or intervention* or assist* or help* or 767
interven*)).tw,sh,fs,kw,ot.
8 International Cooperation/ 40,987
9 (humanitarian adj3 (mission* or coopera* or co-op*)).tw,sh,fs,kw,ot. 239
10 (relief adj3 (work* or program* or assist* or co-op* or cooper* or initiativ*)).tw,sh,fs,kw,ot. 715
11 (international adj3 (cooperat* or co-op* or assist* or relie* or aid* or help* or 5,106
aid*)).tw,sh,fs,kw,ot.
12 or/1-11 [ Intervention] 59,839
13 ((cattle or sheep or cow or mammal* or livestock or live-stock or animal* or llama* or yak* or 18,209
goat* or camel*) adj3 husbandr*).ti,ab,fs,sh,kw,ot,hw.
14 pastoral*.tw,sh,fs,kw,ot,hw. 5151
15 Animal Husbandry/ 17,119
16 exp Agriculture/ 69,832
17 limit 16 to yr="1966 - 1975" 2,748
18 Animals/ 5,889,534
19 animals, domestic/ or livestock/ 16,160
20 agr*-pastoral*.tw,sh,fs,kw,ot. 116
21 (agr* adj3 pastoral*).tw,sh,fs,kw,ot,hw. 233
22 ((nomad* or transhuman* or itineran* or migrat*) adj3 (livelihood* or farm* or societ* or group* 1,457
or lifestyle)).tw,hw,fs,sh,kw,ot.
23 "Transients and Migrants"/ 9,140
24 or/13-23 5,933,380
25 12 and 24 5,950
26 limit 25 to yr="1967 -Current" 5,902
CAB Abstracts 1973 to 2016 Week 20

Results: 327
(((mammal* or vertebrate* or cattle* or sheep* or goat* or camel* or yak* or llama* or animal*
or chordata* or livestock or live-stock*) adj3 husbandr*).mp,sh,ot. or pastoral*.mp,sh,ot or
(agr* adj3 pastoral*).mp,sh,ot OR agr*-pastoral*.mp,sh,ot or pastoralism/ or agropastoral
systems/ or silvopastoral systems/ OR animal husbandry/ or livestock farming/ or ((nomad*
or transhuman* or itineran* or migrat*) adj3 (livelihood* or farm* or societ* or group* or
lifestyle)).mp,sh,ot. or ((cattle or sheep or cow or mammal* or livestock or live-stock or
animal* or llama* or yak* or goat* or camel*) adj3 husbandr*).mp,sh,ot.) and (((food or
nourish* or nutrition* or sustenance or ration* or meal*) adj3 (assist* or relief* or relie* or
help* or aid* or provision*)).mp,ot,sh. or (food adj3 (distribut* or work* or asset*)).mp,ot,sh.
or exp food aid/ or emergency relief/ or (humanitarian adj3 (mission* or coopera* or co-
op*)).mp,sh,ot. or (relief adj3 (work* or program* or assist* or co-op* or cooper* or
initiativ*)).mp,sh,ot. or (international adj3 (cooperat* or co-op* or assist* or relie* or aid* or
help* or aid*)).mp,sh,ot. OR (in-kind adj3 (assist* or aid*)).mp,sh,ot. or (humanitarian* adj3
(relie* or aid* or intervention* or assist* or help* or interven*)).mp,sh,ot. or (in-kind* adj3
(food* or nutrition* or meal* or nourish* or ration*)).mp,sh,ot.)
LILACS
Results: 4142
(tw:(livestock* OR pastoral* OR agro-pastoral* OR animal husbandr* OR nomad* OR
transhuman*)) AND (tw:(relief* OR crisis OR aid* OR in-kind OR food assist* OR
humanitarian OR interven* OR food distribution*))
Web of Science databases (same string for both databases)
BIOSIS
Results: 726
Web of Science core collection

Results: 3,084
TOPIC: ((mammal* or vertebrate* or cattle* or sheep* or goat* or camel* or yak* or llama* or
animal* or chordata* or livestock or live-stock*) NEAR/3 husbandr*) OR TOPIC: (nomad* or
transhuman* or itineran* or migrat* OR pastoral* OR agr*-pastoral*) AND TOPIC: ((food or
nourish* or nutrition* or sustenance or ration* or meal*) NEAR/3 (assist* or relie* or help* or
aid* or provision*)) OR TOPIC: (food NEAR/3 (asset* OR for work OR distribut*)) OR
TOPIC: ((crisis OR emergenc* OR humanitarian) NEAR/3 (aid* OR relie* OR assist* OR
interven* OR help)) OR TOPIC: (relief NEAR/3 (work* or program* or assist* or co-op* or
cooper* or initiativ*))
World Wide Political science abstracts

