Content uploaded by Jeroen Dijkman
Author content
All content in this area was uploaded by Jeroen Dijkman on Dec 22, 2016
Content may be subject to copyright.
1
Innovation
R
esponse Capacity in
Relation to Livestock-Related
Emergencies in Africa
Dijkman, J., Hall, A., Steglich, M., Sones, K.,
Keskin, E., Adwera, A, and Wakungu, J.
IGAD LPI Working Pape
r
No. 03
-
10
A Living
from L
ivestock
2
TABLE OF CONTENTS
ACRONYMS ......................................................................................................... 3
LIST OF TABLES ................................................................................................... 5
1. INTRODUCTION ................................................................................................. 6
1.1 The livestock sector in the IGAD region .............................................................. 6
1.2 Nature of innovation response capacity and the building of policy-relevant
innovation capacity in the context of livestock-related emergencies in East Africa ........ 7
1.3 The innovation systems concept ...................................................................... 7
2. THE CASE STUDIES ........................................................................................... 13
2.1 Selection of Case Studies .............................................................................. 13
2.2 Case Study One: The 2006/07 Rift Valley Fever Outbreak in
East Africa and the Horn ............................................................................... 13
2.2.1 The Kenyan experience .......................................................................... 13
2.2.2 The Ethiopian experience ....................................................................... 19
2.2.3 The outbreak unraveled ......................................................................... 23
2.3 Case Study 2: Drought in South-Eastern Ethiopia, 2005/06 .................................... 31
2.3.1 The Southern Somali Ecosystem ............................................................... 31
2.3.2 The drought of 2005/06 ......................................................................... 33
2.3.3 Drought emergency management in Ethiopia ............................................... 33
2.3.4 Responses to the 2005/06 drought ............................................................ 35
2.3.5 External support to the Ethiopian government ............................................. 38
2.3.6 The drought response unraveled .............................................................. 39
3. THE WORKSHOPS AND ANALYTICAL FRAMEWORK ..................................................... 43
4. SO WHAT DO THE CASE STUDIES AND THE WORKSHOPS TELL US ABOUT
BUILDING INNOVATION RESPONSE CAPACITY? .......................................................... 45
4.1 Results of the Case Studies ........................................................................... 45
4.2 Actors and their roles .................................................................................. 45
4.3 Patterns of interaction ................................................................................. 47
4.4 Habits and practices .................................................................................... 48
4.5 Enabling environment .................................................................................. 49
5. CONCLUSION .................................................................................................. 50
5.1 Where to Start: The 4cs that build Innovation Response Capacity ............................ 50
REFERENCES ..................................................................................................... 53
APPENDIX.…..……………………………………………………………………….....……………………………………………………...54
3
ACRONYMS
ACTS Africa Centre for Technology Studies
CAHWs community-based animal health workers
CBPP contagious bovine pleuropneumonia
CCPP contagious caprine pleuropneumonia
CDC Centers for Disease Control
DFID UK Government Department for International Development
DOD-GEIS US Department of Defence Global Emerging Infections System
DPPA Disaster Prevention and Preparedness Agency, Ethiopia
DPPC Disaster Prevention and Preparedness Commission, Ethiopia
DVS Director of Veterinary Services
EMPRES Emergency Prevention System for Transboundary Animal and Plant Pests and
Diseases of FAO
EWD Early Warning Department, Ethiopia
FAO Food and Agriculture Organization of the United Nations
FMOH Federal Ministry of Health
FSCB Food Security Coordination Bureau, Ethiopia
GALVmed Global Alliance for Livestock Veterinary Medicines
GDP gross domestic product
GOARN Global Outbreak Alert and Response Network
GoE Government of Ethiopia
GoK Government of Kenya
IGAD-LPI Intergovernmental Authority on Development’s Livestock Policy Initiative
IgG Immunoglobulin G
IgM Immunoglobulin M
ILRI International Livestock Research Institute
JH
A
Joint Humanitarian Appeal
KEMRI Kenya Medical Research Institute
LINK Learning, Innovation and Knowledge, UNU-MERIT
MOARD Ministry of Agriculture and Rural Development
MoH Ministry of Health
MoLFD Ministry of Livestock & Fisheries Development, Kenya
MSF Médecins Sans Frontières
NASA National Aeronautics and Space Administration
NDVI normalized density vegetation index
NEP North Eastern Province, Kenya
NGO non-government organization
NHLS National Health Laboratory Service
NICD National Institute of Communicable Diseases
OIE World Organization for Animal Health
PLI Pastoralist Livelihood Initiative
PPLPI FAO’s Pro-poor Livestock Policy Initiative
PP
R
peste des petits ruminants
PRSP poverty reduction strategy papers
4
PSNP Productive Safety Net Programme
RVF Rift Valley fever
SAPs structural adjustment programs
SCF (US) Save the Children Fund (United States)
SNNP
R
Southern Nations, Nationalities, and People's Region, Ethiopia
SPS-LMM Sanitary and Phytosanitary Standards and Livestock and Meat
Marketing
UAE United Arab Emirates
UN/CERF United Nations Central Emergency Relief Fund
UNICEF United Nations Children’s Fund
UN-OCHA United Nations Office for the Coordination of Humanitarian Affairs
USAID United States Agency for International Development
WFP World Food Program
WHO World Health Organization
5
LIST OF TABLES
Table 1: Sequence of events during the 2006/07 Kenya RVF outbreak ................... 17
Table 2: Sequence of events for response in Ethiopia to RVF in East Africa ............. 22
Table 3: An example of the successive levels of assessments and related activities
That took place in Borena zone (Oromiya Region) and Leban zone
(Somali Region) during the early stages of the 2005/06 drought .................. 35
Table 4: Sequence of events in the 2005/06 drought .......................................... 37
6
1. INTRODUCTION
1.1 The livestock sector in the IGAD region
Livestock play important roles in both national economies and households’ livelihoods in the IGAD
region. Excluding Somalia, livestock account for around 15% of the gross domestic product (GDP) of
IGAD member countries. The IGAD region supports an estimated 68 million livestock units, with
Ethiopia having the highest livestock populations in Africa. Livestock products constitute an
important category of exports in the region: for example, in Somalia in abundant rainfall years,
livestock can account for up to 80% of total exports. In Ethiopia, livestock skins and hides are the
second most important export commodity after coffee. It is estimated that there are 43 million
poor livestock keepers in the region, estimated to account for 60% of the total poor.
Livestock-based livelihoods in the IGAD region have recently been reviewed by Sandford and Ashley
(2008). They categorized the region’s livestock-dependent livelihoods into four categories:
• pastoral area
• mixed farming
• large-scale livestock enterprises
• urban and landless
The study focused on pastoral areas. Pastoral livelihoods were further subdivided into pastoral,
agro-pastoral and pastoral drop-outs. Pastoral livelihoods were characterized as being highly
dependent on livestock, including sheep, goats, cattle, camels and donkeys. In recent decades it
was noted that both sources of food and income had diversified: in the past, milk constituted a
major component of diets but now high-value livestock and livestock products were sold to enable
grain to be purchased. Diversified income sources now included trading, sales of wood, charcoal,
palm leaf products, gums and resins, and often also sales of labour. Pastoralists utilize a number of
traditional coping strategies to help them cope with shocks, especially the droughts which are a
pervasive feature of the semi-arid and arid areas in which they live. These strategies include
mobility, whereby herds are trekked to access browse, grazing and water resources; herd
accumulation, which recognizes that owning large herds ahead of a drought will help ensure that a
viable sized herd survives after the drought; and livestock sharing and loan arrangements, whereby
pastoralists loan each other animals to enable faster post-drought recovery, and also share animals
between each others’ herds, thereby spreading the risk amongst more herds.
Agro-pastoral livelihoods in the region include both those who have traditionally combined crop
production with extensive rearing of livestock as well as those who have been ‘forced’ to grow
crops because they can no longer rely entirely on their livestock. As for pastoralists, livestock play
an important role in coping strategies, especially in relation to drought, although crops can also
play a role in drought recovery - in some cases enabling a return to full pastoralism. In addition to
providing food and income, agro-pastoralists often also use cattle and camels for ploughing.
Pastoral drop-outs are often female-headed households who have lost their herds and now have
very few productive assets except their labour. They tend to move to urban areas where they may
7
be able to sell their labour and access food aid, although their incomes are very low and unreliable.
Although some still have a few goats, sheep or in some cases poultry, they are unlikely to ever
return to a pastoralism: neither customary nor external restocking initiatives are usually sufficient
to enable them to be rehabilitated to viable livelihoods dependent on livestock. Frequent drought
is making it increasingly difficult for pastoralists to rebuild their herds and flocks in the decreasing
inter-drought interval, which is likely to create more and more pastoral drop-outs.
1.2 Nature of innovation response capacity and the building of policy-
relevant innovation capacity in the context of livestock-related
emergencies in East Africa
The study explored the nature of innovation response capacity and the building of policy-relevant
innovation capacity in the context of livestock-related emergencies in East Africa.
The work described has been carried out on behalf of the Intergovernmental Authority on
Development’s Livestock Policy Initiative (IGAD-LPI), a regional sister initiative to the FAO’s Pro-
poor Livestock Policy Initiative (PPLPI). IGAD and FAO have formed a partnership with the objective
of enhancing the contribution of livestock to sustainable food security and poverty reduction in the
region through the strengthening of policy formulation and implementation capacity of relevant
organizations and stakeholders, including IGAD and its member state governments.
Through two case studies, the report describes innovation response capacity in relation to
livestock-related emergencies in East Africa using the examples of a recent regional drought and
the zoonotic disease Rift Valley fever, situating these in on-going conceptual and policy debates.
The case studies are described and then analysed using an analytical framework that considers,
actors and their roles, patterns of interaction, habits and practices, and presence or absence of an
enabling environment. As part of this study, policy-relevant innovation capacity was explored
during two workshops which brought together livestock sector policy makers from Djibouti,
Ethiopia, Kenya, Somalia, Sudan and Uganda, together with representatives of international
organizations and NGOs, to focus on process monitoring, policy dialogue and the sharing of
experiences. The report reveals a set of policy relevant lessons and indicative practices for policy
makers and practitioners on building innovation response capacity in the IGAD region.
1.3 The innovation systems concept
The idea of an innovation system is now widely used to explore innovation processes and capacities
at both national and sectoral levels (Lundvall 1992; Freeman 1995; Malerba 2002) in both the
developed and increasingly developing economies (Hall et al 2002; World Bank 2006).
At its simplest, the concept departs from earlier notions of innovation as a research-driven process
of technology transfer. Instead it views innovation as a social process where different sources of
knowledge and ideas are put into use. The concept gives centre stage to two interconnected
dimension of the innovation process.
8
The first is the interaction of different players in economic systems; the roles they play and the
way their interaction facilitates the transmission, adaptation and use of ideas, and enables learning
and innovation.
The second dimension is the way this process is located in, shaped by and responds to various
contexts: the habits and practices – institutions - of the various actors involved in innovation; the
historical, cultural and political setting that shape habits, practice and styles of innovation; and the
enabling environment that includes some of these other contextual elements, but also includes
policies and infrastructure as well as the market itself as a mechanism for providing incentives for
entrepreneurial activity.
Two other important considerations that the innovation systems framework allows one to reveal are
the dynamics of the processes involved and the capacities that emerge at a systems level. So, while
the concept recognises the importance of certain types of relationships and linkage that mediate
information flows, it also recognises that in ever changing environments (markets, policy,
technology, climate), patterns of linkages need to change to meet new conditions and demands.
The recognition of this as a systems phenomenon is arguably the critical point of departure for
contemporary thinking on innovation. Not only does it recognise the interaction of many individual
parts, and the non-linearity of the outcomes of these interactions, but also that these networks of
interacting elements have emergent properties. That is to say, these systems have properties which
are more than the sum of the constituent parts, which cannot be accounted for by analysis of
individual elements of the system. It is for this reason that institutional settings of actors - ways of
working – assume such significance as this is in a sense the ‘hidden hand’ that determines how the
system operates. By the same reasoning it is why science, technology and innovation policy focus is
shifting towards considering capacity development in terms of the behaviour of systems rather than
in terms of quantum of research or the nature of technology transfer elements.
Innovation capacity in highly dynamic environments
The focus of this document is the responsiveness of parts of the livestock sector in East Africa to
changing contexts and, in particular, responses to rapidly changing conditions, caused by drought
and the disease Rift Valley fever (RVF). For natural resources based industries, such as livestock in
developing countries, this has become a particularly important concern. In a sector where rapid
(and often unpredictable) change in market, technological, social and environmental circumstances
calls the tune, more thought needs to be given to the kind of capacity that needs to be developed
to be able to respond to the frequent and often unpredictable changes - without compromising the
contribution that livestock can make to sustainable and inclusive growth. The ever increasing rate
of change in the sector’s markets means that responsiveness is likely to be the critical element of
innovation capacity.
The idea of innovation response capacity is not really well defined in the literature. It would seem,
however, that its broad contours would be very similar to what is generally discussed as innovation
capacity, as has been outlined above, but with specific analytical attention given to two aspect.
First are the factors, arrangements and attributes that enable rapid response; factors that allow
timely responses in a rapidly changing environment are going to be crucial. This implies that the
response capacity must include both mechanisms for early warning of up-coming changes, as well
as mechanisms for dealing with the opportunities and challenges that arise from these.
9
The second aspect that needs attention is the specificity of the response. That is to say the
response has to be tailored to the specific characteristics of the opportunity or challenge as well as
the context of the response: for example, it is no good having a general capacity to respond to
changing consumer demands if the demand is for organic produce - the capacity to innovate must
be specific. Similarly, while a country with a strong analytical science base and effectively
enforced food standards may be able to respond in certain ways, a country without these
competencies and skills would need to respond to consumer demands differently - for example,
with external assistance.
Analytical elements for exploring innovation response capacity
What then might be the specific analytical elements that could be used to explore innovation
response capacity? A useful starting point is to follow the four-point analytical framework
developed by the World Bank to investigate agricultural innovation capacity (World Bank 2006).
These four elements are:
• Actors and their roles
• Patterns of interaction
• Habits and practices
• Enabling environment.
Actors and their roles. Innovation systems’ principles tell us that the process of innovation
requires a diversity of actors, which might include entrepreneurs, research and training
organisations, public policy bodies and civil society organisations. In terms of responsiveness, it is
not possible to be prescriptive about which actors should be present. However the following roles
are expected to be important:
• Sector coordination: This is important as it provides coherence to actions of the different
actors in the sector. Not only does this facilitate the patterns of interaction needed for
innovation, it also helps coordinate and speed the actions needed to respond. This response
might be new groupings of actors, dissemination of new information or the introduction of
new practices. Sometimes, coordination might be more concerned with third party
brokerage rather than sector-wide coordination.
• Entrepreneurship: This role is important because it is by grasping opportunities and taking
risks that sectors innovate in response to changing market opportunities. If this role is not
well developed, responses to opportunities will be weak. While this seems obvious,
entrepreneurship is not uniformly developed across all sectors and countries. Although this
role could, in theory, be played by either the public or private sectors, the private sector is
now usually thought to be better suited to this role. Again, this can vary from country to
country and sector to sector.
• Providing knowledge of the future: This role is important in response capacity as it is the
way in which early warning information about up-coming challenges and opportunities is
collected and transmitted to entrepreneurs and policy makers. Buyers and others in market
chains, for example, might play this role in relation to information about changing market
demands and policy and regulatory changes in distant markets. Informal networks linking
10
sectors to policy makers, meanwhile, might give early warning on up-coming policy
changes. Farmers’ organisations and industry association might also play this role. Some
countries might have formal foresight committees or standing panels on certain export
commodities with the specific purpose of collating expert opinions on future market or
technology conditions.
• Research: Not all innovation requires formal research; by the same logic neither does
responsiveness. Nevertheless, research plays an important role in the innovation process as
a specialist form of knowledge creation: it both addresses unforeseen challenges and
opportunities and provides opportunities through new technology. Both private and public
organisations can play a research role, but in many developing countries the public sector is
dominant. This role is often closely connected to tertiary education.
• Service provision: There are a number of services which are required for innovation and
which contribute to responsiveness, including technical advisory, financial and auditing
services to help with regulatory compliance.
• Ensuring socially desirable outcomes: Ensuring socially desirable outcomes is no longer a
non-market issue. As global consumers become more sensitive to the social and ethical
consequences of the good and services they purchase, non-regulator policing and assistance
with compliance to these standards is an important role in responding to the ever-more
sophisticated demands of the market. The advocacy role of organisations is also important
in this context; for example alerting society to malpractices and social inequity. It is
usually civil society organisations that play this role, but the public and even the private
sectors may also do so in certain sectors and countries.
Patterns of interaction. Again it is dangerous to be too prescriptive of the sorts of patterns of
interaction that should be present in innovation response capacity. Indeed, the innovation systems
concept tells us that patterns of interaction will be shaped by the local context and the particular
innovation challenges or opportunity that are being addressed. These patterns are dynamic and will
change over time, in response to changing patterns of challenges and opportunities. What one
would expect to see is evidence of a loose network of linkages between actors in the sector that
provides coherence and which acts as a foundation for more concrete forms of linkage and
collaboration in times of need. This informal cohesiveness is often discussed in the innovation
literature in terms of social capital (Fukuda-Parr, Lopes and Malik 2002). In analysing patterns of
interaction it is important to bear in mind that forms of interactions that allow two-way flows of
information are the most important. Also, while frequency of interaction is an important
consideration, quality of interaction is equally important. Critical aspects of patterns of interaction
specifically relevant to innovation responsiveness include:
• Links to consumers: Consumers provide critical information about preferences and how
these are changing. Often, especially in the case of export markets, interaction with buyers
(middlemen) is a critical source of information on consumer preferences in distant markets.