Results: 115
(((food OR nourish* OR nutrition* OR sustenance OR ration* OR meal*) NEAR/3 (assist* OR
relief* OR help* OR aid* OR provision*)) OR (food NEAR/3 (asset* OR for work OR
distribute*) OR (crisis OR emergenc* OR humanitarian OR international) NEAR/3 (aid* OR
relief* OR assist* OR interven* OR help OR support))) AND (((mammal* OR vertebrate* OR
animal* OR chordata* OR live-stock OR livestock*) NEAR/3 husbandr*) OR (nomad* OR
transhuman* OR itinerant* OR migrant* OR pastoral* OR agr*-pastoral*))
PAIS International
Results: 146
relief* OR assist* OR interven* OR help OR support*))) AND (((mammal* OR vertebrate* OR
Africa-Wide NiPAD
Results: 595
(((food OR nourish* OR nutrition* OR sustenance OR ration* OR meal*) N3 (assist* OR
relief* OR help* OR aid* OR provision*)) OR (food N3 (asset* OR for work OR distribut*))
OR ((crisis OR emergenc* OR humanitarian OR international) N3 (aid* OR relief* OR assist*
OR interven* OR help OR support*))) AND (((mammal* OR vertebrate* OR animal* OR
chordata* OR live-stock OR livestock*) N3 husbandr*) OR (nomad* OR transhuman* OR
itinerant* OR migrant* OR pastoral* OR agr*-pastoral*))
Google Scholar
Results: 412
(pastoral* OR nomad* OR transient* OR migrant* OR animal husbandry* OR itinerant*)
AND (relief* OR aid* OR food assist* OR food asset OR in-kind OR food distribution OR
meal distribution OR humanitarian OR interven* OR help* OR support*)
Cochrane
Results: 40
Search Name: Oxfam HEP Search
Last saved: 03/06/2016 17:38:51.708
Description: Ran June 3, 2016
ID Search
#1 MeSH descriptor: [Animal Husbandry] explode all trees
#2 MeSH descriptor: [Agriculture] explode all trees
#3 MeSH descriptor: [Animals] explode all trees
#4 MeSH descriptor: [Transients and Migrants] explode all trees
#5 (cattle or sheep or cow or mammal* or livestock or live-stock or animal* or llama* or
yak* or goat* or camel*) near/3 husbandr*
#6 (nomad* or transhuman* or itineran* or migrat*) near/3 (livelihood* or farm* or societ*
or group* or lifestyle)
#7 pastoral*
#8 agro near/3 pastoral
#9 agr*-pastoral*
#10 MeSH descriptor: [Relief Work] explode all trees
#11 MeSH descriptor: [Food Assistance] explode all trees
#12 MeSH descriptor: [Disasters] explode all trees and with qualifier(s): [Prevention &
control PC]
#13 MeSH descriptor: [International Cooperation] explode all trees
#14 (food or nourish* or nutrition* or sustenance or ration* or meal*) near/3 (assist* or
relief* or relie* or help* or aid* or provision*)
#15 in-kind* near/3 (food* or nutrition* or meal* or nourish* or ration*)
#16 in-kind near/3 (assist* or aid*)
#17 food near/3 (distribut* or work* or asset*)
#18 humanitarian* near/3 (relie* or aid* or intervention* or assist* or help* or interven*)
#19 humanitarian near/3 (mission* or coopera* or co-op*)
#20 relief near/3 (work* or program* or assist* or co-op* or cooper* or initiativ*)
#21 international near/3 (cooperat* or co-op* or assist* or relie* or aid* or help* or aid*)
#22 #1 or #2 or #3 or #4 or #5 or #6 or #7 or #8 or #9
#23 #10 or #11 or #12 or #13 or #14 or #15 or #16 or #17 or #18 or #19 or #20 or #21
#24 #22 and #23
IDEAS
Results: 4,101
(pastoral*|nomad*|transient*|migrant*|transhuman*|animal husbandr*|agr*-pastoral) +
(relie*|aid*|"food assis*"|in-kind|interven*|humanitarian*|asset*|food*|help|support*)
International Bibliography of the Social Sciences

Results: 142
relief* OR assist* OR interven* OR help OR support*))) AND (((mammal* OR vertebrate* OR
ARTFL-FRANTEXT (French)
Results: 2,084
Paste in the Search in texts for search bar:
nomad.*|transient.*|transhuman*|itinerant.*|pastoral.*
EMBASE
Results: 1,608
Database(s): Embase Classic+Embase 1947 to 2016 Week 23
(((food or nourish* or nutrition* or sustenance or ration* or meal*) adj3 (assist* or relief* or
relie* or help* or aid* or provision*)).tw,sh,fs,kw,ot,hw or relief work/ or food assistance/ or
(in-kind* adj3 (food* or nutrition* or meal* or nourish* or ration*)).tw,sh,fs,kw,ot,hw. or (in-
kind adj3 (assist* or aid*)).tw,sh,fs,kw,ot,hw. or (humanitarian* adj3 (relie* or aid* or
intervention* or assist* or help* or interven*)).tw,sh,fs,kw,ot,hw. or International Cooperation/
or (humanitarian adj3 (mission* or coopera* or co-op*)).tw,sh,fs,kw,ot,hw or (relief adj3
(work* or program* or assist* or co-op* or cooper* or initiativ*)).tw,sh,fs,kw,ot,hw. or
(international adj3 (cooperat* or co-op* or assist* or relie* or aid* or help* or
aid*)).tw,sh,fs,kw,ot,hw. or disaster/pc) AND (pastoral*.tw,sh,fs,kw,ot,hw or animals,
domestic/ or livestock/ or agr*-pastoral*.tw,sh,fs,kw,ot,hw. or (agr* adj3
pastoral*).tw,sh,fs,kw,ot,hw. or exp animal husbandry/ or *agriculture/ or exp migrant/ or
emigrant/ or immigrant/ or (nomad* or migrant* or transhuman* or
itinerant*).tw,sh,fs,kw,ot,hw. or ((nomad* or transhuman* or itineran* or migrat*) adj3
(livelihood* or farm* or societ * or group* or lifestyle)).tw,hw,fs,sh,kw,ot.)
Grey literature sources and searches

* The following sources were all searched on 6 June 2016.
Addis Ababa University

Searched, nothing found. Keywords used: pastoralism, pastoral, humanitarian.
Center for International Development, Harvard University

Broken down by domains, under Environment and Natural Resources, 2009 (earliest)2016
date range https://research.hks.harvard.edu/publications/Policy_Domain.aspx
Feinstein International Center
FHI 360
Keyword: pastoralism
Results:
https://www.fhi360.org/explore/content?search_api_views_fulltext=pastoralism&op=Search
Institute for Development Studies at the University of Nairobi

Keywords: pastoralism AND humanitarian intervention
Results:
This organization pointed to two repositories for information, which were also searched:
1. Institute of Development Studies, University of Nairobi

Keywords: nomad* OR transient* OR itineran* OR transhuman* OR pastoralis* OR
migrant* OR animal husbandry
Results:
http://opendocs.ids.ac.uk/opendocs/handle/123456789/28/discover?query=nomad*+O
R+transient*+OR+itineran*+OR+transhuman*+OR+pastoralis*+Or+migrant*+OR+%22
animal+husbandry%22&submit=Go
2. University of Nairobi digital repository

Keywords:
nomad* OR transient* OR itineran* OR transhuman* OR pastoralis* OR migrant* OR
animal husbandry
Results:
http://erepository.uonbi.ac.ke:8080/xmlui/discover?filtertype_3=title&filter_relational_o
perator_3=contains&filter_3=&submit_apply_filter=Apply&query=nomad*+OR+transie
nt*+OR+itineran*+OR+transhuman*+OR+pastoralis*+Or+migrant*+OR+%22animal+h
usbandry%22&scope=%2F
Institute for Global Health at University College London

Use keyword: pastoralism. No stable URL available to share results.
Institute of Development Studies at the University of Sussex

Different pockets to search:
1. Publication search:
Keyword: Pastoralism
2. Research on Livestock and Pastoralism:

No keyword needed, area can be browsed.
3. Advanced search (click on it):

Keywords: pastoralis* OR husbandry
International Institute for Environment and Development (IIED)

Keyword: Pastoralis*
Results:
http://pubs.iied.org/search.php?k=pastoralis*&t=&a=&w=&s=&c=&g=&l=&tdB=1&tdC=1&tdA
=1&tdJ=1&tdF=1&tgI=1&tgP=1&tgE=1&tmd=1&tmf=1&d=a&b=d&r=a&z=Search
International Livestock Research Institute (ILRI), Makerere University:

http://www.ilri.org/
Keywords: (pastoralism OR nomad OR transient OR migrant OR husbandry OR itinerant OR
transhuman) AND (food OR aid OR relief OR support OR help OR in-kind OR assistance
OR intervention)
Results:
https://books.google.com/books/p/international_livestock_research_institute?oe=UTF-
8&q=%28pastoralism+OR+nomad+OR+transient+OR+migrant+OR+husbandry+OR+itineran
t+OR+transhuman%29+AND+%28food+OR+aid+OR+relief+OR+support+OR+help+OR+in-
kind+OR+assistance+OR+intervention%29&btnG=Search+Books
Overseas Development Institute, Oxford Policy Management

Keywords: pastoral, pastoralism, husbandry, nomad.
Nothing was retrieved.
Danish International Development Agency (DANIDA)

Keywords: pastoralism OR nomad OR transient OR itinerant OR migrant OR transhuman
Results:
http://um.dk/search?q=pastoralism%20OR%20nomad%20OR%20transient%20OR%20itiner
ant%20OR%20migrant%20OR%20transhuman&filter=0
Deutsche Gesellschaft fr Internationale Zusammenarbeit (GIZ)

Keywords: pastoral* OR nomad* OR transient* OR migrant* OR transient* OR transhuman*
OR husbandr*
Site doesnt support permanent search; go here: http://star-
www.giz.de/starweb/giz/pub/servlet.starweb?path=giz/pub/pub.web&STAR_AppLanguage=1
and copy and paste the keywords above.
European Union Humanitarian Aid and Civil Protection (ECHO)

Other sites identified:
1. European Commission
Results:
http://ec.europa.eu/geninfo/query/index.do?queryText=pastoralism+&summary=summ
ary&more_options_source=global&more_options_date=*&more_options_date_from=&
more_options_date_to=&more_options_language=en&more_options_f_formats=*&swl
ang=en
2. EU Aid Explorer:
https://euaidexplorer.ec.europa.eu/SearchPageAction.do
Use keyword: pastoralist
No ability to save a search string, copy and paste keyword above. Go to both
Development and Humanitarian Projects.
3. International Cooperation and Development

Results: http://ec.europa.eu/europeaid/search/library/pastoralism_en
Swedish International Development Agency

Keywords: pastoralist OR pastoralism OR nomad OR transient OR migrant OR itinerant OR
transhuman
Results:
http://www.sida.se/English/search/?site=www.sida.se%2FEnglish%2F&q=pastoralist+OR+p
astoralism+OR+nomad+OR+transient+OR+migrant+OR+itinerant+OR+transhuman
United Kingdom Department for International Development

Keywords: pastoralism
Results:
https://www.gov.uk/search?q=pastoralism
United States Office of Foreign Disaster Assistance

transhuman
Results:
https://www.usaid.gov/gsearch/%2BOR%2Bpastoralism%2BOR%2Bnomad%2BOR%2Btran
sient%2BOR%2Bmigrant%2BOR%2Bitinerant%2BOR%2Btranshuman
African Development Bank

transhuman
Results:
http://www.afdb.org/en/search/?query=pastoralist+OR+pastoralism+OR+nomad+OR+transie
nt+OR+migrant+OR+itinerant+OR+transhuman
Asian Development Bank

Keywords: pastoralist
Results:
http://www.adb.org/search?keywords=pastoralist
International Committee of the Red Cross

Keywords: (pastoralist OR pastoralism OR nomad OR transient OR migrant OR itinerant OR
transhuman) AND (aid OR relief OR support OR assistance OR in-kind OR food OR help
OR humanitarian OR intervention)
Results:
https://www.icrc.org/en/resource-
centre/result?context=hf%253D7%2526s%253D%2526logic%253Dinternet-
eng%2526q%253D%252528pastoralist%252BOR%252Bpastoralism%252BOR%252Bnoma
d%252BOR%252Btransient%252BOR%252Bmigrant%252BOR%252Bitinerant%252BOR%
252Btranshuman%252529%252BAND%252B%252528aid%252BOR%252Brelief%252BOR
%252Bsupport%252BOR%252Bassistance%252BOR%252Bin-
kind%252BOR%252Bfood%252BOR%252Bhelp%252BOR%252Bhumanitarian%252BOR%
252Bintervention%252529%2526sa%253D0&s=&sa=1&b=0
International Organization for Migration, Office for the Coordination of Humanitarian

Affairs
Results:
http://www.iom.int/search/pastoralist
Office of the United Nations High Commissioner for Refugees, United Nations
Keywords: pastorali*
Results:
http://www.unhcr.org/search?query=pastorali*
UNICEF
Results:
http://www.unicef.org/search/search.php?querystring_en=pastoralist+&hits=&type=&navigati
on=&Go.x=0&Go.y=0
* The following sources were all searched on 7 June 2016.
World Bank
Keywords: (pastoralist OR pastoralism OR nomad OR transient OR migrant OR itinerant OR
transhuman) AND (aid OR relief OR support OR assistance OR in-kind OR food assistance
OR help OR humanitarian OR intervention)
Results (limited to English):
http://search.worldbank.org/all?qterm=%28pastoralist+OR+pastoralism+OR+nomad+OR+tra
nsient+OR+migrant+OR+itinerant+OR+transhuman%29+AND+%28aid+OR+relief+OR+sup
port+OR+assistance+OR+in-
kind+OR+%22food+assistance%22+OR+help+OR+humanitarian+OR+intervention%29+&titl
e=&filetype=&_Top/language=English
Results (limited to French):
http://search.worldbank.org/all?qterm=%28pastoralist+OR+pastoralism+OR+nomad+OR+tra
nsient+OR+migrant+OR+itinerant+OR+transhuman%29+AND+%28aid+OR+relief+OR+sup
port+OR+assistance+OR+in-
kind+OR+%22food+assistance%22+OR+help+OR+humanitarian+OR+intervention%29+&titl
e=&filetype=&_Top/language=French
World Health Organization