Trade fairs can facilitate a similar form of interaction.
11
• Links to specialist technical knowledge: Changing market and policy conditions often
involve the need for technical upgrading. This might be in terms of production and process
technology, production organisation, packaging or marketing. This might require linkages to
research organisations in cases where problems and opportunities require new knowledge.
More often, however, it will involve linkages to technical services and/or organizations
which can help adapt existing technologies and processes to a new situation.
• Mechanisms that facilitate interaction: Sector coordination bodies, such as government
agencies and industry associations exist in some sectors to help foster interaction between
different actors. Since innovation response capacity often involves rapid reconfiguration of
patterns of interaction, mechanisms that facilitate this are likely to be important. One such
mechanism is to have a high degree of organization in the sector, i.e. farmers associations,
women’s associations, self-help groups, industry associations, export associations and
cooperatives.
Habits and practices. Habits and practices, or institutions, are the most intangible element of
innovation capacity, but also the most important. They are the mainly informal rules that
determine how people behave and shape the way they do things. The innovation systems concept
gives particular attention to the institutions that affect the processes of interaction, information
sharing and learning. In relation to innovation response capacity there are a number of basic ways
of working that are likely to be important. These are some of the institutions that underpin them:
• Attitudes towards change: Most people tend to be conservative by nature and fear and
avoid change. Public and private sectors organizations exhibit varying degrees of reluctance
to change. This is most marked in public sector organisations because of the rigid rules and
hierarchies that often characterize these organizations. Although these same
characteristics can sometimes also be seen in the private sector, because this attitude
seriously undermines their ability to respond to markets and competition, these type of
organization do not remain in business long.
• Trust: Trust is an important lubricant in social relationship. Since innovation is a social
process of interaction, information sharing and learning, trust is a necessary ingredient in
the relationships needed to underpin innovation. By the same argument responsiveness to
rapidly changing circumstances requires a sufficient degree of trust among actors in a
sector to ease the creation of the new relationships needed for innovation. Trust often
manifests its self as willingness to cooperate with other actors.
• Shared identity: Another institution that can help coherence, and thus aid responsiveness,
is the degree of shared identity or the sense of ‘belonging’ that exists among actors in a
sector. This is importance for a number of reasons: Increasingly, and particularly for export
sectors, competitive pressures come from other countries rather than domestic
competitors. Responding to this type of competition requires collaboration and the
recognition that survival of the sector is a collective responsibility. An example is the need
to respond to quality demands in distant markets: often this is about building a sector’s
reputation for quality and this requires all actors to recognise this as a shared
responsibility. The ‘we are us’ mentality is therefore an important aspect of innovation
response capacity.
12
• Learningness: The innovation systems concept gives very strong emphasis to learning and
the habits and relationship that enable this process. These are often difficult to distinguish
and may best be revealed by identifying the way practices and behaviour of actors or
sectors have changed as a result of learning from experience, research, interaction or
searching for information. Attitudes that support this include a propensity for
experimentation, risk taking and organizational ‘space’ for trying new things out. Some
organizations may have structured learning practices as part of their management strategy,
or they may utilise external evaluation. Since much learning accumulates in the form of
tacit knowledge, organizations with low staff turn-over may be better placed to learn and
build capacity in the longer term. Learning is particularly important in relation to response
capacity as, in dynamic environments, the ability to learn how to cope with a continuous
succession of largely unpredictable events is an essential attribute for success.
• National culture: The originators of the innovations concept rightly point out that there
are different national styles of innovation and that they emerge from cultural, historical
and political contexts at a national level (Lundvall 1992; Freeman 1995). National culture is
an emotive subject and one needs to be careful not to fall into the trap of dismissing
certain national cultural traits as constraints to innovation and innovation responsiveness.
But national cultures also evolve; certain cultural stereotypes can become quickly
outdated. Nevertheless, attributes like the degree of social cohesiveness, or the degree of
social hierarchies found in society as a whole, can reflect on innovation responsiveness.
Similarly, accepted views on the role of certain actors in society, particularly the public
and private sectors, can similarly be reflected in capacity. Analytically, this allows a more
nuanced diagnosis of existing patterns of capacity; from an intervention perspective this
provides the contextual backdrop for the design of capacity strengthening activities.
Enabling environment. The enabling environment is the wider set of policies and institutions in
which the innovation process is situated. Much of the enabling environment manifests itself through
the factors already discussed. For example, agricultural science and technology policy often
determine the degree of interaction between researchers and actors in the productive sector. More
specific factors might include monetary policy, infrastructure, level of corruption, the
effectiveness of the legal system, education practices, regulatory regimes and sector governance
amongst others. A final aspect of the enabling environment is the presence of a strong market or,
in its absence, policy triggers that provide the incentives for innovation.
13
2. THE CASE STUDIES
2.1 Selection of Case Studies
In 2007/08, IGAD-LPI organized extensive stakeholder meetings in the IGAD member countries to
identify livestock policy priority areas. They also subsequently facilitated the establishment of
policy hubs to support efforts towards the harmonization of a regional animal health policy
framework and to intensify policy engagement for the improved incorporation of livestock sector
concerns in national poverty reduction strategy papers (PRSP) processes. This research study, which
explores innovation response capacity in relation to livestock emergencies, is designed to inform
both the policy hubs as well as the broader group of stakeholders, with the objective of assisting
them in their endeavours to promote pro-poor policy change in the region’s livestock sector.
The choice of case studies for inclusion in this study emerged from national stakeholder meetings
facilitated by IGAD-LPI in 2006/07. A key objective of these meetings was to identify priority
livestock policy areas. Two such areas that emerged from this process were recent regional
droughts and the 2006/07 RVF outbreak as relevant scenarios to examine response capacities. To
compare livestock emergency responses across countries the RVF case study focuses on Kenya and
Ethiopia.
2.2 Case Study One: The 2006/07 Rift Valley Fever Outbreak in East
Africa and the Horn
2.2.1 The Kenyan experience
• Introduction
North-east Kenya is a semi-arid region inhabited mainly by ethnically Somali pastoralists. The
region is remote with poor infrastructure and communication links, and has low levels of public
services. The security situation is hazardous; violent cross-border cattle raiding incidents and
bloody disputes over scarce grazing and water resources, exacerbated by the ready availability of
automatic weapons from neighbouring Somalia, are not uncommon.
The region has erratic and unreliable rainfall: in early 2006 the Kenya Government declared the
drought a national disaster – reports indicated that in north-eastern Kenya three-quarters of
livestock had died and two-thirds of the population were reliant on food aid for their survival; in
late 2006 the three-year long drought gave way to a prolonged period of heavy rainfall and
widespread and persistent flooding.
Pastoralists’ livelihoods in the region depend on the extensive rearing of sheep, goats, cattle,
camels and donkeys. These animals are subject to large-scale die-offs due to droughts, and in
recent decades the increased frequency of extreme weather events has caused traditional coping
mechanisms, such as building up herd and flock numbers during the good years, to fail. Livestock
are also at risk from various diseases including: in cattle, lumpy skin disease, foot-and-mouth
disease and contagious bovine pleuropneumonia (CBPP); in sheep and goats, peste des petits
14
ruminants (PPR), sheep and goat pox, contagious caprine pleuropneumonia (CCPP) and foot-rot
during floods; in camels, Surra and a number of as yet uncharacterized diseases; and for all these
species Rift Valley fever.
• The 2006/07 RVF outbreak
In early October 2006, Somali pastoralists in Sangailo, a small village in Ijara District, North Eastern
Province, Kenya first noticed worrying symptoms in their livestock: abortion occurring at more than
four-times the rate during years when no drought or major disease epidemic occurs - (more than 8
out of every 10 pregnant sheep and goats aborted), animals with bloody noses which were also
frothing at the mouth, and a very high mortality rate amongst lambs and kids. Coming shortly after
a period of heavy and persistent rain and widespread flooding, and coinciding with the presence of
swarms of a rarely-seen, large, black and white mosquito they knew this was bad news – it was
undoubtedly the disease known locally as sandik or abbur and more widely as Rift Valley fever
(RVF).
What is Rift Valley fever?
Rift Valley fever is a viral disease that affects sheep, goats, cattle and camels as well as people.
The epidemiology of the disease varies: in forested zones of Africa it tends to be endemic,
occurring after the onset of the rainy season; in high rainfall grassland areas it occurs epidemically
in 5 to 15 year cycles; in the semi-arid zones of the Greater Horn of Africa – with which this case
study is concerned – it occurs as epidemics in years in which heavy rainfall leads to flooding with
surface water persisting for at least four weeks. This provides ideal conditions for mosquito eggs,
which have lain dormant in the soil since the previous flood, to hatch. The RVF virus can survive in
the eggs of certain species of Aedes mosquitoes for many years. Once hatched, the emerging adult
female mosquitoes infect livestock with the virus as they feed on their blood. The floodwaters also
provide ideal breeding conditions for a wide variety of other mosquito and midge species which act
as secondary vectors, further transmitting the virus to susceptible livestock.
Typically, herders, slaughterhouse workers, butchers and animal health workers are most affected
by RVF. This suggests that infection in humans occurs mainly through direct contact with blood and
tissues, such as aborted foetuses, of RVF-infected animals. In livestock, the disease mainly results
in abortions and the death of young animals, especially of sheep and goats. In people, although the
disease is usually mild and flu-like, in a minority of cases it can cause serious complications and
even death. There is no effective treatment for RVF; affected patients (human or animal) can only
be provided with supportive therapies and be treated for inter-current infections. Vaccines are
commercially available for use in livestock, although the currently available ones are not ideal: for
example, the vaccine used during outbreaks causes abortion in pregnant animals. Vaccination of
animals in areas experiencing active infections is not recommended as the act of vaccination can
further transmit the virus between animals via contaminated multi-use needles. No human vaccines
are currently commercially available. In Kenya, RVF outbreaks have previously been recorded in
1931, 1951/53, 1961/63, 1967/68, 1977/79, 1997/98 and most recently 2006/07.
Around the same time, in other villages in Ijara and the neighbouring district of Garissa, pastoralists
were observing similar symptoms. Many of these pastoralists remembered the previous occasion
when they had witnessed similar symptoms and conditions: in 1997/98 an RVF outbreak had caused
15
major losses in their herds and flocks and resulted in the death of family and clan members.
Initially, pastoralists and community-based animal health workers (CAHWs) tried to treat their
animals using a combination of traditional and more modern (though inappropriate) methods. Only
when these failed did the CAHWs report the disease to the local district veterinary officer on behalf
of their communities. In some cases this was done via radio-calls provided by the local
administration; in other cases the reports were simply passed through local chiefs or the police.
Some villages were unable to report the disease outbreak because they were too remote and lacked
communication infrastructure.
District- and provincial-level veterinary officers reported that they were unable to mount an
immediate investigation or response. Due to flooding, many of the roads were impassable and in
addition they lacked sufficient serviceable vehicles, equipment and protective clothing, or
adequate personnel and funds. It also appears that, initially, they assumed that the diagnosis by
the pastoralists and CAHWs was wrong; that the probable cause of the disease was trypanosomiasis
- which also causes abortion in livestock.
Meanwhile, in the United States, scientists at the Department of Defence Global Emerging
Infections System (GEIS) were compiling and publishing on the internet their regular monthly
reports and maps of the relative risk of RVF. These are based on analysis and interpretation of
satellite-derived observations of sea-surface temperatures, cloudiness, rainfall and vegetation
dynamics. The data are collected daily by several satellites as part of the global climate observing
efforts of the National Aeronautics and Space Administration (NASA) and the National Oceanic and
Atmospheric Administration and are associated with heavy rainfall. Their September 2006 report
gave the impression that, overall, the RVF risk in East Africa was relatively low. Only in their
October report did GEIS suggest that there was an elevated risk of an RVF outbreak in East Africa,
including northern Kenya, and that “ground surveillance and response to disease outbreaks” was
likely to be required during the next 2-6 months. This warning was picked up by the Emergency
Prevention System for Transboundary Animal and Plant Pests and Diseases (EMPRES) of the Food and
Agriculture Organization (FAO) in Rome. A special issue of EMPRES Watch was published on their
website and FAO and the World Health Organization (WHO) also sent warnings directly to the
relevant national veterinary and public health authorities. However, this did not occur until
November - more than a month after the pastoralists observed the first suspected cases in livestock
and even after the first suspected cases in people started to emerge.
As suspected cases of RVF in people started to be reported or present to health facilities in North
Eastern Province in mid December, the Kenyan Government’s medical and veterinary authorities
initiated their response, with the support of a wide range of partners from international
organizations and NGOs. A major constraint to the government in mounting its response was the
lack of an RVF contingency plan and the absence of emergency funding accessible to the veterinary
department: Because the situation was not officially declared a national disaster1, this prevented
the release of government emergency funding for a joint medical/veterinary response. These
difficulties were to some extent overcome by the support and active participation of donors,
international organizations and medical and veterinary NGOs.
1Toputthisincontext,theKenyaGovernmentdeclaredfournationaldisastersduetodroughtorflood
between1993and2003.
16
In an early response, a helicopter was provided by the Ministry of Health (MoH) and dispatched to
the flooded North Eastern Province. Specimens were collected from severely ill patients who were
exhibiting haemorrhagic symptoms and sent to the Centers for Disease Control (CDC)-Kenya Medical
Research Institute (KEMRI) laboratory in Nairobi in mid-December where they were confirmed
positive for RVF. Immediately thereafter a task force, headed up by the district commissioner, was
formed. The task force coordinated the interventions of the various agencies, including provincial
and district medical and veterinary services, police, international organizations and NGOs. Several
technical teams, each comprised of two state veterinary surgeons, five medical doctors and three
NGO personnel, were formed in early January 2007 and sent to the affected areas to manage relief
and emergency interventions, including the distribution of food.
RVF is a World Organization for Animal Health (OIE) notifiable disease; the Kenya Director of
Veterinary Services (DVS) formally reported the suspected outbreak to OIE on 4th December 2006.
The diagnosis of RVF in samples collected from sick livestock was confirmed on the 22nd December.
An early veterinary intervention was the closure of Garissa livestock market, the largest in north-
eastern Kenya, on the 17th December. This was followed by banning the slaughter and movement of
livestock in the affected areas. The timing of these measures was delicate, coming as they did
shortly before Eid-Ul-Adha, a religious festival marking the end of the Hajj, when the district’s
predominantly Muslim population usually slaughter livestock. The Garissa District Chief Medical
Officer estimated that more than 20,000 goats would usually be slaughtered in the district at this
time. Urgent meetings were held with local religious leaders in an attempt to dissuade local
residents from undertaking their customary home slaughter rites.
Vaccination of livestock against RVF in at-risk but thus far uninfected areas started on the 8th
January 2007. Livestock were also treated for other infections and administered with pour-on
insecticide. Interventions led by the MoH included treating human cases and distributing mosquito
nets and insecticide.
The DVS and MoH also disseminated public health information: initially these were contradictory
although later the two organizations’ messages were aligned. Local leaders, including religious
leaders, played an important role in communicating key public health messages about RVF and
debunking circulating rumours and conspiracy theories. For example, some local pastoralists
believed that the disease control operations were a guise to reduce the size of their herds. Their
suspicions were fanned by the timing of the interventions, which came shortly after the ousting of
the Islamic Courts in neighbouring Somalia.
Local radio stations were also utilized to broadcast public health messages, such as warning against
drinking raw milk, slaughtering animals or eating uncertified meat. Subsequent surveys identified
these as one of the most common sources of information for pastoralists. Kenyan mass media
provided considerable coverage of the RVF outbreak. This was often critical of the government’s
response, and was considered by some officials to be at times misleading and inaccurate,
contributing to a degree of panic amongst the general public: The demand for red meat in Nairobi
collapsed during the outbreak, although consumers were in fact at little or no risk. The outbreak
resulted in the imposition of a ban on the importation of livestock from the entire Horn of Africa by
trading partners in the Middle East. This ban had relatively little impact on Kenya, whose livestock
and meat exports to this region is relatively small, but had much greater impact in Ethiopia (see
Repercussions of the 2006/07 Horn of Africa Rift Valley fever outbreak in Ethiopia, below), and
17
even more in Somalia, where livelihoods and the national economy are highly dependent on the
export of live animals to the Middle East. Whilst OIE’s guidelines do not require such a ban, their
recommendations for importation of live animals or meat from countries with current or recent RVF
outbreaks are so stringent that this is not an unreasonable response.