Keywords: pastoralism AND (aid OR relief OR support OR assistance OR in-kind OR food
assistance OR help OR humanitarian OR intervention)
Pastoralist AND (aid OR relief OR support OR assistance OR in-kind OR food assistance
OR help OR humanitarian OR intervention)
Transhumance AND (aid OR relief OR support OR assistance OR in-kind OR food
assistance OR help OR humanitarian OR intervention)
Advanced Google search, look into site: who.int
In find any of these words typed keywords above. Limit by year to 19672016
Action Against Hunger

Keywords: pastoralist OR pastoralism OR nomad OR transient OR itinerant OR
transhumance
Results:
http://www.actionagainsthunger.org/search/node/pastoralist%20OR%20pastoralism%20OR
%20nomad%20OR%20transient%20OR%20itinerant%20OR%20transhumance
International Rescue Committee

transhumance
Results:
http://www.rescue.org/search/node/pastoralist%20OR%20pastoralism%20OR%20nomad%2
0OR%20transient%20OR%20itinerant%20OR%20transhumance
Mdecins Sans Frontires/Doctors Without Borders

transhumance
Results:
http://www.msf.org/en/search?keyword=pastoralist%20OR%20pastoralism%20OR%20noma
d%20OR%20transient%20OR%20itinerant%20OR%20transhumance
Oxfam Canada
transhumance
Results:
http://www.oxfam.ca/search/node/pastoralist%20OR%20pastoralism%20OR%20nomad%20
OR%20transient%20OR%20itinerant%20OR%20transhumance
Save the Children

transhumance
Results:
https://www.savethechildren.net/search/node/pastoralist%20OR%20pastoralism%20OR%20
nomad%20OR%20transient%20OR%20itinerant%20OR%20transhumance
Vtrinaires Sans Frontires

transhumance
Results:
Advanced Google search, look into site: vsf-international.org
The Bill & Melinda Gates Foundation

transhumance
Results:
http://www.gatesfoundation.org/search#q/k=pastoralist%20OR%20pastoralism%20OR%20n
omad%20OR%20transient%20OR%20itinerant%20OR%20transhumance
The Clinton Foundation

transhumance
Results:
https://www.clintonfoundation.org/search/node/pastoralist%20OR%20pastoralism%20OR%2
0nomad%20OR%20transient%20OR%20itinerant%20OR%20transhumance
The Gatsby Charitable Foundation

transhumance
Results (no results):
http://www.gatsby.org.uk/search-
results?term=pastoralist+OR+pastoralism+OR+nomad+OR+transient+OR+itinerant+OR+tra
nshumance+&sort_by=relevant
The Rockefeller Foundation

transhumance
Results:
Advanced Google search, look into site: rockefellerfoundation.org
Open Grey
Keywords: (pastoralist OR pastoralism OR nomad* OR transient* OR migrant* OR itinerant*
OR transhuman*) AND (aid OR relie* OR support OR assist* OR in-kind OR "food
assistance" OR help OR humanitarian OR interven*)
Results:
http://www.opengrey.eu/search/request?q=%28pastoralist+OR+pastoralism+OR+nomad*+O
R+transient*+OR+migrant*+OR+itinerant*+OR+transhuman*%29+AND+%28aid+OR+relie*+
OR+support+OR+assist*+OR+in-
kind+OR+%22food+assistance%22+OR+help+OR+humanitarian+OR+interven*%29
GreyLit
transhumance
Results (no results):
http://www.greylit.org/library/search#wt=json&facet=true&q=pastoralist%20OR%20pastoralis
m%20OR%20nomad%20OR%20transient%20OR%20itinerant%20OR%20transhumance%
20&qt=dismax&fl=id&qf=full_text&facet.field=publisher&facet.field=full_subjects&q.op=AND&
start=0
APPENDIX 2: INCLUDED PUBLICATIONS
AND REPORTED INTERVENTIONS AND
IMPACTS
Publication Type of Reported Description of Impact(s)
Intervention Impact(s)
HORN OF AFRICA
Taylor (1983) Unconditional Health Children enrolled in food programmes gained weight over the course of access to
food aid food relief.
Snow (1984) Unconditional Use of food Once food aid was implemented, there was an increase in alcoholism due to
food aid and assistance distillation of corn, intended for relief, by people with adequate food supplies who
food-for-work set up stills.
Livelihood Established food-for-work projects leading to a life away from pastoralism towards
strategies agriculture. A possible shift to a cash crop economy was also observed.
Gender roles There was a growing tendency for young Turkana women to seek alternatives to
nomadic lifestyles.
Social Food aid may have led to the creation of new power structures and power brokers
relations who challenge the authority of and respect given to elders. Furthermore, greater
concentration of people was observed in camps.
Government Greater and increasing power of the Kenyan state and its ability to control the
Turkana was observed through the provision of food aid.
Adams and Unconditional Security During the crisis in Darfur, food aid by non-governmental intervention was widely
Hawksley (1989) food aid lauded to have saved a great many lives during the emergency.
Ngunjiri (1989) Unconditional Mobility The provision of food aid contributed to internal distribution of both assets and
food aid patterns responsibilities: Stock owners tended to divide their herds in two in response to the
drought. Women, children, and older members of the household were often left with
the milk herd near centers where relief was likely to be provided. The men often
moved the remainder of the herd in search of better grazing in areas distant from
the center. Though this meant that the project tended to reach the most vulnerable
members of the household, it also meant that much of the stock put forward to
slaughter was female stock, which would have better have been retained for
rebuilding the herd after the drought.
Asmarom (1989) Unconditional Social Demoralization of the community receiving food assistance was observed: It is the
food aid relations closest thing to an Ik type of non-community that I ever observed in the Boran
country.
McCabe (1990) Unconditional Livelihood Compared with the nomadic Ngisonyoka of Southern Turkana, those in the
food aid strategies Ngilukumong, Ghiyapakuno, Ngikamatak and Ngibocheros sections of the northern
Turkana witnessed a breakdown of indigenous drought coping strategies.
Mobility The attraction of free food influenced traditional patterns of movement for many
patterns individual herd-owners. If members of the family were receiving food aid, there was
a strong incentive for the herd-owner to try to remain near the relief centre. Once
new rules requiring whole families to apply for food aid were instituted, this became
a necessity. Since the environment close to the settlements was rapidly becoming
denuded, livestock which in many instances were already stressed were forced to
forage on sparser and less nutritious plants than they would have away from the
settlements. This practice inevitably led to livestock losses through starvation or
disease. Thus, the famine relief centres were, to some extent, creating their own
clients.
Farzin (1991) Unconditional Access to The provision of food aid led to dependence upon its existence while having a
food aid food disincentive effect on the domestic food supply.
Cutler (1991) Not reported Income Undermining the domestic food economy: grain prices dropped; labourers failed to
dynamics migrate where needed to assist with the harvest; and crops were left unharvested
because the market for them became too weak.
Kilby (1993) Unconditional Mobility As a result of food aid, those from the Red Sea Hills had increased rates of
food aid patterns sedentarization.
Income The food assistance programmes in Sudan were compared with the Oxfam
dynamics restocking project in Kenya (Moris, 1988): The results suggest that it is still far more
economic to restock than to provide food aid for four years. Furthermore, food aid
had little positive effect on human mortality levels.
Publication Type of Reported Description of Impact(s)

Intervention Impact(s)
Bush (1995) Unconditional Livelihood Food aid led to a strengthening of herd growth and filled income gaps among
food aid patterns recipients.
Health There was a decline in rates of childhood malnutrition.
Social Food aid led to a strengthening of social networks.

relations
Buchanan-Smith Unconditional Livelihood The provision of food aid resulted in a reduction in livestock sales while reducing
and Barton food aid strategies overall individual-level and household debt.
(1999)
Health Similarly, rates of malnutrition decreased during the provision of food aid.
No author (2001) Not reported Health Health was substantially impacted through food aid: Even though food aid and
feeding centers are a priority during famine, attracting a large concentration of
susceptible persons to feeding centers may increase transmission of infectious
diseases such as measles and diarrhea.
Pantuliano (2002) Food-for-work Livelihood Changes in livelihood strategy patterns were observed, in particular a weakening of
strategy the Beja livelihood system.
Mobility Food aid contributed to increased sedentarization and urban drift among recipients.
patterns
Social Aid distribution created a new class of food shaikhs, which brought about a
relations modification of the existing social structure and the traditional inter-diwab dynamics
founded on exchange of resources and reciprocity.
Overseas Not reported Livelihood Although limited details were provided on the impact of food aid, the authors noted:
Development strategies The humanitarian response to the crisis in the Greater Horn of Africa has done very
Institute (2006) little, to date, to protect livelihoods.
Pantuliano (2007) Unconditional Mobility Sedentarization increased as a result of food aid, not only leading to a
food aid patterns strengthening of internal political divisions but also establishing power brokers.
Environment Potentially negative environmental impacts, but no further details reported.

Adow (2008) Food aid Livelihood Movement patterns changed in response to food aid: Many move near urban
strategies centres to seek emergency food aid.
Nangulu (2009) Not reported Livelihood Patterns in livelihood strategies have been substantially undermined by the receipt
strategies of food aid.
Dependency Food aid may lead to dependence and increase poverty among recipients.
Stockton (2012) Not reported Mobility Through the receipt of food aid, populations are reported to be increasingly
patterns sedentary. Although sedentarization is claimed to be the consequence of food aid, it
may be more a result of the development of water resources, which takes place
alongside the provision of food aid.
Lekapana (2013) Not reported Dependency Successive drought episodes have forced pastoral communities to rely on
emergency food aid, which have reinforced the cycle of dependency.
Bersaglio et al. Unconditional Livelihood Livelihood patterns have been modified in response to food aid, including the
(2015) food aid strategies creation of new spaces for community collaboration. However, this has also
contributed to a movement away from pastoralism.
Well-being A reduction in anxiety, improvement of living standards and enhanced spirituality

were observed with food aid receipt.
Access to Although food aid was offered to all, unequal distribution of available goods was
food reported.
AFRICA, OTHER
News24 (2005) Unconditional Livelihood Changes in income dynamics were reported: Many international and non-
food aid strategies governmental agencies, including Catholic Relief Services and the European Union,
have expressed concern that free food distributions after the harvests will drive millet
prices below market value and damage the chances of recovery for indebted farmers.
Cicalese et al. Not reported Security Food aid was provided during the Moroccan invasion of the Western Sahara.
(2009)
Access to Food aid provided was reported to be insufficient and unbalanced, leading to
food widespread malnutrition among the recipient population.
ASIA
No author (2002) Not reported Use of food Compared to those herders in districts only slightly or not affected: Relief efforts to
assistance distribute food to affected areas might have lessened the nutritional impact of
livestock losses among herders and their families.
Fernandez- Unconditional Dependency Food aid may have created dependence among recipients: Relief aid that helps
Gimenez et al. food aid prevent loss of life, suffering, and impoverishment in the short-term may contribute
(2012) to long-term dependence syndromes, social disparities, and lack of initiatives on the
part of both herders and local government.
REFERENCES
INCLUDED PUBLICATIONS
Adams, M.E. and Hawksley, E. (1989). Merging Relief and Development: The Case of
Darfur. Development Policy Review 7: 14369.
Adow, M. (2008). Pastoralists in Kenya. Forced Migration Review, no. 31: 34.
Asmarom Legesse (1989). Adaptation, drought, and development: Boran and Gabra
pastoralists of northern Kenya. In: Huss Ashmore, R. and Katz, S.H. (eds), African Food
Systems in Crisis. Part one: Microperspectives. Gordon and Breach, New York, USA,
pp.261-280.
Bersaglio, B., Devlin, J. and Yap, N. (2015). Contextualising Emergency Responses to

Famine among Turkana Pastoralists in Kenya. Development in Practice 25 (5). Taylor &
Francis: 688702.
Buchanan-Smith, M. and Barton, D. (1999). Evaluation of the Wajir Relief Programme:

19961998. Oxford: Oxfam.
Bush, J. (1995). The Role of Food Aid in Drought and Recovery: Oxfams North Turkana
(Kenya) Drought Relief Programme, 199294. Disasters 19 (3).
Cicalese, M.P., Ranucci, G. et al. (2009). Nutritional Impact Analysis of the Food Aids
Distributed in the Saharawi Refugee Camps: Possible Implication with Celiac Disease.
Digestive and Liver Disease 41S. Editrice Gastroenterologica Italiana S.r.l.: S22122.
doi:10.1016/S1590-8658(09)60504-1
Cutler, P. (1991). The Political Economy of Famine in Ethiopia and Sudan. Ambio 20 (5):
17678.
Farzin, Y.H. (1991). Food Aid: Positive or Negative Economic Effects in Somalia? The
Journal of Developing Areas 25 (2): 26182.
Fernandez-Gimenez, M.E., Batkhishig, B. and Batbuyan, B. (2012) Cross-Boundary and