Table 1: Sequence of events during the 2006/07 Kenya RVF outbreak
Details of the sequence of events during the Kenyan RVF outbreak are presented in the following
table:
Date Event
2003-2006 Ongoing drought in northern and eastern Kenya
August 2006 Earliest recollection of heavy rainfall by pastoralists in North Eastern Province
(NEP)
September 2006 Monthly GEIS report gave the impression that overall the risk in East Africa was
relatively low
Early October 2006 Somali pastoralists in Sangailo village, Ijara District, notice symptoms consistent
with RVF in their livestock
Earliest recollection by pastoralists of appearance of mosquito swarms
October 2006
GEIS’s October report suggested there was an elevated risk of an RVF outbreak
in East Africa, including northern Kenya, and that “ground surveillance and
response to disease outbreaks” would likely be required during the next 2-6
months
November 2006 EMPRES publish
RVF warning in special issue of EMPRES Watch
FAO and WHO sent warnings directly to the relevant national authorities
30
th
November 2006 Estimated onset of symptoms in human RVF Index Case
4
th
December 2006 Kenya DVS formally reports suspected RVF outbreak to OIE
14
th
December 2006 Herdsman from Garissa admitted to hospital coughing/vomiting blood
17
th
December 2006 Garissa Livestock Market closed
20th December 2006 11 deaths reported with unexplained febrile haemorrhagic illness in NEP
RVF diagnosis confirmed in 10 of 19 patients
20th December 2006
Outbreak investigation launched by MoH, Kenya Field Epidemiology and
Laboratory Training Program (FELTP), KEMRI, the Walter Reed Project of the
U.S. Army Medical Research Unit, CDC-Kenya's Global Disease Detection Center,
and other partners, including WHO and Médecins Sans Frontières (MSF)
Prior to 20
th
December 2006 Most cases were in young men who herded cattle
After 20th December
2006
Distribution of cases by age/sex broadened: young women disproportionally
affected, perhaps because they handle raw meat Children under five years and
the elderly rarely affected
21s
t
December 2006 MoH formally declares RVF outbreak
22n
d
December RVF confirmed in samples taken from sick livestock in Garissa District
24
th
December 2006 Number of human cases being reported peaked (date of onset of symptoms)
Late December 2006 Field laboratory set up at Garissa Provincial Hospital by CDC, KEMRI and MoH
Suspect and confirmed case definitions established
27t
h
Decembe
r
2006 Ban on slaughter in Garissa District
27th December 2006
Local leaders in Garissa appeal to government to declare floods and RVF
outbreak a national disaster
Meetings between Garissa District officials and local religious leaders to
publicise need to avoid home slaughter ahead of Eid-ul-azha
31s
t
December 2006 Eid-ul-azha – Muslim festival when 20,000 goats would usually be slaughtered in
18
Date Event
homes across Garissa District
8th January 2007
Livestock vaccination campaign started: initial target, 500,000 shoats and
600,000 cattle in Garissa, Ijara, Wajir, Isiolo and Tana River Districts
3 NGO vaccination teams operating in each district supported by FAO
Policy was to vaccinate apparently unaffected herds of livestock in districts in
which human or livestock RVF disease had been confirmed and also in adjacent
districts
Vaccination targeted 20 km wide belt surrounding affected areas
10
th
January 2007 Additional 450,000 doses provided with USAID funds
12th January 2007
Over 8000 animals vaccinated to date under the supervision of the MoLFD and
FAO
Vector control spraying, human and animal surveillance and social mobilization
activities continued
15
th
January 2007 Initial 100,000 doses of Kenyan manufactured vaccine used up
15
th
January 2007 MoH and MoLFD hold joint
press conference
15th January 2007
As of 15
th
January 2007, a total of 248 cases and 95 deaths had been reported
from the affected districts of Garissa, Ijara, Wajir, Tana River, Kilifi, Malindi
and Taita Taveta
Samples of suspected animal and human cases collected from Lamu district
One confirmed case and one suspect case reported in the coastal district of
Malindi
15th January 2007
MoH set up a team at national level to work on a national communication social
mobilization strategy. The Kenya Red Cross and UNICEF worked with the team.
National messages “would be finalised soon”. A WHO consultant for social
mobilization was made available in Garissa
16th January A team from Canada was deployed and set up a mobile lab in Malindi, Coastal
Province
19
th
January 2007 RVF outbreak spread to Isiolo, Wajir and Tana River districts
29
th
January 2007 GoK reports it has spent US$ 1.4 million on medical care for RVF patients
30
th
January 2007 USAID pledges additional 800,000 doses vaccine for delivery 1s
t
February
31s
t
January 2007 400,000 head livestock vaccinated to date
Late January 2007 RVF confirmed in: 3 districts in North Eastern Province (NEP), 5 districts Coast
Province; 2 districts in Central Province; one district Rift Valley Province; one
case in Nairobi but patient had visited NEP
May 2007 684 human cases reported, 155 deaths reported
• External support to the Government of Kenya
Although the RVF response was led by the Kenyan Government, donors, international
organizations and veterinary and medical NGOs provided a wide range of support.
Funding included:
• United Nations Central Emergency Relief Fund (UN/CERF) support to WHO and FAO totalling
US$ 1.9 million
• USAID/UnitedNationsOfficefortheCoordinationofHumanitarianAffairs (UN-OCHA):
1,250,000 doses of livestock vaccine, worth US$ 480,000, and US$ 35,000 to support CDC’s
assistance to MoLFD and MoH, plus additional support to the World Food Program (WFP)
• While practical support included:
19
• Under the direction of the MoH, CDC/KEMRI set up a field laboratory at Garissa Provincial
Hospital to facilitate rapid diagnosis and CDC also provided funds for government veterinary
personnel’s field allowances
• WHO provided a mobile laboratory, training in case management, and assistance with public
health education
• WFP provided food aid to people in flooded areas and helicopters to transport medical staff,
equipment and medicines to affected areas
• Various veterinary NGOs received funds from FAO and formed teams of technical staff with
their partner organizations and government technicians to carry out livestock vaccinations.
They also provided training for veterinary staff on RVF disease recognition, surveillance and
control, and funded other veterinary treatments
• Various medical NGOs provided transport to support the RVF sensitization campaign and to
transport suspect cases to hospital
• FAO provided training to veterinary staff to enhance ongoing surveillance and provided the
veterinary laboratory in Kabete with ELISA kits (IgG and IgM) and other laboratory equipment.
Additional technical assistance was also provided by an international team from the Global
Outbreak Alert and Response Network (GOARN). An 11-member team from GOARN partner
institutions and WHO (Country Office, Regional Office for Africa and Headquarters) supported the
MoH, provincial and district health authorities in implementing public health measures to control
the outbreak. GOARN partners included: a mobile laboratory team and two epidemiologists from
the Public Health Agency of Canada (National Microbiology Laboratory, Canadian Science Centre for
Human and Animal Health, and Canadian Field Epidemiology Programme), a technical expert on Rift
Valley Fever and a case management/infection control expert from the National Institute of
Communicable Diseases (NICD) and Department of Clinical Microbiology and Infectious Diseases,
National Health Laboratory Service (NHLS) and the School of Pathology of the University of the
Witwatersrand, Johannesburg, South Africa, and a three member social mobilization team from the
US CDC and WHO Centre for Vulnerability Reduction, Tunisia. CDC-Kenya coordinated bilateral
assistance from the US Departments of Agriculture, Defence and Health and Human Services.
• In the end
By March 2007, when the RVF outbreak in Kenya had abated, official reports showed that 155
people had died and economic losses were estimated to amount to US$ 30 million. The impacts of
the disease were felt not just by pastoralist producers but along the entire meat value chain. Some
livestock traders and butchers were unable to resume business even after the bans on movement
and slaughter of livestock in the affected areas were eventually lifted due to the depletion of their
working capital.
20
2.2.2 The Ethiopian experience
• Introduction
Over the past 75 years or so, RVF has been reported and in most cases confirmed from a swathe of
countries down the eastern side of Africa and also from across the Red Sea in the west of the
Arabian Peninsula. This pattern broadly follows the Rift Valley - from Egypt in the north to South
Africa in the south. However, cases of the disease in livestock have never been officially reported
in Ethiopia or Eritrea, nor has the presence of the RVF virus been unequivocally demonstrated in
these countries, although some reports mention possible human cases in southern Ethiopia during
the RVF outbreak in the Horn of Africa in 1997/98, and there is some evidence of cryptic virus
activity in the country. However, all the surrounding and a number of nearby countries, including
Egypt, Sudan, Kenya, Somalia, Saudi Arabia and the Yemen, have reported cases in people and
livestock during the past 30 years. Within the Horn of Africa in general there is considered to be a
lack of transparency in reporting and publishing RVF surveillance results because of fear of negative
impact upon the livestock trade, which has been estimated to be worth around US$ 600 million
annually throughout the region.
The ecology of the region of Saudi Arabia and Yemen (the Tihama ecotype of the eastern Rift Valley
zone) affected by RVF is reported to be identical to the west floor of the Rift Valley in Ethiopia and
Eritrea. Aedes mosquitoes - the primary vector of RVF and the species responsible for maintaining
the virus between outbreaks - are very widely distributed, including in Ethiopia, and low-lying,
semi-arid areas of the country are flood-prone. FAO considers that RVF virus “is probably present in
all countries of sub-Saharan Africa”.
The conditions necessary for the onset of RVF virus activity in an area are the occurrence of heavy
and persistent rainfall over several months, together with the presence of the virus and susceptible
livestock species. Heavy rainfall can lead to a rise in the water table which ultimately causes
flooding. This is seen in East Africa in geomorphic formations called dambos - depressions found in
grasslands prone to flooding. Rainfall of between two and ten-times mean annual values have been
associated with periods of epizootic RVF. Flooding in semi-arid and arid zones can also occur in
floodplains downstream from where the rainfall occurs, which can be on plateaus or mountain
forest zones; examples of such at-risk areas include the watersheds of the Wabi Shabelle and
Genale rivers in the Ethiopian plateau, which are prime pastoralist areas.
Although RVF has never been formally confirmed in livestock in Ethiopia, outbreaks of the disease
in other countries in the region none-the-less have dramatic impacts on Ethiopia’s export trade in
live animals (both formal and informal ‘cross-border’ trade) and the growing trade in chilled and
frozen beef and mutton carcasses and cuts. In the event of an RVF outbreak in East Africa and/or
the Horn, bans on imports from the entire region tend to be imposed by the major importing
countries, especially the Gulf States and Egypt, and these bans are often retained for lengthy
periods.
21
• Ethiopia and livestock trade
Ethiopia’s international livestock trade takes two main forms: export of live animals and export of
chilled and frozen beef and mutton carcasses and cuts. The former is dominated by informal cross-
border trade in which livestock passes from ethnically Somali pastoralist regions of Ethiopia,
especially Zone IV, into Somalia in exchange for consumer goods, from where live animals are
mostly re-exported via the Djibouti quarantine and export facility to the Gulf States. It is estimated
that this informal cross-border ‘contraband trade’ is up to 100-times larger than the formal,
government-sanctioned trade in live animals.
The export meat trade is centred on some six export abattoirs and four major private meat
exporters. Cattle are purchased from pastoralists, finished in feedlots for three to four months,
slaughtered and processed in the private abattoirs, and then exported as chilled or frozen carcasses
and cuts. Sheep and goats are also slaughtered, processed and exported. Beef is mainly exported to
Egypt, while mutton is mainly exported to the Gulf States, principally Saudi Arabia and the UAE.
Official figures show that in 2006/07, Ethiopia earned more than US$ 56 million from the export of
some 295,000 head of livestock and 6000 tonnes of meat products: Saudi Arabia and UAE accounted
for 48% and 46% of the meat exports, respectively.
The on-going Ethiopia Sanitary and Phytosanitary Standards and Livestock and Meat
Marketing (SPS-LMM) programme, financed by USAID and implemented by Texas A&M University in
cooperation with MOARD, has as its ambitious goal to increase exports of meat and meat products
five-fold to 30,000 tonnes annually.
The Ethiopian livestock and meat export trade is, however, extremely vulnerable to bans imposed
by the importing countries: recently these have been triggered by disease outbreaks (actual or
potential) including foot-and-mouth disease as well as RVF, and deficiencies observed at the export
abattoirs during routine inspections by the importing countries. The demands of importing countries
are not always clear, transparent or in accordance with the laid-down OIE requirements: for
example, during the recent RVF outbreak in the Horn of Africa, the UAE imposed the ban on
importation of meat from Ethiopia in mid-January - after the end of the Hajj, the annual pilgrimage
to Mecca - although the highest risk period was probably during December 2006/early January 2007:
neighbouring Kenya formally reported RVF in the north-east of the country in early December 2006.
• Response in Ethiopia to RVF in East Africa
Following the outbreak of RVF in Kenya in late 2006, the response in Ethiopia had three major
components: an awareness raising and sensitization campaign targeted at the Kenya border area;
some preventive measures including spraying livestock with ‘acaricide’ (most likely synthetic
pyrethroids which are effective against biting insects as well as ticks); and official delegations
dispatched to the major livestock and meat importing countries, mainly Egypt, Saudi Arabia and
UAE, to lobby for and negotiate the lifting of the livestock and meat import bans imposed by these
countries.
The main actors in Ethiopia’s response to the RVF outbreak in neighbouring countries were the
Government of Ethiopia supported by a number of international and national agencies, donor-
funded programmes and NGOs, including USAID, FAO, WHO and Save the Children.
22
The sensitization campaign was organized by the government, supported by FAO, USAID, and NGOs,
including the SPS-LLM programme, and aimed to ensure that the at-risk population were aware of
the symptoms of RVF so that they could report any cases encountered. The campaign included the
production and distribution of pamphlets, which described RVF symptoms in local languages, by
USAID and NGOs.
The NGO Save the Children (US) undertook spraying of livestock with acaricide in the border area as
a precautionary measure. At the regional level, the organizations that responded to the outbreak
were: WHO, FAO, UNICEF, CDC, NGOs and the region’s governments. Coordination was done by
WHO and FAO. These agencies together with their partners formed a regional task force, the Inter-
agency Standing Health Cluster, which helped in the dissemination of information to countries in
the region, including Ethiopia.
Table 2: Sequence of events
Date Event
Early October 2006 Somali pastoralists in Sangailo village, Ijara District, Kenya notice
symptoms consistent with RVF in their livestock
October 2006 GEIS’s October report suggested there was an elevated risk of an RVF
outbreak in East Africa, including northern Kenya, and that “ground
surveillance and response to disease outbreaks” would likely be required
during the next 2-6 months
November 2006 EMPRES publish RVF warning in special issue of EMPRES Watch: it was
noted that there had been “increased and widespread rainfall over the last
three months over most of Eastern and coastal Kenya, Somalia and
Southern Ethiopia” but Ethiopia was not included in the list of countries
which had an “elevated risk of RVF”
FAO and WHO sent warnings directly to the relevant national authorities
30t
h
November 2006 Estimated onset of symptoms in human RVF Index Case in Kenya
4t
h
December 2006 Kenya DVS formally reports suspected RVF outbreak to OIE
20t
h
December 2006 Active surveillance for RVF commenced by Ethiopian MoARD and FMOH in
areas bordering Kenya
Assessment teams sent to Somali Region and South Omo Zone, SNNPR. The
teams, comprised of personnel from MoARD, FMOH, WHO and FAO, report
back to the National Coordination Committee, chaired by MoARD
January 2007 Meeting held between WHO, FMOH, MoARD and FAO
WHO provided technical support through consultants sent to SNNPR,
Tigray, Afar, Somali, Oromia
4 January 2007 A Nairobi-based FAO team, drawn from animal health experts in a number
of countries of the Horn of Africa, work with veterinary departments in
Kenya, Somalia and Ethiopia to address the RVF outbreak in the region
In Ethiopia, officials from WHO and various international aid agencies
present in the area assist an FAO team to draw up preparedness,
communication, surveillance and response activities
29 December 2006 – 3
January 2007 Hajj – annual
pilgrimage to Mecca
17t
h
January Preventive measures to control RVF taken in Ethiopia, but no cases have
23
Date Event
yet been reported
Livelihood impacts of RVF in the form of drastically reduced market access
for pastoralists in Ethiopia
UAE announces a ban on all animal products from the Horn of Africa
January 2007 Import bans imposed by Egypt and Saudi Arabia
20t
h
February 2007 “Despite the ban imposed by UAE, which has affected our livestock
market, no threat of Rift Valley Fever has been detected anywhere within
the country" Amsalu Demisse, acting head, Ethiopian Veterinary
Department
22n
d
February 2007 Ethiopian delegation led by Abera Deresa, state Minister of Agriculture and
Rural Development, travel to UAE to push for lifting of import ban by UAE:
“there was no sign that the fever had or would hit Ethiopia”
UAE government promised to lift ban
Saudi Arabia later sent their own delegation to Ethiopia to assess the
situation
March 2007 Meat imports from Ethiopia still banned by UAE
Live animal imports to UAE from Ethiopia allowed once they had
completed quarantine/testing in Djibouti
Meat prices in Ethiopia drop: producer prices for beef down from Birr 8 to
6.5; mutton down from Birr 9 to 6.5-7
March 2007 Intensification of RVF surveillance
Funds reported to have been received from UN-OCHA to assist FMOH in
preventing and controlling a possible RVF outbreak
WHO reported to be recruiting and deploying two national consultants to
provide technical support in high risk regions
Information on current situation in Ethiopia is monitored in collaboration
with FMOH, MoARD, FAO and partners
WHO ensure continued information exchange amongst partners
March 2007 RVF outbreak in neighbouring Kenya and Somalia abates: no cases of RVF in
either livestock or people detected in Ethiopia
May 2007 Preparation underway by WHO to conduct training for health workers on
RVF surveillance in identified high risk zones
Areas targeted for training include Afder and Liben zones in Somali Region,
South Omo zone in SNNPR and Borena zone in Oromiya Region
RVF surveillance guidelines for health facility and community workers and
Standard Operating Procedure for specimen collection, storage and
transportation have now been developed by WHO
2.2.3 The outbreak unraveled
• Kenya
RVF outbreaks in the Horn of Africa are intrinsically linked to unusually heavy rainfall and flooding
events, which can be forecast with some degree of accuracy. It would thus appear that the disease
24
would be well suited to early warning linked to appropriate, timely and pre-planned responses.