Cross-Level Dynamics Increase Vulnerability to Severe Winter Disasters (Dzud) in Mongolia.
Global Environmental Change 22 (4): 83651. doi:10.1016/j.gloenvcha.2012.07.001
Kilby, P. (1993). Emergency Relief Programmes for Pastoral Communities. Development in

Practice 3 (2): 92102. doi:10.1080/096145249100077101
Lekapana, P.L. (2013). Socioeconomic Impacts of Drought on Pastoralists, Their Coping

Strategies, and Government Interventions in Marsabit County, Kenya. University of Nairobi.
McCabe, T. (1990). Success and Failure: The Breakdown of Traditional Drought Coping
Institutions Among the Pastoral Turkana of Kenya. Journal of Asian and African Studies XXV
(3-4): 14660.
Nangulu, A.K. (2009). Food Security and Coping Mechanisms in Marginal Areas. The Case
of West Pokot, Kenya, 19201995. Leiden: African Studies Centre.
News24 (2005). UN: Niger Food Aid to Continue. News24, 19 May.
Ngunjiri, E. (1989) Relief and recovery for pastoralists: Oxfams experience. In Downing,
T.E., Gitu, K.W. and Kamau, C.M. (eds.) Coping with Drought in Kenya: National and Local
Strategies, pp. 344351. Lynne Rienner, Boulder.
No author (2001). Mortality During a Famine Gode District, Ethiopia, July 2000. Morbidity
and Mortality Weekly Report 50 (15).
No author (2002). Nutritional Assessment of Children after Severe Winter Weather

Mongolia, June 2001. Morbidity and Mortality Weekly Report 51 (1).
Overseas Development Institute (2006). Saving Lives through Livelihoods: Critical Gaps in
the Response to the Drought in the Greater Horn of Africa. London: Humanitarian Policy
Group, ODI. Retrieved 28 April 2016, from https://www.odi.org/publications/1381-livelihoods-
response-drought-greater-horn-africa
Pantuliano, S. (2007). From Food Aid to Livelihoods Support: Rethinking the Role of WFP in
Eastern Sudan. Disasters 31 (S1): 7790.
Pantuliano, S. (2002). Sustaining Livelihoods across the RuralUrban Divide: Changes and
Challenges Facing the Beja Pastoralists of North Eastern Sudan. London: International
Institute for Environment and Development (IIED).
Snow, R. (1984). Famine relief: some unanswered questions from Africa. In: Tietze, W. (ed.),
Famine as a Geographical Phenomenon. D. Reidel Publishing, Holland.
Stockton, G. (2012). Sugar for the Tea: Assistance and the State of Pastoralism in the Horn
of Africa. Pastoralism: Research, Policy and Practice 2 (6): 110.
Taylor, W.R. (1983). An Evaluation of Supplementary Feeding in Somali Refugee Camps.

International Journal of Epidemiology 12 (4): 43336. doi:10.1093/ije/12.4.433
OTHER
Abbink, J. (1993). Famine, Gold and Guns: The Suri of Southwestern Ethiopia, 198591.
Disasters 17 (3): 21825.
Abbink, J., Askew, K. et al. (2014). Lands of the Future: Transforming Pastoral Lands and
Livelihoods in Eastern Africa. Halle/Saale: Max Planck Institute for Social Anthropology
Working Paper 154.
Abebe, D., Cullis, A. et al. (2008). Impact of a Commercial Destocking Relief Intervention in
Moyale District, Southern Ethiopia. Disasters 32 (2): 16789.
Arksey, H. and O'Malley, L. (2005) Scoping studies: towards a methodological framework.

International Journal of Social Research Methodology 8: 19-32
Action Against Hunger (2011). Final Independent Evaluation of the Provision of Food Aid
and Livelihood Disaster Prevention to the Most Dzud-Affected Herders in Mongolia. New
York: Action Against Hunger. Retrieved 27 April 2016, from
http://www.alnap.org/resource/21074
Aker, J. (2008). How Can We Avoid Another Food Crisis in Niger? Washington, D.C.: Center
for Global Development. Retrieved 27 April 2016, from
http://www.cgdev.org/publication/how-can-we-avoid-another-food-crisis-niger-essay
Bassi, M. (2010). The Politics of Space in Borana Oromo, Ethiopia: Demographics,

Elections, Identity and Customary Institutions. Journal of Eastern African Studies, 4(2), 221
46.
Bellos, A., Mulholland, K. et al. (2010). The Burden of Acute Respiratory Infections in Crisis-
Affected Populations: A Systematic Review. Conflict and Health, 4(3). Retrieved 27 April
2016, from http://conflictandhealth.biomedcentral.com/articles/10.1186/1752-1505-4-3
Bishop, C. and Hilhorst, D. (2010). From Food Aid to Food Security: The Case of the Safety
Net Policy in Ethiopia. Journal of Modern African Studies, 48(02):181202.
Blackwell, P.J. (2010). East Africas Pastoralist Emergency: Is Climate Change the Straw
That Breaks the Camels Back? Third World Quarterly 31 (8): 132138.
Boulton, J. (2012). Turkana through the Lens of Complexity: Final Report for Oxfam GB
Kenya. Nairobi: Oxfam.
Butt, B., Shortridge, A. and WinklerPrins, A.M.G.A. (2009). Pastoral Herd Management,
Drought Coping Strategies, and Cattle Mobility in Southern Kenya. Annals of the Association
of American Geographers, 99(2):30934.
Central Bank of Somalia. (n.d.) Economy and Finance. Retrieved 27 January 2016, from
http://www.somalbanca.org/economy-and-finance.html
Chi, P., Urdal, H. et al. (2015). Improving Maternal, Newborn and Womens Reproductive
Health in Crisis Settings. London: Cochrane Collaboration. Retrieved 28 April 2016, from
http://www.cochrane.org/CD011829/EPOC_improving-maternal-newborn-and-womens-
reproductive-health-crisis-settings
Clay, E.J. (2010). A Future Food Aid or Food Assistance Convention? London: Overseas
Development Institute (ODI). Retrieved 28 April 2016, from
https://www.odi.org/publications/4941-food-food-aid
Crumlish, N. and ORourke, K. (2010). A Systematic Review of Treatments for Post-