However, this was not the case in 2006/07: why not?
The RVF early warning is generated through a joint initiative of the NASA Goddard Space Flight
Center and the US Department of Defence Global Emerging Infections Surveillance and Response
System (DoD GEIS) that utilises various remotely-sensed data to which these organizations have
access. It appears that this is an example of a technology seeking an application rather than
necessarily the most appropriate solution to a problem. The monthly RVF risk bulletins present a
variety of remotely-sensed measurements, including abnormal sea surface temperatures, cloudiness
and outgoing long-wave radiation anomalies, all of which have been linked to heavier than normal
rainfall in East Africa. The actual monthly RVF risk map - perhaps the most important part of the
bulletin - is based on persistently positive anomalies in an index of vegetation growth (the
normalized density vegetation index: NDVI). NDVI time-series show a positive response to rainfall,
especially in semi-arid areas, which means that NDVI data can be used as a measure of the
magnitude and persistence of widespread and prolonged rainfall events. On the RVF risk maps,
pixels that show a positive NDVI anomaly above a specified threshold value for three successive
months and which lie in RVF endemic areas are shown as ‘RVF risk areas’. The RVF risk maps,
therefore, are not forecasts of where heavy and persistent rainfall might occur, and cause flooding
and RVF outbreaks; rather they indicate where heavy and persistent rainfall has occurred, where
flooding might occur soon (or possibly already has), and where RVF outbreaks could occur (or might
already be present).
The US team behind the RVF early warning system are both physically and culturally distant from
the actual situation on the ground: the scientists clearly inhabit a very different world to that of
the pastoralists of the Horn. The RVF early warning mechanisms are not embedded in local,
national or regional knowledge networks, nor are they provided directly to the pastoralists. Rather,
the RVF risk assessments are simply generated on a monthly basis and posted on the DoD GEIS
website for anyone to use as they wish. They come with a strong ‘health warning’; it is explained
that the information is presented as a ‘statement of research for the purpose of scientific
validation and review’. Specialists at the FAO and WHO pick up on these monthly bulletins, analyse
and reinterpret them and, when appropriately repackaged, distribute them via an internet-based
newsletter and also by direct contact with the responsible authorities in the at-risk countries. In
2006/07, by the time the RVF warnings had been generated and distributed, more than a month
had passed since pastoralists had reported the first suspected cases in their livestock.
The RVF early warning system does not, therefore, seem to respond to clear demand from the
affected countries to provide a sufficiently early ‘early’ warning, or to be linked to any response
procedure. To date, there have not been any attempts to combine the local knowledge and
observations of those at the front-line – pastoralists, CAHWs, local-level government officials, local
NGO staff, community-based organization members and others – with the remotely sensed data. In
the 2006/07 outbreak, pastoralists including CAHWs reported suspected cases well before the
‘early’ warning was issued. A mechanism or process that would facilitate integrating field-level
observations with remotely sensed data would seem to offer a significant advance over the
currently used system. In addition, better use of the available heavy rainfall foresight technology
could enable increasing levels of risk to be identified and acted upon earlier – assuming that the
appropriate response capacity would be available.
25
A participatory assessment undertaken shortly after the RVF outbreak demonstrated that the
Somali pastoralists and the CAHWs in Kenya, who themselves are pastoralists and livestock keepers
who have received some rudimentary animal health training, were highly adept at recognizing the
symptoms of RVF, correctly associating the disease with risk factors such as heavy rainfall, flooding
and swarms of a rarely seen mosquito, and they recalled that this was similar to the previous RVF
outbreak that had occurred some 10 years earlier. Some pastoralists reported the situation to local
government officers, but prior experience had taught them to have low expectations; with
previously reported livestock disease outbreaks they had often witnessed no response from the
veterinary department.
The first response to the disease in Kenya was therefore at the local level; pastoralists resorted to
various available, albeit mostly inappropriate, traditional and modern treatments, including use of
smoke to repel the clouds of mosquitoes, reading from the Koran, and administration of
trypanocidal drugs and antibiotics. Pastoralists and CAHWs are able to treat many disease
conditions that occur locally using a combination of their indigenous knowledge and treatments,
and modern medicines. In the case of RVF, however, there are no effective curative remedies that
can be deployed and they do not currently have access to RVF vaccines.
Some of the district veterinary offices which received the first reports from the field thought that
the abortion storms, which are usually the first symptom of RVF in livestock, could have been
caused by another disease, trypanosomiasis. This suggests that the RVF warnings received from the
FAO by the chief veterinary officer had not been forwarded to the districts at most risk or, if they
had, that insufficient attention had been drawn to them. However, even if the warnings and field
reports had been linked up, there would have been little the front-line veterinary staff could have
done. There was no contingency plan for RVF in Kenya. Funds for routine operations are limited and
there was no provision at any level of the veterinary services (national, provincial or district) for
access to emergency funding to allow an effective and timely investigation and response to be
launched. Access to national and international emergency funding could have been triggered if the
RVF outbreak had been declared a national disaster – but this was not done on this occasion.
Perhaps the reason for this was that, whilst the situation was undoubtedly serious, this act of last
resort is reserved for other more widespread and, arguably, even more serious disasters, such as
severe and widespread drought – although politics and vested interest are equally likely to have
been a factor. The actual physical ability to respond was further constrained by the challenging
practical situation in the affected area: wide-spread flooding and impassable dirt roads made it
virtually impossible to move around by vehicles, and in any case very few running vehicles were
available to veterinary department personnel. Only when cases of people with severe haemorrhagic
symptoms started to report at local health facilities did an official response finally swing into
action. Helicopters, the only effective means of moving around the affected areas, were then made
available to the Ministry of Health, and veterinary department personnel were able to hitch a ride
and gain access to the affected area.
Although livestock play an important role in the livelihoods of millions of people in Kenya, livestock
have long been under-valued and under-resourced by the Government, and the veterinary
department enjoys little influence in the corridors of power. Pastoralist areas have never been well
served by state veterinary services: in the colonial era, services were largely focused on the needs
of settler farmers; post-independence, although veterinary services were refocused to better meet
the needs of local small-scale farmers, these services generally had little impact in remote
pastoralist areas. Following the advent of World Bank-driven structural adjustment programs
26
(SAPs), state veterinary services were scaled back and the roles of the public and private sector
better defined. Clinical services, for example, were now regarded as a private good. Although it
was hoped that the private sector would fill in the gaps left by state veterinary service, there have
been few incentives for private practitioners to establish private practices other than in and around
major urban centres. The prospect of establishing private practices in the harsh, remote
environments where pastoralists live was particularly unattractive. NGOs established community-
based animal health services in response to the needs of pastoralists. Initially the veterinary
profession and the state veterinary service were largely hostile to CAHWs. Over time, however, this
hostility has dissipated and today the role of CAHWs is formally recognized in [thus far unendorsed]
policy documents both within Kenya and internationally: the OIE, for example, now accepts that
CAHWs, in certain circumstances, can be part of a country’s formal disease surveillance system.
A key feature of pastoralism is mobility; rainfall tends to be very localized and erratic and
pastoralists move their herds and flocks to take advantage of pockets of rainfall and the resultant
growth of pasture and browse. Access to traditional wet and dry season grazing areas has, however,
been increasingly restricted over recent decades, including through the creation of national parks
and reserves, settlement of farmers and agro-pastoralists in valleys and along rivers, as well as
through the generally increased pressure from human and livestock population growth. All of this is
exacerbated by a long-term trend of decreased rainfall and an increased frequency of extreme
weather events, including both drought and floods. Insecurity and regional conflict, both
exacerbated by the ready availability of weapons, and the presence in the region of a failed state
(Somalia) all serve to further restrict mobility. In the more distant past, pastoralists would have
had far greater freedom to move their animals and families to avoid floods - and thus RVF-,
localized droughts or other disease outbreaks, albeit within a framework imposed by traditional
rangeland management systems. But with the current restrictions on mobility, and strong
encouragement by the government to adopt a more sedentary lifestyle, this is now less of an
option.
The Somali pastoralist of north-eastern Kenya are isolated from decision makers in Nairobi not just
by long distances and bad roads but also culturally; they speak a different language and practice a
different religion to the majority of Kenyans and are more closely aligned to neighbouring Somalia
than to the rest of Kenya. North-eastern Kenya is one of the poorest parts of the country: its people
are marginalized and are poorly served by the state in terms of education, health or any other
public services. Even with the explosive growth of mobile phone networks in Kenya (from very few
to more than 12.5 million subscribers in less than 5 years), large parts of north-eastern Kenya
remain outside the network. They also have very few alternative livelihood options beyond
livestock rearing: even there the long-term prospects look increasingly bleak given the trend of
movement restrictions, decreasing rainfall and the increasing frequency of extreme weather
events.
Once news of the RVF outbreak started to emerge, there was considerable coverage by the media.
Not all the information disseminated in the media was accurate: the technical experts in the line
ministries considered that this misinformation was a contributory factor in the panic that gripped
Nairobi. This contributed to a collapse in the market for red meat in the capital, although in fact
the risk to consumers was virtually nil. The panic in Nairobi was also fuelled by reports that one
patient with RVF had been admitted to the main government hospital in the city. Whilst this was
true, the patient had been infected in north-eastern Kenya, from where he travelled to the capital
and where he was taken ill. Adding to the confusion in the early stages of the outbreak, the state
27
veterinary and medical services initially put out contradictory information regarding the safety of
consuming livestock products in the at-risk area: whilst the medical authorities suggested that meat
and milk should not be consumed, the veterinary authorities recommended that inspected and
properly cooked meat and boiled milk were safe. Later, the two services aligned their messages;
however, lack of pre-prepared, tested and unequivocal messages targeted at the different at-risk
groups was a clear constraint. Traditionally, the veterinary and medical services have operated
separately from each other, reporting to entirely different line ministries. Rather disappointingly,
despite the vast amount of expenditure on preparedness for highly pathogenic avian influenza,
including a number of initiatives in Kenya, there appears to have been little beneficial spill-over of
these approaches to the management of another zoonotic disease, RVF.
The media also found it difficult in the early stages to gain access to well-informed representatives
of the veterinary and medical services to provide the information they needed and to answer their
questions. The fact that the outbreak occurred during the Christmas and New Year holiday period,
the main annual break for the majority of Kenyans, added to the difficulties.
As awareness of the RVF outbreak spread and criticism of the speed and adequacy of the
government response mounted, the situation became increasingly politicized and there was
pressure to be seen to be doing something – anything! RVF outbreaks in the Horn are essentially
self-limiting – they persist only as long as floodwater provides the conditions needed for mosquitoes
to breed: by the time cases have been detected the epidemic curve is likely to be reaching or has
already reached its peak, so whatever action is taken -including doing nothing - the outbreak will
inevitably come to an end.
Following the 2006/07 outbreak a number of interventions were made, including closure of
livestock markets, bans of movement of livestock from the affected areas, bans on the slaughter of
livestock, warnings about the danger of consuming raw milk or uncertified meat, vaccination of
livestock, treatment of livestock with pour-on or spray formulations of insecticide and distribution
of mosquito nets. Whilst these interventions demonstrated to the affected population and the
wider ‘worried well’ that public action was being taken, and is likely to have decreased panic, and
some may have had additional benefits associated with them – such as application of insecticide
controlling ticks on livestock and distribution of insecticide-treated mosquito nets helping to
prevent malaria – it is questionable how appropriate they were in the control of RVF. In the first
described case of RVF, which occurred in a flock of sheep near Naivasha in 1930, the outbreak was
effectively brought to an end by moving the sheep up the escarpment to an altitude where the
vector (Aedes mosquitoes) were absent. This suggests that the disease probably cannot be
introduced into an area through the movement of infected livestock; that the right environmental
conditions need to be present simultaneously: the virus, vectors and a population of susceptible
livestock. The outbreak in Tanzania in 2007, for example, is unlikely to have ‘spread’ from Kenya as
has sometimes been inaccurately suggested; rather the necessary conditions (heavy rainfall and
persistent flooding) occurred in Tanzania a little later than in Kenya and, since the RVF virus and
susceptible livestock were present, an outbreak ensued. If this is indeed correct, the necessity for
movement control is perhaps questionable.
This illustrates, however, that there is a general absence of sound evidence to guide which
interventions are appropriate in the event of an RVF outbreak. During the outbreak, butchers in
Nairobi started to display meat inspection certificates in their shop windows. This seemed to
reassure customers that the meat being offered for sale was safe. Although demand for red meat in
28
general declined during the outbreak, sales from one supplier actually increased as the newly
revived parastatal Kenya Meat Commission apparently came to be seen as a trusted source of meat.
RVF is a difficult disease to do research on, which perhaps explains why there are a surprisingly
large number of gaps in the understanding of relatively basic aspects of the disease’s epidemiology.
For example, much of the disease literature suggests that human cases can occur as a result of
transmission of the virus via mosquitoes, whereas records of actual cases show that these occur
mainly in those people directly exposed to infected livestock or tissues derived from them including
aborted foetuses, i.e. herders, slaughterhouse workers, butchers, animal health workers and, to a
lesser extent, women who handle raw meat: the elderly and children, for example, rarely become
infected, although presumably they too are exposed to mosquitoes. Two factors that make doing
research on RVF difficult are the challenging physical conditions that accompany disease outbreaks,
such as widespread flooding and impassable roads, and the episodic nature of the disease: it is
difficult to establish an effective field-based research programme for a disease that may not occur
for a decade or more. The disease is also difficult and costly to research in a laboratory context
because of the risk the virus presents: health and safety regulations and good laboratory practice
would demand the highest level of containment.
The episodic nature of the disease also presents specific challenges in the application of the lessons
learned from previous outbreaks. With an interval between outbreaks of up to 20 years, personnel
who were actively involved in previous outbreaks are unlikely to still be in their posts for the next.
As a result there is little or no institutional memory on the disease. Coupled with the lack of
written contingency plans or specific mechanisms to capture lessons learned, each new outbreak
exposes the responsible authorities and front-line staff to a very steep learning curve.
The currently available vaccines for RVF, which have been on the market for 50 years or more, are
far from ideal but in the past there have been few incentives to develop new ones. Low uptake of
vaccination by pastoralists was reported during the 2006/2007 outbreak. This was partly because
there was confusion about what classes of livestock should be presented for vaccination, but also
because the pastoralists knew, or rapidly found out through experience, that vaccinated animals
tended to abort. Some pastoralists complained that they had not been warned that this was a likely
side-effect of vaccination. Vaccination against RVF in the Greater Horn of Africa presents a number
of challenges. Indeed some experts consider that these are so great that they effectively preclude
the use of vaccines to prevent/control RVF outbreaks in this region – although they are effectively
used in other regions where the epidemiology of the disease, environmental conditions and
infrastructure are different, e.g. southern Africa.
The current Smithburn vaccine has a shelf-life of around 4 years, while the average interval
between outbreaks in the Greater Horn has been around 10 years, it has been closer to 20 years
during some inter-epizootic periods. Hard pressed veterinary authorities with many demands on
their scarce resources are understandably reluctant to maintain vaccine stocks for a disease which
occurs intermittently and which are likely to expire before they are used. For sound commercial
reasons the manufacturers also avoid maintaining large stocks which are likely to reach their expiry
dates before they can be sold. However, the lead-time needed by manufacturers to produce new
batches of vaccine can be several months. Waiting until an RVF outbreak is highly likely or actually
occurring will leave too little time for the manufacturers to respond. Even if the manufacturers did
have adequate vaccine stocks, waiting until the heavy rains and flooding have begun means that it
is then very difficult, often impossible, to transport and distribute vaccine in remote areas which
29
have no all-weather roads. In a recent assessment, carried out by the International Livestock
Research Institute (ILRI), it was estimated that for a successful vaccination programme to be
carried out in an at-risk area, an early warning would need to be issued and acted upon more than
140 days before the first case occurred. This timing was based on the assumption that stocks of
vaccine would need to be ordered, manufactured, shipped, distributed within the at-risk area, and
the vaccination campaign carried out on a sufficiently large-scale to prevent the outbreak.
Several recent global developments have increased interest in RVF. RVF is regarded as a possible
weapon for bioterrorists and as such the disease has attracted the attention of the US military. As a
result, new vaccines have been developed for use in people - primarily to protect US soldiers - and
also for use in livestock, but these are not yet widely available. GALVmed, a new public-private
partnership funded by DFID and the Bill & Melinda Gates Foundation, established to develop and
make available livestock vaccines and other animal health products that address the problems
facing poor livestock keepers, has prioritized RVF as one of its first targets. It plans to introduce
two new RVF livestock vaccines: a multivalent one that will provide simultaneous protection against
lumpy skin, sheep and goat pox and RVF, and one that can be used in the face of an RVF outbreak.
Because of the often lengthy intervals between RVF outbreaks, and the available vaccine’s
undesirable side-effects, it has not been considered desirable or feasible to routinely vaccinate
livestock in at-risk areas in the Horn. However, the idea behind the new multivalent vaccine is that
protection against the other diseases, all of which are significant on-going threats to livestock in
the areas at risk from RVF in the Horn, will provide the incentive to use the vaccine routinely. In
doing so, RVF protection will be provided almost as a bonus. This approach offers the possibility of
avoiding future outbreaks, and also of preventing a number of other economically important
livestock diseases. Some challenges remain, however, such as who pays – pastoralists or
government, i.e. is the herd immunity this approach could provide a public or private good? Who
will carry out the regular vaccination campaigns? If vaccine usage is to be mainstreamed in these
areas there also needs to be a major change in the way pastoralists view disease control: they are
often more likely to wait until a disease occurs and then attempt to treat it than adopt a
preventive approach – particularly if they are unconvinced of the benefits vaccination is likely to
bring.
The potential spread of diseases, especially vector-borne diseases, beyond their traditional
geographical foci under various climate change scenarios, also means that multinational veterinary
companies are now taking interest in diseases such as RVF which have the potential to spread to
new areas. In this respect, the recent experience with bluetongue in Europe is relevant: whilst the
disease was largely confined to the tropics, the disease was neglected. However, as soon as it
became established and started to spread in Europe, governments and veterinary pharmaceutical
companies took interest. Within a few years a new and effective vaccine had been developed and
was being widely used in the newly affected areas.
Shortly before the 2006/07 outbreak Ethiopia, supported by the US, had ousted the Islamic Courts
from Mogadishu. The Muslim pastoralists in north-eastern Kenya were apprehensive when the Kenya
veterinary authorities wanted to vaccinate their animals, suspecting this may be a ploy to reduce
their herds. This is illustrative of the general lack of trust between the pastoralists and government
officials. In this case the government engaged with the local imams who played a key role in
convincing the pastoralists that there was no hidden agenda behind the vaccination campaign.
30
As the narrative has highlighted, a major problem is that the various actors with a stake or interest
in RVF control remain largely unconnected. Divisional and district veterinary offices in the
northeast have poor linkages with both the widely dispersed pastoralist population and the
veterinary department in Nairobi. At headquarters level, the Director of Veterinary Services and
Director of Medical Services do not have a tradition of close cooperation and there are few
mechanisms to promote closer collaboration and integration. There is a disconnect between the
instigators of the early warning system and the veterinary and medical authorities, who are their
prime target audience. NGOs which specialize in veterinary interventions often coordinate their
activities poorly, both with other NGOs and with the state veterinary service. They do have the
advantage, however, of being able to respond relatively quickly. Donors, on the other hand, are
often poorly equipped to react to fast-onset disasters. And whilst the coordination between the
main actors improved as the outbreak progressed, the most critical period is during the early
warning stage and at the start of the outbreak.
The 2006/07 RVF outbreak affected not just Kenya but also Somalia and Tanzania. In addition,
although the disease was not detected in Ethiopia, the country suffered nonetheless when the Gulf
States banned the importation of livestock from the entire Horn of Africa (see below). As the
disease is likely to occur in more than one country in the region at one time, and to have
implications for the entire region’s trade, a regional approach would appear to be beneficial.
During this outbreak Kenya bought up the entire stock of RVF vaccine available in South Africa;
when the disease occurred some weeks later in Tanzania there was no further vaccine available.
The Gulf States are major international players in the livestock sector of the Horn of Africa through
their imports of large numbers of live animals and, increasingly, also of meat. Although Kenya is not
involved in the export of live animals, it does export some meat. It has long been thought that the
Gulf States use RVF as a blunt political or economic tool: bans imposed in response to actual or
suspected outbreaks of RVF are often kept in place for much longer than can be justified on disease
risk grounds. In the process alternative suppliers emerge to take advantage of the market
opportunities the bans provide.
• Ethiopia
Not surprisingly given the different experiences and potential risks of RVF in the two countries, the
response to the threat of RVF in Ethiopia differed from that in Kenya. The immediate response by
the Ethiopian authorities to the outbreak of RVF in neighbouring Kenya was to initiate surveillance
and awareness raising activities in the border zone and at-risk pastoralist areas. This was done by
the MoARD and FMOH supported by WHO, FAO and various international aid agencies, who also
drew up plans for response activities; in the event these were not required, although SCF (US)
undertook precautionary measures in the form of administration of insecticide to livestock.
Surveillance in the Ethiopia-Somalia border area was considered to have been constrained by the
then on-going conflict between Ethiopia and Somalia.
Although sector coordination appeared to be reasonably good, the dominant players were
government and international agencies. The private sector is relatively under-developed in Ethiopia
and actors from this sector played little part in the countries response to RVF. Also, those most
directly affected by the RVF ban – the pastoralist livestock keepers and traders - are marginalized
in Ethiopia: they have little voice or influence in government circles. As in Kenya, absence of a
carefully considered contingency plan hampered a timely and effective response. For example,
31
WHO only developed surveillance guidelines and standard operating procedures for specimen
collection, storage and transport in May 2007 – several months after the peak risk.
Livestock export bans from the Horn of Africa are highly controversial and are regarded by many as
being highly politicized measures. Although RVF has never been diagnosed in livestock in Ethiopia,
and was not detected in the country during the 2006/07 outbreak that occurred in Kenya, Somalia
and Tanzania, none-the-less Ethiopia was included in the blanket bans on imports of live animals
and meat by the Gulf States. It is significant, however, that the bans were imposed only after the
annual hajj, when up to 15 million small ruminants are required for ritual slaughter – with the
majority of these being imported from East Africa and the Horn of Africa. The 2006/07 hajj took
place in late December and early January, thereby coinciding with the probable peak of the
epidemic curve in Kenya and the highest risk of an outbreak in Ethiopia. But imports bans were only
imposed by the Gulf State authorities in mid January, well after the hajj. This clearly illustrates the
imbalance of power between the exporting countries in East Africa and the Horn and the importing
countries in the Gulf States, with the latter calling the shots and responding to the potential or
actual threat from RVF in a manner that primarily met their own needs. In particular it
demonstrates that the Ethiopia authorities were unable to demonstrate to the Gulf States that they
were RVF-free. Although a high-level Ethiopian delegation travelled to UAE to argue for the ban to
be lifted, and apparently received a promise from the UAE authorities that this would occur, a
month or so later the ban was still in place. An important element in innovation response capacity,
that of trust, was apparently largely absent between the Ethiopian and Gulf State authorities.
There is little evidence to suggest that any significant learning occurred during the 2006/07 RVF
threat: in all probability the response in Ethiopia to the next RVF outbreak in the Horn will not have
benefited from experience gained during the previous one.
2.3 Case Study 2: Drought in South-Eastern Ethiopia, 2005/06
2.3.1 The Southern Somali Ecosystem
“Drought is no longer a slow onset disaster in the region, but a chronic emergency.”
Source: FAO (2006) FAO Horn of Africa Regional Drought Response, FAO, Rome
The southern Somali ecosystem, straddling north-western Kenya, south-eastern Ethiopia and
southern Somalia, is an arid and semi-arid region predominantly inhabited by pastoralists and agro-
pastoralists, with some settled farmers and a few urban centres. Rainfall, in the range of 400-600
mm per annum, is usually distributed between two rainy seasons: one from the end of March to July
(known as the Gu season in Ethiopia and Somalia) and the second from October to early December
(known as the Deyr season).
Localized droughts have occurred every 3 to 5 years, whilst major droughts have been reported to
occur anywhere between four or five times a century to every 10 years or so.
A recent analysis by NASA of rainfall records from the 1950s onwards has shown that rainfall in the
East African region during this period has steadily declined and that this is linked to rising Indian
Ocean temperatures: rain that previously fell on land is now falling in the ocean. This has resulted
in particularly significant declines in rainfall in the already dry Somali ecosystem. The current
32
predictions are that this decline will continue, with a 15% reduction in rainfall occurring every 20-
25 years (NASA 2008). Major drought events have been recorded in Ethiopia in 1973-74, 1984-85,
1999-2000, 2001-2002 and 2005-2006 and again at the present time.
The dynamics of cattle populations in these regions are typically one of ‘boom and bust’, but the
periodic crashes do not appear to be simply related to the occurrence of drought. In a study
undertaken (Desta 2001) in the Borana Plateau of southern Ethiopia over a 17 year period from 1980
to 1997, marked decreases in cattle numbers occurred in 1983/85 and 1991/92 (and also just after
the study period in 1998/99) when herd sizes dropped by between 37% and 62%. Whilst the
decreases in 1983/85 coincided with a severe drought, from 1986 to 1997 annual rainfall varied
relatively little, and yet the cattle population crashed again in 1991/92 - a year when rainfall was
close to the long-term average of 706 mm. The study concluded that the crashes in cattle
populations were due to interactions between the stocking density and rainfall: if stocking densities
exceeded 20 head of cattle per square kilometre, even modest reductions in rainfall resulted in
severe decreases in cattle numbers.
The main cause of these losses was death due to starvation resulting from a ‘forage crisis’: the
higher the stocking rate and the greater the rainfall deficit, the higher the rate of loss. The study
also suggested that system stability was declining: between 1980 and 1997, the average household
cattle holding decreased by 37% from 92 to 58 head. Population growth during this period,
however, means that the number of households will have increased.
In addition to the harsh environment and decreasing rainfall, and increasing frequency of drought,
pastoralists in the region face other threats and challenges. The region is remote and has poorly
developed infrastructure, limited availability of services and low levels of education and literacy.
Access to traditional grazing land has been steadily reduced due to settlement and the expansion of
arable farming, especially irrigation schemes in the fertile valleys that have occupied areas that
were previously used for dry season grazing.
Widespread encroachment by the alien invasive shrub Prosopis spp, originally introduced to
stabilize banks of irrigation channels, has created dense thickets which are now inaccessible to
livestock, whilst indiscriminate development of bore-holes has contributed to degradation of some
wet season grazing zones. The pastoralists of the Somali Region of south-eastern Ethiopia are more
closely integrated with neighbouring Somalia and the Gulf States than the rest of Ethiopia.
Unofficial cross-border trade in livestock from Ethiopia to Somalia - from where they are re-
exported to Gulf States - is estimated to exceed formal trade by a factor of 100. The Ethiopian
government, however, regards this historic trade, in which cattle are often swapped for consumer
goods which are then sold on, as ‘contraband trade’ and attempts to suppress it. There have also
been a number of bans by the Gulf States on the import of livestock from the Horn of Africa region
over recent decades due to outbreaks of Rift Valley fever. There is perceived to be an anti-
pastoralist bias by the dominant highland culture in Ethiopia, and civil unrest and general insecurity
in the region compound an already difficult situation.
This case study is focused on livestock-related aspects of the emergency response in south-eastern
Ethiopia during the drought of 2005/06.
33
2.3.2 The drought of 2005/06
The 2005/06 drought is estimated to have affected 1.7 million pastoralists in south-eastern
Ethiopia, the vast majority of who were ethnically Somali with the remainder being members of the
Borana community. The worst affected areas were the Somali, Oromiya and Afar regions.
2.3.3 Drought emergency management in Ethiopia
Ethiopia has a tiered government system: in addition to the national level, a federal government
oversees regional states, zones, districts (woredas) and neighborhoods (kebele). Under the
constitution, considerable power is delegated to regional states which can establish their own
government and implement their own policies through an executive committee and regional
sectoral bureaus. Following on from the devastating droughts and associated famines of the 1980s,
drought response in Ethiopia is now regulated by the government’s 1993 National Policy for Disaster
Prevention, Preparedness and Management, while the Directives for Disaster Prevention and
Management articulates implementation modalities and the associated institutional set up.
Since 1993 to the present day there have been a series of changes to the agencies responsible for
implementing these policies. First the Relief and Rehabilitation Commission, later the Disaster
Prevention and Preparedness Commission (DPPC), and more recently the Disaster Prevention and
Preparedness Agency (DPPA) have in turn been responsible for developing national infrastructure
and systems to assess and respond to food aid needs. In early 2004, the Food Security Coordination
Bureau was created to focus on the causes of chronic food insecurity, and the mandate of the DPPC
(now the DPPA) shifted to responding only to emergencies and unpredictable events. So,
responsibility for interventions are divided between the DPPA, which responds to the acute needs
of the unpredictable food insecure, and the Food Security Coordination Bureau (FSCB) within the
Ministry of Agriculture & Rural Development (MoARD), which focuses on improving long-term food
and livelihood security for the chronically food insecure through safety net interventions and other
food security programmes. The DPPA retains responsibility for coordinating emergency responses
for both food and non-food assistance in Ethiopia. The Early Warning Department (EWD) within the
DPPA is responsible for early warning activities in Ethiopia. The diagram below shows the linkages
at the various levels from local to national. The Early Warning Working Group, made up of line
ministries, UN agencies and NGOs, contributes to the early warning monitoring capacity through
four main mechanisms:
• regular monitoring of key standard indicators collected in disaster prone woredas
• pre and post-harvest assessments
• disaster area assessment (for assessing the impact of rapid-onset disasters such as flooding)
• pastoral area assessment
• In 2005, in an attempt to change the focus from emergency assistance towards longer-term
development, the government launched the Productive Safety Net Programme (PSNP). Under the
PSNP, the Ethiopian government makes a distinction between two groups:
• The ‘unpredictably food insecure’, who face transitory food deficits because of erratic weather
or other livelihood shocks. This group continues to receive food aid and other humanitarian
assistance as needed.
• The ‘predictably food insecure’, who face chronic food deficits because of poverty rather than
food shocks, receive cash or food transfers, for work or freely, on a regular, predictable basis
34
for a fixed period of five years. Together with complementary interventions such as ‘livelihoods
packages’, it is envisaged that this should enable these households to escape from their chronic
food insecurity over time after which they will no longer receive any social assistance except
during emergencies.
However, by the onset of the 2005/06 drought the PSNP had not yet been implemented in the main
pastoralist areas. The main mechanism for resource mobilization during an emergency is an appeal
made by the Government of Ethiopia, UN agencies and humanitarian partners, known as the Joint
Humanitarian Appeal (JHA). Such ‘emergency’ appeals have been launched every year for the past
three decades.
Structure of DPPA :
NationalDisasterPreventionandPreparednessAgency(DPPA)
(DPPAassecretariat)
NationalEarly
WarningSystem
NationalCrisis
Management
RegionalDisasterPrevention
andPreparednessCommittee
RegionalEarly
WarningSystem
RegionalCrisis
Management
ZonalDisasterPreventionand
PreparednessCommittee
ZonalEarlyWarning
System
ZonalCrisis
Management
WoredaDisasterPrevention
andPreparednessCommittee
WoredaEarly
WarningSystemat
MoA
Development
Agents(DA)
Communit
y
NGONational
Office
NGORegional
Office
NGOZonal
Office
NGOatWoreda
Source: Based on Abate 2003
35
As mentioned earlier, under the drought early warning system, monthly monitoring provides
primary data on food production, price and market situation, human and animal health, and onset
and distribution of rain at the woreda level. In addition there are multiple seasonal assessment
exercises for verification. In addition, several NGOs have their own early warning mechanisms, with
different types of data collected as a ‘by-product’ of the field monitoring of their regular
programme operations. The following table provides an example of the successive levels of
assessments and related activities that took place in Borena zone (Oromiya Region) and Leban zone
(Somali Region) during the early stages of the 2005/06 drought. CARE is an international NGO.
Table 3: The successive levels of assessments and related activities
Week 1 CARE monthly drought monitoring report triggers the convening of the Emergency
Coordination Meeting
Week 2 A rapid assessment is undertaken, led by government zonal authorities
Week 3 Assessment report discussed at the Emergency Coordination Meeting
Week 4 Report forwarded to Oromiya Regional Government for discussion
Week 5 Oromiya regional government discusses report and forwards to the federal
government
Weeks 1–5 CARE and others undertake a nutritional survey in affected areas
Week 6 CARE presents the nutritional survey report to the Emergency Coordination Meeting.
Federal and the regional authorities hold consultations with the zonal team
Week 7 The federal and regional governments assemble an assessment team to visit the
field. This assessment team is joined by the zonal team? who are
Week 8 The Federal and Regional Assessment teams provides feedback to the Zonal
Coordination Meeting
Week 9 Assessment report submitted to the federal government and discussed
Week 10 The federal government issues an appeal and allows
response by public zonal,
regional and federal agencies. CARE had begun water trucking for domestic use
before the launching of the government appeal and the declaration of emergency.
Woreda officials were also undertaking water trucking. Under normal circumstances
NGOs must wait for a federal declaration and appeal before they can intervene.
Source: Pantuliano and Wekesa (2008)
2.3.4 Responses to the 2005/06 drought
Responses to the drought included both life-saving and livelihood support-based interventions,
although overall the majority of resources went to food aid. Non-food interventions included
health, nutrition, water and sanitation, and agriculture - including livestock-focused activities.