Traumatic Stress Disorder Among Refugees and Asylum-Seekers. The Journal of Nervous
and Mental Disease, 198(4):237251. Retrieved 28 April 2016, from
http://journals.lww.com/jonmd/Abstract/2010/04000/A_Systematic_Review_of_Treatments_f
or.1.aspx
Czuba, K. and ONeill, T.J. (2016). The impact of food assistance on pastoralist livelihoods in
humanitarian crises: an evidence synthesis protocol. Oxford: Oxfam GB.
DiCenso, A., Martin-Misener, R. et al. (2010). Advanced Practice Nursing in Canada:

Overview of a Decision Support Synthesis. Nursing Leadership, 23:1534. Retrieved 28
April 2016, from http://www.longwoods.com/content/22267
Doocy, S. and Tappis, H. (2015). The Effectiveness and Efficiency of Cash-based

Approaches in Emergencies: A Systematic Review. Oslo: The Campbell Collaboration.
Food and Agriculture Organization of the United Nations (FAO). (n.d.). Pastoral Systems
Worldwide. Rome: FAO. Retrieved 28 April 2016, from
http://www.fao.org/docrep/005/y2647e/y2647e02.htm
Fratkin, E. (1992). Drought and Development in Marsabit District, Kenya. Disasters 16 (2):
11930.
Fratkin, E. (2001). East African Pastoralism in Transition: Maasai, Boran, and Rendille
Cases. African Studies Review, 44(3):125. Retrieved 28 April 2016, from
http://journals.cambridge.org/action/displayAbstract?aid=9124324
Fratkin, E., Roth, E.A. and Nathan, M.A. (2004). Pastoral Sedentarization and its Effects on
Childrens Diet, Health, and Growth among Rendille of Northern Kenya. Human Ecology,
32(5):53159.
Harvey, P., Proudlock, K. et al. (2010). Food Aid and Food Assistance in Emergency and
Transitional Contexts: A Review of Current Thinking. London: Humanitarian Policy Group,
ODI. Retrieved 28 April 2016, from http://www.odi.org/resources/docs/6038.pdf
Harvey, P. and Bailey, S. (2011). Cash Transfer Programming in Emergencies. London:

Humanitarian Practice Network, ODI. Retrieved 28 April 2016, from
http://odihpn.org/resources/cash-transfer-programming-in-emergencies/
Helland, J. (1998). Institutional Erosion in the Drylands: Case of the Borana Pastoralists.
Eastern Africa Social Science Research Review, 14(2):4972.
Hesse, C. and MacGregor, J. (2006). Pastoralism: Drylands Invisible Asset? Issue Paper
142. London: International Institute for Environment and Development (IIED). Retrieved 28
April 2016, from http://pubs.iied.org/pdfs/12534IIED.pdf
Hesse, C. and Pattison, J. (2013). Ensuring Devolution Supports Adaptation and Climate
Resilient Growth in Kenya. London: IIED. Retrieved 28 April 2016, from
http://pubs.iied.org/17161IIED.html
Hesse, C. and Thbaud, B. (2006). Will Pastoral Legislation Disempower Pastoralists in the
Sahel? Indigenous Affairs. 1(6):1423.
Krtli, S. (2009). ABEK (Alternative Basic Education for Karamoja): Strategic Review.
Kampala: Save the Children in Uganda.
Krtli, S., Kaufmann, B. et al. (2015). A House Full of Trap Doors: Identifying Barriers to
Resilient Drylands in the Toolbox of Pastoral Development. London: IIED. Retrieved 28 April
2016, from http://pubs.iied.org/pdfs/10112IIED.pdf
Levine, S. (2008). Trigger Happy? Why we need to rethink the signals which determine
humanitarian intervention for more timely response. Washington, D.C.: USAID. Retrieved 28
April 2016, from
http://www.fao.org/fileadmin/user_upload/drought/docs/trigger%20happy%2012%2011%200
8%20FINAL.pdf
Levine, S. (2010). An Unromantic Look at Pastoralism in Karamoja: How Hard-Hearted

Economics Shows That Pastoral Systems Remain the Solution, and Not the Problem.
Nomadic Peoples. 14(2):14753.
London School of Hygiene and Tropical Medicine (LSHTM). (2013). The Humanitarian
Health Evidence Review. London: LSHTM.
Manley, J., Gitter, S. and Slavchevska, V. (2011). How Effective are Cash Transfer
Programs at Improving Nutritional Status. World Development, 48:133155. Retrieved 28
April 2016, from
https://www.researchgate.net/publication/228420737_How_Effective_are_Cash_Transfer_Pr
ograms_at_Improving_Nutritional_Status
Markakis, J. (2004). Pastoralism on the Margin. London: Minority Rights Group International.
Maxwell, D., Majid, N. et al. (2014). Lessons Learned from the Somalia Famine and the
Greater Horn of Africa Crisis 20112012: Desk Review of Literature. Medford,
Massachusetts: Feinstein International Center, Tufts University. Retrieved 28 April 2016,
from http://www.alnap.org/resource/12736
McPeak, J., Little, P.D. and Doss, C.R. (2012). Risk and Social Change in an African Rural
Economy: Livelihoods in pastoralist communities. Abingdon, Oxon: Routledge.
Merten, S. and Haller, T. (2009). Whose Logic? The Local Redistribution of Food Aid
Targeting Old and Chronically Sick People in Zambia. Human Organization 68 (1): 89102.
Moris, J.R. (1988). Oxfams Kenya Restocking Projects. Oxford: Oxfam.
Mkutu, K. (2001). Pastoralism and Conflict in the Horn of Africa: Report. Nairobi: Africa
Peace Forum.
Morton, J. (2006). Pastoralist Coping Strategies and Emergency Livestock Market