Livestock-focused initiatives were undertaken on a relatively limited scale and targeted at a few
woredas. In total it was estimated that some 1.7 million pastoralists were affected by the drought
and the DPPA estimated that 54 million livestock required emergency and recovery interventions:
with the exception of vaccination, the livestock interventions described below, however, benefited
at most a few tens of thousands of households with practical considerations, such as accessibility
and proximity to paved roads, being amongst the main determinants in being reached. Vaccination
of livestock against pasteurellosis, blackleg and anthrax, which involved by far the greatest number
36
of livestock, still only covered around 4% of the at-risk population. Livestock-related interventions
included:
• Commercial destocking: This involved the facilitation by an NGO (Save the Children USA)
and MOARD of commercial livestock traders to engage with pastoralists. The NGO provided funds
for an awareness raising campaign targeted at livestock traders and abattoir owners, which
included radio and television announcements. This campaign resulted in 40 or so traders
attending a meeting held in mid-January 2006. As a direct result, 21 of the traders travelled to
Somalia Region State in early February in transport provided by the NGO. Ultimately, however,
just two traders went on to purchase cattle. Thin cattle were bought by the traders using their
own money, although this was done before the condition of the cattle had deteriorated too
much due to the drought. The cattle were transported out of the drought area, fattened and
subsequently exported as beef to Egypt. This activity was undertaken in the Moyale woreda of
Oromiya region. In total an estimated 20,000 cattle were purchased by the two traders for a
total of US$ 1.01 million: 5400 households benefited by, on average, US$ 186 per household.
This is equivalent to an average of just US$ 50 per head of cattle purchased, which would
appear to be a highly discounted price.
• Slaughter destocking: Various NGOs also undertook slaughter destocking operations, in which
cattle were purchased, slaughtered and processed into dry meat which was distributed to
vulnerable households. In Dire woreda, Borana Zone, a destocking initiative involved the
slaughter of 2411 animals from 1121 households, with 2814 kg of dried meat being distributed to
1301 households.
• In total, through both commercial and NGO-implemented slaughter destocking (see above), one
estimate suggests that up to 75,000 cattle were destocked although precise figures are lacking.
• Supplementary feeding: Various initiatives targeted the feeding of breeding animals (cattle or
goats) with hay, straw or concentrates, combined with provision of water, deworming and
other veterinary treatments and vaccinations. For example, in parts of Dire woreda, Borana
Zone, supplementary feeding was undertaken for 10,763 animals with 18,126 bales of hay or
teff straw. A financial cost-benefit study of one intervention, in which cattle from 800
households in Zone 4 of Afar region were fed for two months with concentrates, showed that
the supplementary feeding programme was more than five-times less expensive than allowing
the animals to die and then undertaking restocking. In a similar intervention targeted at goats,
2300 households in Zone 1 of Afar region were covered.
• Livestock treatment and vaccination: In addition to the treatments and vaccinations
administered as a component of the above mentioned supplementary feeding programmes, up
to 2 million animals were vaccinated in Borana zone using FAO funds. For example, in an
initiative in parts of Dire woreda, Borana Zone, livestock from 10,763 households were covered
by vaccination, spraying, deworming and treatment of their livestock.
37
Table 4: Sequence of events in the 2005/06 drought
Date Events with a focus on livestock interventions
On-going Each woreda
develops drought contingency plans
Mid-March to end May 2005 Gena
rains below average
July 2005 First signs of drought in Borana Zone
August 2005 First early warning of drought
Water trucking started in some locations
Mid-September to mid-
November 2005 Hageya
rains fail
October 2005 Pastoralist Livelihood Initiative (PLI) funded by USAID as a
drought mitigation and preparedness programme
Save the Children US field staff reported the onset of drought
November DPPA Crop and Needs Assessment: indicated 1.5M people will
need food aid; 0.6M in acute water shortage
Livestock start to die in Moyale woreda
December UNICEF begin to prepare Regional Multisectoral Response Plan
First emergency meeting at Regional Presidency
Dams in Moyale woreda completely dry
NGOs launch initial emergency livestock interventions
31 December 2005 Kenyan Government declare drought a national disaster
November 2005 through to
February 2006 Interventions,
including livestock health,
initiated by agencies
which were not dependent upon the humanitarian appeal and
which could access funding through alternative mechanisms
Mid December Save the Children US undertook drought assessments in Liben
and Afder Zones in the Somali Region: concluded that Moyale
District was in the alert phase of the drought cycle and without
rain or appropriate intervention, could move to the alarm
phase within the following four to six weeks
Pastoralists employing drought coping strategies, including
movement of livestock to Hudet and Filtu where slightly better
deyr rains and pasture were recorded
December 2005 to March 2006 Key interventions conducted in southern areas of Somali region
and Borena zone of Oromiya Region: vaccination of up to 2
million animals and destocking of approximately 75,000 head
of livestock
17 January 2006 Awareness-raising meeting for livestock traders: number of
traders expressed an interest in travelling to the drought-
affected areas to explore the possibility of purchasing drought-
affected livestock
23 January 2006 GoE emergency international appeal launched
January 2006 Large numbers of cattle die in Moyale woreda
January/February 2006 Multi-agency assessment
Late January/early February
2006 PLI partners established a Commercial De-stocking Working
Group
Draft Modalities of the Provision of Short-Term Loans to
Livestock Traders prepared and circulated by PLI
February Commercial destocking undertaken in Moyale’s two districts
38
Date Events with a focus on livestock interventions
(Oromiya and Somali)
Pastoralists in Moyale move some of their animals to other
places
Livestock feed supplement programme started by NGOs CARE
and GAYO
15 February 2006 FAO appeals for US$
18.5 million for livestock and agricultural
assistance throughout country
9 March UN agencies agree on division of the first US$ 10m tranche of
CERF funds (for the Horn of Africa): majority earmarked for
health and nutrition, water and sanitation, and livestock
16 March Appeal by Kjell Magne Bondevik, the UN Special Humanitarian
Envoy for the Horn of Africa
March Showers occur in Oromiya region
Destocking programme stopped
End March 2006 Humanitarian Response Fund (HRF)
established
April Gu
rains begin in Somali Region
Migrated livestock return
Emergency livestock vaccination and treatment started in Dire
woreda
May 2006 Dry meat distributed in Dire woreda
Late September 2006
HRF had disbursed US$
6.6 million since established in March
2.3.5 External support to the Ethiopian government
A variety of sources and instruments were utilized to fund the emergency response activities. The
JHA was launched in late January 2006. In total the appeal was for US$ 166 million, made up of US$
55 million for food aid and US$ 111 million for non-food aid (health, and nutrition, water and
sanitation, and agriculture including livestock). The relatively low proportional share for food aid
was due to a combination of a large carryover from 2005 of food stocks and pledges, and the
establishment in 2005 of the PSNP. The actual donor response to the appeal, however, resulted in a
significant shortfall, with just US$ 38.5 million being realized for food aid and US$ 36.6 million for
non-food aid. The FAO’s component of the JHA included US$ 2 million for emergency livestock
health interventions and US$ 1 million for destocking activities.
Four sources of funding were of particular importance to livestock-focused interventions. The
Pastoralist Livelihood Initiative (PLI), a two-year long USAID-funded drought mitigation and
preparedness project, was established in October 2005, together with funds held by FAO from the
Belgian Government, and this provided an opportunity to test livelihood-based relief opportunities,
including livestock-oriented ones. In addition, some NGOs utilized their own funding to enable
timely implementation of initiatives including destocking and provision of supplementary livestock
feed. Finally, use of the expanded Central Emergency Response Fund (CERF) saw the UN agencies
share US$ 10 million of CERF funds on 9th March, with the majority earmarked for health and
nutrition, water and sanitation and livestock interventions, and minimal support (US$ 250,000) for
food assistance.
39
2.3.6 The drought response unraveled
The major actors involved in the emergency drought response were the Ethiopian Government at
the various levels, from the most local, woreda, through zonal, regional and federal, to national;
international NGOs based in Ethiopia; UN organizations; the Ethiopian private sector; and the
pastoralist communities in the drought affected areas.
Ethiopia’s federal system, where policy and decision-making powers are devolved to the
autonomous states, makes it difficult to have a coordinated national response. As a result there
was no national-level contingency plan or funding system, and while there were contingency plans
and funding arrangements in place at local and federal levels, these were not coordinated and
there was no overall coherent approach. As part of the drought preparedness policy the woredas
are expected to develop their own contingency plans but they largely lack the capacity to do so.
The government institution responsible for emergency drought response in Ethiopia has changed
several times since new policies were put in place in 1993, which in turn were a response to the
devastating droughts and associated famines of the 1980s. The latest changes, which are intended
to direct more attention to tackling the chronic causes of food insecurity, were in the process of
being introduced as the 2005/06 drought took hold, but at that time had not been implemented in
the pastoralist areas that were worst affected. The private sector in Ethiopia is relatively poorly
developed and Ethiopia is regarded as a difficult place to do business, ranking 107th out of 184
countries listed in the World Bank’s Ease of Doing Business rankings for 2009. Although reforms
aimed at transforming Ethiopia from a centrally planned economy to a market-oriented one were
launched in 1991, after the overthrow of the former pro-Soviet Derg regime, the potential
contributions of industry and the private sector remain largely untapped.
As the drought began to unfold there were a number of monitoring and early warning systems in
operation. The official system, implemented by the DDPA, required information to flow from the
most local level, woreda, up through the various tiers of government, eventually reaching the
national level. The DPPA early warning system is often constrained by inadequate resources, in
which case NGOs try to step in to fill the gap. And NGOs and international organizations also carry
out their own regular, independent assessments as part of the on-going monitoring of their
interventions and development programmes. During the 2005/06 drought there seemed to be clear
duplication of effort: federal and regional government repeated and verified assessments already
undertaken by zonal governments or NGOs, delaying decision-making and timely response. The
reasons for this seems to be lack of trust between the tiers of government, and also between
government and NGOs, lack of coordination between the various players, and a general lack of
confidence by government in data collected in the field. Marginalization of pastoralist communities
and their lack of influence in government and political power are also likely to be important
contributory factors. And despite all this emphasis on, and investment in, monitoring and early
warning, in the 2005/06 drought this failed to translate into timely action, especially for livestock-
based interventions, some of which were only beginning to take off as the rains returned.
Most emergency interventions consisted of food relief, with only limited livelihood-based
interventions. Assessing emergency non-food needs is considerably more difficult than assessing
food needs and is a relatively new and emerging approach that requires new and different skills
which are generally lacking, especially at the more local levels to which responsibilities tend to be
devolved.
40
Despite the first signs of drought being apparent and flagged up in July 2005, interventions in
general and the livelihood-based interventions in particular – including destocking and feeding of
livestock- were generally late: the first externally implemented, livestock-oriented interventions
occurred in November with most not happening until 2006. Reasons for the tardiness of response, in
addition to the lack of capacity noted above, included inflexibility and unresponsiveness of
procurement procedures, and poor coordination, especially amongst NGOs. In addition there was a
lack of policy, institutions and legislation to support timely delivery of livelihood-based
interventions, and a systematic failure to facilitate the sharing of ideas and experiences and in
doing so create the framework conditions needed to mobilise knowledge to meet the diverse and
ever-changing needs of the sector. Exposure to the knowledge and experience of others can help
organizations and policy makers translate principles into operational strategies relevant to their
own contexts.
During the 2005/06 drought various livestock-oriented interventions took place, although their scale
and area of impact was relatively limited. A significant source of the funding for them came from a
short duration, donor-funded project which, by chance, was just being established as the drought
took hold. Fortunately this project was able to be sufficiently flexible to allow a change of a
direction which created the ‘space’ for some of the livestock-oriented emergency interventions to
take place; however, this source of funding will be not be available next time a drought occurs, and
so is not part of a sustainable solution. Although many of these livestock-based interventions could
be viewed as pilots, it is difficult to see how most could be significantly scaled up. In addition,
given that these initiatives were not set up in way that would enable rigorous evaluation and
learning – both at the practice and policy level- there is a risk that the same mistakes will be made
next time around.
The campaign to create awareness of the opportunity for commercial destocking, promoted by the
NGO Save the Children US, initially attracted the interest of around 40 livestock traders, but of
these just two eventually purchased cattle from pastoralists and the activity was focused in only a
few locations close to main roads. Traders were aware that roads in the more remote areas were in
poor conditions and they knew that transporters would charge more to operate in these areas; in
any case the two traders involved could buy the relatively small number of cattle they needed from
the more accessible areas and so had no reason or incentive to travel further into the hinterland.
The poor state of roads in the more remote pastoralist areas – a glaring indicator of the
marginalization of pastoralist areas in Ethiopia - is clearly a major impediment to larger scale or
more widespread commercial destocking; it has been reported that Ethiopia has the lowest road
density per capita in the world. The pastoralists were initially sceptical that the traders would buy
their thin cattle, but once they had witnessed this happening they soon realized that this was an
opportunity to sell cattle in poor condition that, as the drought progressed, were likely to die, and
to use the money obtained to buy food for their families, feed for their remaining core livestock,
pay for veterinary services or purchase more drought-tolerant sheep and goats in place of cattle.
Many pastoralists lacked confidence in dealing with livestock traders and were unaware of what
constituted a fair price. They therefore relied on trusted and more market-savvy fellow pastoralists
to negotiate with the traders on their behalf. This system also suited the traders as they were able
to purchase large batches of cattle in single transactions. During the commercial destocking
initiative cattle were not sold in a conventional livestock market setting: with just two buyers the
usual competitive marketing system clearly could not operate. However, surprisingly, it was
41
reported that although prices paid varied widely and in some cases these were below ‘normal’
market prices, in other cases relatively high prices were paid for the thin cattle with these prices
apparently driven by the strong Egyptian export market for beef. With the emergency slaughter
initiative, in which a number of NGOs purchased thin cattle, slaughtered them, dried the meat and
distributed this to vulnerable local households, the scale of operation was even smaller than the
commercial destocking operation. Traditionally, Borana pastoralists avoid eating meat from
emaciated animals and there was a widely held preconception that the dried meat would not be
culturally acceptable. However, the pastoralists who received dried meat reported that they liked
the taste, indeed that it represented an improvement on their normal, rather bland maize-based
diets. In a study undertaken in Dire woreda it was found that dried meat made up almost a quarter
of the household food of beneficiaries. It was even reported that although the NGOs targeted the
most needy households (as judged by local people), even the better-off households appreciated the
dried meat and, in keeping with traditional practice, the meat was shared throughout the entire
community. The combined impact of the two destocking schemes was highly localized and overall
benefited less than 2% of the total number of pastoralist households affected by the drought.
Despite the small scale, those pastoralists who did benefit reported that they considered the
destocking interventions as amongst the most innovative and beneficial of all interventions,
although they would have preferred that they had occurred earlier in the drought cycle. Provision
of livestock feed by external agencies, whether as dried grass, teff straw or concentrate feed, was
also highly appreciated by the pastoralists, although again this activity began too late and occurred
on a relatively small scale. Clearly the greatest need for such emergency feeds will occur during
the most severe droughts when supplies will be most restricted and, as in the commercial
destocking intervention, the poor state of feeder roads in the more remote rural areas will act as a
major barrier.
The largest scale livestock-based emergency drought intervention, vaccination, still only involved
less than 4% of the livestock population thought to be affected by the drought, and the
appropriateness of this intervention is questionable. The vaccines used in 2005/06 were against the
diseases pasteurellosis, blackleg and anthrax. A recent study (Feinstein International Centre 2007)
of the impact of vaccination of livestock in droughts in Ethiopia concluded that during drought
years there was no significant difference in livestock mortality, for any species, in vaccinated
compared with non-vaccinated herds. Also, the vaccines administered in 2005/06 were not for the
diseases that pastoralists ranked as most important, nor were they necessarily for the disease that
veterinary authorities believed caused the highest mortality rates in the targeted areas, which
tended to be CBPP for cattle and CCPP and PPR for sheep and goats. The study authors concluded
that while vaccination should be a standard preventive measure during normal years, poorly
designed and implemented vaccination campaigns undertaken during droughts were of doubtful
utility. A major driver of livestock vaccination during droughts appears to be that they are a
relatively cheap and easy intervention to implement rather than any evidence that vaccination at
this time can reduce mortality.
In addition to the external interventions, pastoralists themselves adopted a number of their own
coping strategies and implemented these much earlier in the drought cycle than the external ones.
Key amongst these was moving cattle to areas where water and grazing were more available,
although this caused problems when pastoralists moved onto land belonging to other clans. In
addition, on-going insecurity meant that some land that could usefully have been used as dry
season grazing went unused. Pastoralists suggested that NGOs could work with traditional
organizations to help overcome such access problems. In addition to trekking animals in the
42
traditional manner, during the 2005/06 drought some pastoralists were innovative and hired
vehicles to truck breeding animals out of the drought area. Other local innovations observed during
the drought were for irrigated maize crops to be cut early by agropastoralists for use as
supplementary livestock feed. Some pastoralists also reported that they would be willing to
purchase concentrate feed at full market prices to help ensure key breeding animals survived, but
that such feeds were not available locally (Feinstein International Centre 2007).
43
3. THE WORKSHOPS AND ANALYTICAL FRAMEWORK
A parallel component of this programme of work consisted of two workshops organized by IGAD–LPI
in mid 2008. The workshops brought together senior civil servants with responsibility for livestock
and animal health issues, with parliamentarians, representatives of livestock marketing
organizations and NGOs from Djibouti, Ethiopia, Kenya, Somalia, Sudan and Uganda, in addition to
representatives of IGAD-LPI, FAO, Africa Centre for Technology Studies (ACTS), LINK ( Learning,
Innovation and Knowledge) and donors. The overall objective was to gain a better understanding of
the types of competencies and skills that contribute to policy-relevant capacity in the region’s
livestock sectors through sharing knowledge and experience on livestock policy processes.
During the workshops, the participants revisited the World Bank’s basic agricultural innovation
capacity analytical framework (actors and their roles; patterns of interaction; habits and practices;
enabling environment) and developed this further by adding sub-headings under the four main
headings as follows:
Actors and their roles
- Diversity / relevance of actors involved – stakeholder analysis
- Power and influence of different actors
- Roles performed by different actors, evolution of roles;
- Levels of participation;
- Presence of initiators
- Different actors involved in different stages of policy process;
- Sector coordination;
- Presence of ‘champions’
Patterns of interactions
- Dynamic x respond to challenges or opportunity
- Network of linkages – transparent, evolving, appropriate
- Frequency of interaction – evolution of interactions
- Appropriate mechanisms (locations, issues, actors)
- Platforms
- Rules of engagement
- Continuity of engagement; institutionalization of engagement
Habits and practices
- Processes of interaction
- Diversity and quality of knowledge sources and pools
- Information sharing
- Monitoring and review
- Learning
- Quality of interactions
- Culture, culture of persons as pertains to social interaction and networks
- Time frame
- Institutional history/background/baggage
44
Presence or absence of an enabling environment
- Incentives (both positive and negative)
- Needs and triggers
- Context
- Goodwill / Political will
- Good governance
- Anti-corruption
- Room for arbitration
- Capacity (competencies, skills, resources, information)
- Institutional capacity – capacity to participate
- Social standing / influence of policy champions/advocates
Participants were also introduced to practical approaches that can be used to track change in
policy processes. The long-term goal is to facilitate changes that make policy processes operate in
more pro-poor ways – and thus that social and economic outcomes are mediated through the
process of institutional change. Participants learned, however, that, as there is generally a long
time lag before tangible social and economic benefits arise that can be assessed, there is a need, in
the mean time, to monitor and report changes in process. Hence, lessons that emerge from
interventions with institutional change become key indicators of progress and a qualitative means
to assess the direction of institutional change (Appendix 2).
45
4. SO WHAT DO THE CASE STUDIES AND THE WORKSHOPS TELL US
ABOUT BUILDING INNOVATION RESPONSE CAPACITY?
4.1 Results of the Case Studies
The results from the case studies largely endorse the elements identified as essential
characteristics of innovation response capacity. The analysis during the workshops also highlighted
additional issues that are of importance, but rather than providing new analytical elements these
were generally a more detailed sub-set of the main characteristics identified in the conceptual
framework. In the ‘enabling environment’, some additional features were identified and the
innovation response capacity conceptual framework has been revisited and revised to reflect this:
4.2 Actors and their roles
Sector coordination: The case studies revealed a broad array of sector coordination arrangements;
ranging from its virtual absence to the presence of ‘over’ coordination. In both cases, the speed
and coherence with which decisions are taken, and appropriate action is implemented, suffers. In
the few examples where sector coordination was used to broker new relationships, such action
significantly improved coherence and creativity within the response. However, in many cases,
coordination was used to control, rather than to facilitate, and as such prevented rather than aided
the building of innovation response capacity.
In cases where rapid, coherent action ensues or where newly formed coalitions of actors allow new
practices and approaches to flourish, this is often a result of third-party brokering, rather than
sector-wide coordination. Whilst such third-party negotiation is often necessary and vital, it is in
most cases not sufficient to establish the sector-wide coordination needed to generate the required
trust and social capital through interactions and linkages that change ways of working, or to ensure
the evolution of the myriad of complementary roles and skills through the cross-fertilization of
knowledge stocks that contributes to the capacity to innovate.
Sector coordination, however, involves more than just the brokering of the necessary linkages or
the elimination of obstructions and similar task - now commonly described as innovation brokering.
It also provides urgency, direction, specificity and identity to a network of actors as it confronts
particular issues. It thus requires actors that command the sector’s broad trust, have an
institutional and organisational overview, and skills that can lubricate an appropriate patchwork of
interaction and competencies. In examples where this has functioned well, such tasks have either
been fulfilled by a body set up specifically through the concerted actions of a critical mass of
sector actors, or where such tasks have gravitated organically towards a broadly recognised sector
champion. Moreover, the two case studies could also clearly benefit from greater regional
coordination. Thus far the national, regional and international organizations involved have
glamorously failed to even suggest a hint of moving towards this.
Champions and brokers. Whilst additional issues identified in the workshops were largely
additional qualifiers, one that does require specific mention is case of ‘champions’. Not only can
such actors play an important role in sector coordination, advocacy and the championing of new
46
approaches, they are also of vital importance in an innovation brokerage role to actively link
different actors. In the case of the drought episode, this role was played by some projects and
NGOs that, albeit on a limited scale, brokered linkages and set up mechanisms for interaction
between diverse actors. Unfortunately, the ‘project’ manner in which this was operationalised
largely prevented the building of the appropriate institutional capacity. It did, however, clearly
indicate the potential that the inclusion of such an innovation brokerage role within the
institutional architecture could have.
Entrepreneurship: The current ‘emergency industry’ has not been kind to entrepreneurs and
entrepreneurship in general. The investment ‘climate’ revealed by the case studies clearly shows
that under the current conditions any signs or budding involvement by entrepreneurs in the sector
are prevented or destroyed through the wholesale take-over of the sector by both the public and
tertiary sector, often with the tacit agreement of the donor community, during ‘emergencies’.
Whilst ‘experiments’ in other developing regions and sectors actively look to stimulate a nascent
private sector through the strategic allocation of public and/or donor monies, the emergency
industry in the Horn of Africa is still largely controlled by the public sector. As indicated, civil
society and donors are often equally guilty of such actions, as their short-term project
interventions frequently destroys existing or burgeoning systems and linkages. The attempt at
getting livestock traders involved in the emergency response in Ethiopia was interesting; as in
general, traders are scared away by the take over of markets during severe droughts by the public
and tertiary sector. Whilst the ‘experiment’ was relatively small and not necessarily an unqualified
success, it clearly showed a potential role that for entrepreneurship in emergency responses. It also
indicated one of the many potential manners in which the public and tertiary sector can facilitate
and broker private sector involvement, rather than temporarily performing such roles themselves -
creating and leaving institutional voids, rather than building institutional capacity.
Thus far, the opportunities for the private sector involvement in RVF responses have been relatively
limited. New technology, however, which permits the combination of various vaccines in one single
dose, is likely to provide treatment opportunities that will be able to combine public good with
private interests for private gain. Thus far, however, this remains a distant promise; innovation in
livestock production, processing, utilisation, and distribution only takes place where different
players in the sector are well networked together allowing them to make creative use of ideas,
technologies and information from different sources, including from research.
Providing knowledge of the future: Both episodes studied deal with sophisticated systems
designed to provide early warning and, essentially, in both cases such early warning was indeed
provided. The analyses also clearly indicate, however, that the value of such knowledge of the
future, without being anchored to field realities and response capacities, is esoteric: In neither
case did it assist the timeliness, or local specificity of the eventual response, nor did it contribute
significantly to the building of the required capacity.
Research: Whilst it is true that not all innovation requires research, it is also clear that a
technological breakthrough in terms of a longer-lasting or ‘cocktail’ vaccination would potentially
provide additional options in dealing with RVF. Whilst, as indicated, some moves are being made in
that direction, in the absence of concomitant changes, such technological breakthroughs are
unlikely to provide the social and economical impacts sought.
Systems changes cannot be adopted by individuals; rather they have to be agreed and delivered, by
definition, by the broad network of players involved in turning inventions into innovation.
47
Public research on weather related ‘emergencies’, both national and international, has thus far
largely preoccupied itself with early warning and emergency food aid distribution. Whilst there
have been some attempts to move beyond this, in general research and research thinking around
these areas is still firmly stuck in the old paradigm, often generously incentified through specific
research investments for those topics.
Service provision: In geographically remote areas, where services are basic at best, expectations in
respect of their contribution to innovation responsive capacity should be similarly basic. Even the
oft touted veterinary services are little exception to this rule. Whilst undoubtedly necessary, in the
absence of a broader interpretation of service provision, and with the recurrent lack of the
integration of community-based services with actors higher up the veterinary chain of command,
they contribute little to innovation and innovation response capacity. So, whilst officially accepted,
some of the findings in these studies suggest that their continuing tacit rejection and lack of
integration may actually be contributing to the worsening of surveillance and response capacity.
Ensuring socially desirable outcomes: Ensuring socially desirable outcomes have become an
important international political currency. The thinking in relation to this in respect of
‘emergencies’, however, has evolved little beyond rescue and relief. In other words, the one
socially desirable outcome pursued is the absence and prevention of death. Whilst clearly
understandable from a humanitarian point of view, this limited interpretation also obviously entails
that long-term socially desirable outcomes will only be achieved if the underlying thinking that
currently drives behaviour in the aid and emergency sector changes. Moreover, such limited views,
and the concomitant actions such as settlement and the creation of economic dependence, have
often shown to be instrumental in perpetuating social inequities and play into the hands of a
political agenda which looks to better control peripatetic social groupings.
4.3 Patterns of interaction
Links to consumers: Although often maligned, there are few doubts that traders (buyers) or
middlemen are a critical source of information on the changing preferences in distant markets.
They are also an integral part of the intricate network of interactions among actors essential to the
functioning of pastoral systems and the maintenance of trade links between the Horn and the
Arabian peninsular. Whilst such linkages have proven to be resilient and can be reconfigured
relatively quickly, the interventionist nature of the measures taken during livestock sector-related
‘emergencies’, has thus far generally prevented the amalgamation of existing networks and the
strengthening of available capacity through the exploitation of potential synergies.
Links to specialist technical knowledge: Changing conditions often require responses through the
upgrading of technology, capacities and systems. Although the studies revealed some evidence of
‘upgrading’, ‘mainstream’ technical knowledge has thus far not been able to fruitfully engage, or
link in an operationally sensible manner, to these processes. This is not only the result of the
marginalization of sector actors in the generation of appropriate knowledge, but it is also
reinforced through the lack of recognition that fundamental changes have taken place which
represent a new normality, and that significant revision of the paradigms that have, thus far,
dominated sector thinking is required.
48
Mechanisms that facilitate interaction: There appear to have been few real attempts at setting up
mechanisms that could knit an innovation fabric amenable to rapid reconfiguration. In the limited
‘experimentation’ with such mechanisms that was observed in the case studies, there were signs
that this indeed aided reconfiguration of patterns of interaction. Unfortunately, ‘membership’ in
such experiments was rather limited and focussed predominantly on satisfying the immediate needs
of pastoralists, rather than also attempt to provide a degree of systemic coherence and interaction.
Moreover, the mechanisms were set up in a ‘project’ way; rather than building institutional
capacity they just filled institutional gaps for a limited period. Limited as the experiences gained
may be, thus far it is clear that mechanisms which facilitate linkages between diverse sector actors
and knowledge stocks can be instrumental in enhancing the timeliness and appropriateness of
responses.
4.4 Habits and practices
Attitudes towards change: Public, private and tertiary sector organizations exhibit varying degrees
of reluctance to change. Although this is generally believed to be most marked in public sector
organisations, the case studies revealed many of the same characteristic in civil society
organizations. With vested interests continuing to be shored up by an ample flow of donor funding –
apparently oblivious to their gate keeper role in this respect - progress in these areas will remain
slow as there is no real incentive for any of the ‘emergency’ industry actors to reflect, share turf,
engage in debate, or to change behaviour.
Trust: Whilst a cursory look at the relationships between the parties involved in ‘emergency’
responses may suggest a willingness to cooperate, evidence form the case studies suggest
differently. Collaboration and coordination generally does not stretch beyond the partition and
allocation of geographical areas to specific organizations. The glaring lack of trust revealed by the
studies, however, is best illustrated by a general ‘we’ll have to see for ourselves’ attitude, rather
than to accept opinions, assessments and diagnoses by other actors.
Shared identity: Much as in the case of trust, it is clear from the studies that in most cases there
does not seem to be enough social capital among actors to be speaking of a shared identity that
would aid coherence and responsiveness. And whilst there may be degrees of shared identity among
some of the same actor groupings, even here the case studies reveal more incidences of turf
protection rather than a shared sense of purpose. It is also clear from the studies, however, that
when this is achieved, even in small measures, it greatly aids the potential to find creative and
sensible solutions, often to long-standing problems.
Learningness: In the case of the drought case study there appear to be mechanisms that could aid
learning; there is, however, limited actual evidence of any learning occurring. This should not come
as a surprise, as it is largely a reflection of the observed relationships among sector actors.
Moreover, recent ‘experiments’ that broke the ‘rescue and relief’ mould, whilst clearly showing
potential, have, thus far, been too limited in nature and scope to have a significant impact on the
emergency industry. It is also undoubtedly true that there is intrinsic learning within staff. It is
unclear, however, to what extent such tacit knowledge is retained due to the relatively high
turnover of personnel and organisations.
49
National culture: Whilst it is not useful to engage in a discussion as to whether certain cultural
traits are constraints to innovation and innovation responsiveness, it is, at the same time,
important to point out that prevailing opinions among governments and non-pastoralist society,
coupled with the geographical remoteness and poor access to infrastructure and services, add to
the political and social marginalisation of the areas and its inhabitants.
4.5 Enabling environment
Clearly the enabling environment, or rather the absence of an enabling environment, manifests
itself through the factors already discussed. The two cases clearly indicate that existing policy
frameworks, respective responsibilities of ministry and governance arrangements of the sector are
hardly helpful. However, rather than measures being taken to remediate these issues, there
appears to be indication that many of these barriers are actively kept in place, or strengthened for
social and political reasons.
In the concomitant absence of incentives for change, and in the continuing reluctance of donors
and civil society to fulfil part of their responsibility as gate keepers, it is hard to see how the wider
set of policies and institutions in which the innovation process is situated will contribute to the
capacity to innovate and innovation responsiveness.
50
5. CONCLUSION
5.1 Where to Start: The 4cs that build Innovation Response Capacity
With the case studies confirming the specific analytical elements that can be used to explore
innovation response capacity, it is probably useful to explore the implications of how one could
best start to think about building this capacity and the indicative actions that investment in such
capacity should be directed towards.
Comprehensiveness
Wide-ranging perspectives assist sectors and countries to deal with the challenges and demands of
today’s changing world. Such viewpoints not only promote inclusiveness, systemic sector
understanding – including links with other sectors and the wider economy - and interaction between
diverse actors, it also instils an understanding that interventions need to have a long-term
perspective and give sufficient emphasis to facilitating an experienced-based, incremental process
that develops sector capacity.
Generally, neither the evolving set of opportunities presented to, nor the problems faced by, a
sector can be exploited or solved by just one actor or actor group. Diverse actors, skills,
technologies, operating environments, knowledge stocks and approaches will be required. Sector
innovation requires new ‘hardware’, ‘software’ and social arrangement at multiple levels. This is
not to say that continuous involvement by all actors is required. Whilst it is important that all
opinions, voices – particularly those of the poor - and walks of life are included and connected
through a loose network of linkages which can be easily reactivated, involvement at specific times
in different innovation trajectories will have to evolve according to needs and approaches.
Similarly, it does not mean that all ‘solutions’ have to be found at the same time, but that the
outlook has to be sufficiently articulated and complete to allow the brokering of rapid responses
and adaptations to services, products and production arrangements as society’s social, economic
and environmental goals evolve. In the absence of this ‘comprehensiveness’, innovation response
capacity is unlikely to flourish.
Indicative activities that can contribute to and stimulate the creation of such capacities include,
among others, interventions that increase the degree of organization of specific actor groups;
interventions that build capacity for groups to participate in policy and practice processes and
dialogues; regular, broadly shared sector reviews and analyses; the establishment of sector
foresight mechanisms; and policy and financial incentives that stimulate cross-sectoral working
arrangements and the exchange of information.
Coherence
Whilst working together is essential, it requires actors to do so in ways that allows them to actively
identify complementarities and to exploit synergies. It calls for sensitivity on the part of the actors
to recognize hiatus or failings and the identification of the competencies and skills that are lacking.
Networks of interacting elements have properties which are more than the sum of the constituent
51
parts and which cannot be accounted for by analysis of individual elements of the system. It is for
this reason that ways of working assume such significance as this determines how the system
operates. True partnership relationships are based on systemic necessity, rather than on vested
interests, and occur only in situations where learning and reflection are an integral part of
management arrangements and where corrective action as a result of such reflection is immediate
and effective.
Similarly, coordination involves more than just the brokering of the necessary contacts or the
elimination of obstructions and similar task. It also provides urgency, direction, specificity and
identity to a network of actors as it confronts particular issues. It requires actors that command the
sector’s broad trust, the institutional and organisational overview, and skills that can lubricate an
appropriate patchwork of interaction and competencies. In examples where this has functioned
well, such tasks have either been fulfilled by a body set up specifically through the concerted
actions of a critical mass of sector actors, or where such tasks have gravitated organically towards
a broadly recognised sector champion. In the absence of this ‘coherence’ innovation response
capacity is unlikely to flourish
Indicative activities that can contribute to and stimulate the creation of such capacities, among
others, include the organisation of regular, sector-wide consultative meetings, which may require
facilitation initially; the creation of a sector coordination body/association; financing and
establishment of sector brokering agents; policy and financial incentives schemes for sector
collaboration and joint programming; and regular, broadly shared sector reviews and analyses.