Intervention. In McPeak, J.G and Little, P.D. (eds), Pastoral Livestock Marketing in Eastern
Africa: Research and Policy Challenges. ITDG Publishing, Warwickshire, pp. 22746.
Oba, G. (1992). Ecological Factors in Land Use Conflicts, Land Administration and Food
Security in Turkana, Kenya. London: Overseas Development Institute.
Parkinson, J. (2009). A Review of the Evidence Base for WASH Interventions in Emergency
Responses. Oxfam/UNICEF Wash Cluster. http://washcluster.net/wp-
content/uploads/sites/5/2014/04/GWC-Review-on-the-Evidence-Base-for-WASH-
interventions-in-Emergency-Responses-200109.pdf
Rose, D. (2012). Assessing Food Security at WFP: Towards a Unified Approach. Geneva:
World Food Programme. Retrieved 19 September 2016, from
https://resources.vam.wfp.org/sites/default/files/WFP%20Fd%20Sec%20Assessment_Desig
n%20Phase%201%20Report.pdf
Salzman P. (ed.) (1980). When Nomads Settle: Processes of Sedentarization as Adaptation

and Response. New York: Praeger.
Scoones, I. (1998). Sustainable Rural Livelihoods: A Framework for Analysis. Brighton,

Sussex: Institute of Development Studies. Retrieved 28 April 2016, from
https://www.ids.ac.uk/publication/sustainable-rural-livelihoods-a-framework-for-analysis
Sharp, K. (1999). Food Aid Targeting in East Africa. Nairobi: FEWS Project.
Smith, G., MacAuslan, I. et al. (2011). New Technologies in Cash Transfer Programming
and Humanitarian Assistance. Oxford: Cash Learning Partnership (CaLP). Retrieved 28 April
2016, from http://cashlearning.org/resources/library/272-new-technologies-in-cash-transfer-
programming-and-humanitarian-assistance
Spangaro, J., Zwi, A.B. et al. (2013). What is the Evidence of the Impact of Initiatives to
Reduce Risk and Incidence of Sexual Violence in Conflict and Post-Conflict Zones and
Other Humanitarian Crises in Lower and Middle Income Countries? London: Institute of
Education, University of London. Retrieved 28 April 2016, from
http://eppi.ioe.ac.uk/cms/LinkClick.aspx?fileticket=Q-WhIlmx17U%3d
Turnbull, C. (1974). The Mountain People. London: Pan Books.
Department for International Development (DFID). (2014). Assessing the Strength of

Evidence. London: DFID. Retrieved 13 July 2016, from
https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/291982/HTN-
strength-evidence-march2014.pdf
Vu, A., Atif, A. et al. (2014). The Prevalence of Sexual Violence Among Female Refugees in
Complex Humanitarian Emergencies: A Systematic Review and Meta-Analysis. Public
Library of Science Currents 6. Retrieved 28 April 2016, from
http://currents.plos.org/disasters/article/the-prevalence-of-sexual-violence-among-female-
refugees-in-complex-humanitarian-emergencies-a-systematic-review-and-meta-analysis/
World Food Programme (WFP). (n.d.) Food Assistance for Assets. Retrieved 27 January
2016, from https://www.wfp.org/food-assets
Published by Oxfam GB under ISBN 978-0-85598-876-0 in January 2017.
Oxfam GB, Oxfam House, John Smith Drive, Cowley, Oxford, OX4 2JY, UK.
OXFAM
Oxfam is an international confederation of 20 organizations networked together in more than 90 countries, as
part of a global movement for change, to build a future free from the injustice of poverty. Please write to any
of the agencies for further information, or visit www.oxfam.org.

The Impact of In-Kind Food Assistance On Pastoralist Livelihoods in Humanitarian Crises

Uploaded by

Copyright:

Available Formats

The Impact of In-Kind Food Assistance On Pastoralist Livelihoods in Humanitarian Crises

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

The Impact of In-Kind Food Assistance On Pastoralist Livelihoods in Humanitarian Crises

Uploaded by

Copyright:

Available Formats

JANUARY 2017

THE IMPACT OF IN-KIND FOOD ASSISTANCE

About the research team

Copyright Oxfam GB 2017

APPENDIX 1: DATABASE SEARCH STRATEGY STRINGS 30

What are the impacts of in-kind food assistance in humanitarian

Thematic Number of Evidence Research question*** Findings

What evidence was eligible for synthesis?

are mainly primary studies (n=20)

lasted between one and 20 years

involved unconditional provision of food in-kind (n=23) or in exchange for assets

What is the state of the evidence?

Definition of common terms included in this review

1.2 RATIONALE AND PURPOSE

The feasibility of investigating these possible effects is dependent on the availability of

1.3 RESEARCH QUESTIONS

Population, intervention, control and outcome (PICO) parameters

The research process involved the following steps:

implementation of the search strategy

assessment of the strength of evidence and causal inferences in included publications

data extraction and evidence synthesis.

2.1 PROTOCOL DEVELOPMENT

2.2 IMPLEMENTATION OF THE SEARCH STRATEGY

2.3 DOCUMENT SCREENING

Round 1 (R1): Title and abstract review for potential relevance

Does the study report on a pastoralist population (Appendix 2)? (Yes/No/Unclear)

Round 2 (R2): Full text review for potential relevance

2.4 EVIDENCE ASSESSMENT

the size of the body of evidence:

the context in which the evidence is set:

the overall strength of the body of evidence:

Figure 2.1: DFID strength of evidence assessment criteria for individual

Principles of Associated questions

Source: DFID (2014: 14)

For a summary of the findings, see Figure 4.8.

2.5 DATA EXTRACTION AND EVIDENCE SYNTHESIS

3.1 SUMMARY OF REVIEWED PUBLICATIONS

Figure 3.1: Flowchart demonstrating the process of identification through to

Records identified Additional records identified

Records screened (R1):

Records screened (R2):

Documents included in review

3.2 INCLUDED PUBLICATIONS

3.3 OUTCOMES REPORTED BY INCLUDED PUBLICATIONS

3.4 ASSESSMENT OF THE STRENGTH OF EVIDENCE AND

Because of the large number of low-strength-of-evidence studies among the included

Figure 3.2: Overall strength of the body of evidence

Quality of the body of evidence Low

3.5 NARRATIVE SYNTHESIS OF RESULTS

Changes in livelihood strategies and asset and income dynamics

Changes in livelihood strategies

Changes in asset dynamics and income

The results reported in this section are summarized in Figure 4.2.

The results reported in this section are summarized in Figure 4.3.

Access to food assistance, use of food assistance and

The results reported in this section are summarized in Figure 4.4.

Household- and individual-level socio-demographic shifts

There is effectively no evidence in the publications on household- and individual-level shifts