Clarity
Information sharing and openness in decision-making greatly aids learning. The associated
exchanges also foster the recognition and integration of both the diverse and shared actor
objectives that the sector needs to satisfy and generate the acknowledgement of specific rights and
responsibilities. The trust created through such interaction is an essential ingredient in the
relationships needed to underpin innovation: Responsiveness to rapidly changing conditions requires
a sufficient degree of transparency among sector actors to ease the creation of new, or the
rekindling of old, links. In the absence of this ‘clarity’, innovation response capacity is unlikely to
flourish.
Indicative activities that can contribute to and stimulate the creation of such capacities include,
among others, sector-wide programming and visioning exercises; negotiation and advocacy skill
acquisition; sector transparency policy and financial incentive schemes; and inter-sectoral
placement and employment schemes.
Creativity
Seizing opportunities, taking risks and solving existing or emergent problems are essential for any
sector to compete, cope and prosper in today’s dynamic rural development scenario. If such
capacities are not well developed, responses to opportunities and challenges will be lacking or, at
best, weak. This role is generally recognised to be best suited to private sector operators.
Nevertheless, when entrepreneurship is not uniformly developed across all sectors, it is important
that such institutional hiatus are filled – even if temporarily - by the public or tertiary sector whilst
remedial action to stimulate private/financial incentive is taken.
52
Foresight mechanisms and close linkages beyond the immediate sector and national boundaries can
assist in the early recognition and action to anticipate social, economic, technical and
environmental sector trends and policies. Such mechanisms are particularly relevant in an
environment were experimentation and sector evolution is actively encouraged.
In the absence of this ‘creativity’, innovation response capacity is unlikely to flourish.
Indicative activities that can contribute to and stimulate the creation of such capacities include,
among others, policy and financial incentives for flexible, embedded and service-oriented research
capacity; creation of foresight processes and mechanisms; policy and financial incentives for
entrepreneurship including social venture capital, and public, private tertiary sector partnerships;
international sector ‘twinning’ and membership of international sector associations; and the
organisation and participation of international sector visits and trade fairs.
53
REFERENCES
Abate, G (2003) Early warning activities in Ethiopian. Early Warning Department, Disaster Prevention &
Preparedness Commission, Ethiopia. At:
www.napa-pana.org/.../presentation%203.%20Early_Warning_Ethiopia.ppt
Desta, S (2001) Cattle Population Dynamics in the Southern Ethiopian Rangelands. The Global Livestock CRSP
Feinstein International Centre (2007) Impact Assessments of Livelihoods-based Drought Interventions in Moyale
and Dire Woredas. A Pastoralist Livelihoods Initiative report produced by the Feinstein International
Center in partnership with: CARE, Save the Children 9USA), USAID Ethiopia
Freeman, C. (1995) The ”National System of Innovation” in Historical Perspective. Cambridge Journal of
Economics, 19, 5-24.
Fakuda-Parr, Sakiko, Carlos Lopes, and Khalid Malik (2002) ‘Institutional innovations for capacity
development’, in S. Fukuda-Parr, C. Lopes, and K. Malik (eds.), Capacity for Development: New
Solutions to Old Problems, London: Earthscan and UNDP.
Hall A.J., , M.V.K. Sivamohan, N. Clark, S. Taylor and G. Bockett. 2002. “Why Research Partnerships Really
Matter: Innovation Theory, Institutional Arrangements and Implications for the Developing New
Technology for the Poor,”World Development, 29:5, pp783-797.
Lundvall, B. –A. (1992). National Systems of Innovation. Towards a Theory of Innovation and Interactive
Learning. London: Pinter Publishers .
Malerba, F. (2002). Sectoral systems of innovation and production. Research Policy 31 (2002) pp 247-264.
NASA (2008) NASA Data Show Some African Drought Linked to Warmer Indian Ocean. Goddard View Newsletter,
May 2008
Pantuliano, S and Wekesa, M (2008) Improving drought response in pastoral regions of Ethiopia.
Somali and Afar regions and Borena Zone in Oromiya Region. Humanitarian Policy Group
Overseas Development Institute, London
Sandford, J., Ashley, S., 2008. Livestock Livelihoods and Institutions in the IGAD Region. IGAD Livestock Policy
Initiative Working Paper 10 2008. At:
http://www.igad-lpi.org/publication/docs/IGADLPI_WP10_08_2.pdf.
World Bank.2006.“Enhancing Agricultural Innovation: How to Go Beyond the Strengthening of Research
Systems.” World Bank, Washington, D.C.
54
APPENDIX: A BRIEF OPERATIONAL GUIDE TO PROCESS MONITORING
Why we need process monitoring
The purpose of many research and development (R&D) projects in the past years have undergone a
clear shift from direct impacts on poor people only, to the facilitation of changes that make
processes operate in more pro-poor ways – thus recognizing that social and economic outcomes are
mediated through the process of institutional change.
Since there is generally a long lag-time before tangible benefits arise that can be assessed, there is
a need to monitor and report changes in process. Thus, lessons that emerge from interventions with
institutional change become a key indicator of progress and a qualitative means to assess the
direction of institutional change. If impact is increasingly about institutional change, as is our main
argument here, one needs to be far more serious about how such change is monitored - and thus
expand the perspective of normal M&E and impact assessments - which grossly underestimate
change because it views this in terms of short-term tangible economic terms, only.
The identification of some typologies of desirable institutional change and institutional change
trajectories -although others may emerge- allows, the monitoring of such trajectories over time
through methodologies such as stages of progress/monitoring domains, socio-economic
benchmarking, episode analyses, thus providing the multiple sources and types of information that
would build up plausible connections between particular types of institutional change and socially
desirable outcomes.
The emphasis of such activities require an action research / action development orientation and
the need to think about progressive change in these processes, where the different progressive
stages need to be defined and redefined throughout the project. The tool of choice in this case is
the stages of progress / monitoring domains approach.
Process monitoring – instructions for use
R&D processes are generally to complex to monitor or analyse in their entirety to obtain any
meaningful and useable results, it is thus proposed to break up the processes that are being
addressed into a number of distinct monitoring domains. In each domain, essential questions are
asked that will need to be revisited as projects and processes evolve.
To achieve this, LINK proposes experimentation with two simple approaches which are explained in
detail below:
The ‘artisanal’ approach
Step: Baseline brainstorm
To initiate this approach, a meeting needs to be organised that brings together partners and other
key stakeholders, to agree on a baseline situation in respect of the R&D process concerned, as it is
conventionally organised. To make such an initial step really effective, we suggest that users of the
approach and their partners facilitate a brainstorm on this. Some of the current state
characteristics that can result from such a brainstorm may, for example, include the following:
(i) Roles of various actors are fixed. Limited scope for new actors to join, be included, or play
new roles;
(ii) Institutional arrangements, nature of relationships and political economy issues not examined
and reported on. No incentives for institutional learning and change;
55
(iii) Definition of the problem(s) and solution(s) to be evaluated, defined by limited set of actors.
Limited consultation and / or participation by other stakeholders;
(iv) etc.
Step: Stages of progress
Following agreement on the current state characteristics, the project/meeting can move to
brainstorm about what the future vision for each of the agreed characteristic of the current policy
process would be. Once agreement has been reached on this vision of the future, intermediate
progressive stages need to be identified and agreed. Next, the critical questions that will have to
be asked to be able to assess progress through its progressive stages from each of the identified
baseline characteristics to its future vision need to be enumerated. The outcome of such a
brainstorming session, in respect of the current state characteristic identified during the baseline
step, may look something like the following examples:
(i) Roles and involvement of actors
- Future vision: Roles and involvement of various actors evolve according to the needs of the
policy process and in accordance with the capacity of actors to perform specific functions
- Progressive change stages e.g.:
Set and fixed Decided in consultation Iterative/evolving -----------
--------------------------------------------------------------Æ
Progressive change
- Indicative questions e.g.: Are ‘new’ actors performing roles traditionally performed by others?
Are different actors performing multiple or new roles? Who? In which specific activities? How do
these changes manifest themselves?
(ii) Nature of relationships and linkages.
- Future vision: Relationships and patterns of interaction among sector actors in the policy
process are efficient, effective and respectful and conducive to sharing of experiences and
institutional change.
- Progressive change stages e.g.:
One dominant member Joint implementation Joint responsibility
-------------------------------------------------------------------------Æ
Progressive change
- Indicative questions e.g.: What are the key relationships in the programme/project? What do
individual partners think about these relationships? Are these relationships and the rules that
govern them evolving? How do such changes manifest themselves?
(iii) Process management and structures by specific activity.
- Future vision: The policy process and related activities are managed by consensus by a broad
range of key stakeholders.
56
Progressive change stages e.g.:
Use of sanctions Powers of veto By consensus
-------------------------------------------------------------------------Æ
Progressive change
- Indicative questions e.g.: How are decisions on different activities reached? Who is consulted?
Who participates?
Step: Assessment and adjustment
The action research orientation that lies at the basis of this process monitoring approach, in
practice means that the results from this continuous assessment drive the adjustment of project
activities and the revision and redefinition of the essential questions as the process evolves and
institutional learning and change proceeds. The approach thus contributes to integrating learning
into livestock sector policy formulation and to devising interventions where the links between
operational activities and R&D processes are explicit and where discovering how best to deal with
new challenges is a common task that connects development interventions and relevant areas of
policy making.
In the setting of a specific programme or project, this assessment and adjustment process can be
facilitated as an central part of regular workshops/meetings, and basically involves repeating the
previously described steps to assess movement on the ’progressive change’ trajectories and to
revisit and adjust, if necessary, the essential questions that are asked to assess process evolution.
The ‘flat-pack’ approach
Prefab process monitoring
As the name implies, this approach makes use of pre-prepared standard monitoring domains that
encapsulate the 4 pillars of the innovation systems analytical framework (World Bank, 2006) and
can be extended for any of the specific objectives of the project or programme. Assessments are
made through the use of standard critical questions for each domain, and a standardized
progressive change barometer which allows process evolution in each domain to be assessed. The
proposed monitoring domains are as follows:
• Role of different actors
• Patterns of interaction
• Quality of interactions
• Enabling environment
• Poverty relevance
The power of process monitoring lies, as mentioned earlier, in the promotion of iterative learning
and subsequent change among diverse stakeholders and the explicit linking of R&D processes to
operational activities and impact. Whilst the attached prefab sheets should thus be used in that
manner, they can, in principle, also be used in smaller groups or by individuals. If used in that way,
the assessment should still form the basis for interaction among different stakeholders during
workshops, or serve as a focus for discussions in programme related dialogues.
57
Instructions for use
Step: Answering critical questions per domain
Each domain sheet contains three critical questions to be ‘answered’. Following the careful
evaluation of each of these questions, the domain’s progressive change barometer allows the
current ‘situation’ in each of the domain to be indicated. At the same time, important comments
and observations can be noted down on the sheets as well [and additional paper if required...]. As
described in Approach A, under Step: Assessment and adjustment, the use of this approach should
ideally be facilitated as an integral part of the programme’s / project’s regular workshops and
meetings, - and provide the basis from which to adjust and devise project activities- and basically
involves a repeat of the described steps to assess movement on the ’progressive change’
trajectories and to revisit and adjust, if necessary, the essential questions that are asked to assess
process evolution.
The following page shows a brief example of ‘used’ prefab process monitoring sheets. The pages
thereafter provide potential users with ‘clean’ prefab sheets for their own use:
58
Domain: Role of different actors
Date: 17 January 2009
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Actors’ roles are fixed and not challenged
x
New and diverse actors become involved as needs dictate
x
Process is actively coordinated and facilitated
x
The progressive change barometer
x
set and fixed decided in consultation iterative/evolving
Comments/observations: The key relationships in the project are largely ‘old’ relationships and there is little evolution in rules and practices. Most project partners are
resigned to their tasks in light of history rather than in light of needs and current capacity. Erosion of influence and introduction of new partners is strongly resisted by
some key stakeholders. Partners not treated or regarded as equals but overpowering influence by public service officials.
59
Date: 9 March 2009
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Actors’ roles are fixed and not challenged
x
New and diverse actors become involved as needs dictate
x
Process is actively coordinated and facilitated
x
The progressive change barometer
X
Comments/observations: Although key relationships in the project are still largely ‘old’ relationships and there has been some change, as new actors have been brought
into the process. Notwithstanding these changes, most project partners are resigned to their tasks in light of history rather than in light of needs and current capacity.
Although there was an initial backlash from the traditional partners, there appears to be a slow recognition of the value of different knowledge stocks.
60
Domain: Role of different actors
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Actors’ roles are fixed and not challenged
New and diverse actors become involved as needs dictate
Process is actively coordinated and facilitated
The progressive change barometer
set and fixed decided in consultation iterative/evolving
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Actors’ roles are fixed and not challenged
New and diverse actors become involved as needs dictate
Process is actively coordinated and facilitated
The progressive change barometer
Comments/observations:
61
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Actors’ roles are fixed and not challenged
New and diverse actors become involved as needs dictate
Process is actively coordinated and facilitated
The progressive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Actors’ roles are fixed and not challenged
New and diverse actors become involved as needs dictate
Process is actively coordinated and facilitated
The progressive change barometer
Comments/observations:
62
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Actors’ roles are fixed and not challenged
New and diverse actors become involved as needs dictate
Process is actively coordinated and facilitated
The progressive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Actors’ roles are fixed and not challenged
New and diverse actors become involved as needs dictate
Process is actively coordinated and facilitated
The progressive change barometer
Comments/observations:
63
Domain: Patterns of interaction
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions between a wide range of actors are frequent and efficient
Linkages among actors are transparent, dynamic and evolving
Responsibility for R&D process shared by wide range of actors
The progressive change barometer
Dominant member(s) joint implementation joint responsibility
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions between a wide range of actors are frequent and efficient
Linkages among actors are transparent, respectful and evolving
Responsibility for R&D process shared by wide range of actors
The progressive change barometer
Comments/observations:
64
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions between a wide range of actors are frequent and efficient
Linkages among actors are transparent, respectful and evolving
Responsibility for R&D process shared by wide range of actors
The progressive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions between a wide range of actors are frequent and efficient
Linkages among actors are transparent, respectful and evolving
Responsibility for R&D process shared by wide range of actors
The progressive change barometer
Comments/observations:
65
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions between a wide range of actors are frequent and efficient
Linkages among actors are transparent, respectful and evolving
Responsibility for R&D process shared by wide range of actors
The progressive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Existing mechanisms facilitate frequent interaction among a wide range of actors
Linkages among actors are transparent, respectful and evolving
Responsibility for R&D process shared by wide range of actors
The progressive change barometer
Comments/observations:
66
Domain: Quality of interactions
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions among actors are efficient and conducive to learning
Processes draw on a wide variety of knowledge stocks
Existing culture promotes sharing of experiences among actors
The progressive change barometer
use of sanctions powers of veto consensus
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions among actors are efficient and conducive to learning
Processes draw on a wide variety of knowledge stocks
Existing culture promotes sharing of experiences among actors
The progressive change barometer
Comments/observations:
67
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions among actors are efficient and conducive to learning
Processes draw on a wide variety of knowledge stocks
Existing culture promotes sharing of experiences among actors
The prog
r
essive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions among actors are efficient and conducive to learning
Processes draw on a wide variety of knowledge stocks
Existing culture promotes sharing of experiences among actors
The progressive change barometer
Comments/observations:
68
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions among actors are efficient and conducive to learning
Processes draw on a wide variety of knowledge stocks
Existing culture promotes sharing of experiences among actors
The progressive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Interactions among actors are efficient and conducive to learning
Processes draw on a wide variety of knowledge stocks
Existing culture promotes sharing of experiences among actors
The progressive change barometer
Comments/observations:
69
Domain: Enabling environment
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
St
r
ongly agree
Competencies and skills of sector actors are conducive to interaction
Sector governance fosters participation, experimentation and learning
Incentives that promote dialogues among sector actors are actively pursued
The progressive change barometer
negative environment neutral environment positive environment
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Competencies and skills of sector actors are conducive to interaction
Sector governance fosters participation, experimentation and learning
Incentives that promote dialogues among sector actors are actively pursued
The progressive change barometer
Comments/observations:
70
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Competencies and skills of sector actors are conducive to interaction
Sector governance fosters participation, experimentation and learning
Incentives that promote dialogues among sector actors are actively pursued
The progressive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Competencies and skills of sector actors are conducive to interaction
Sector governance fosters participation, experimentation and learning
Incentives that promote dialogues among sector actors are actively pursued
The progressive change barometer
Comments/observations:
71
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Competencies and skills of sector actors are conducive to interaction
Sector governance fosters participation, experimentation and learning
Incentives that promote dialogues among sector actors are actively pursued
The progressive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
Competencies and skills of sector actors are conducive to
interaction
Sector governance fosters participation, experimentation and learning
Incentives that promote dialogues among sector actors are actively pursued
The progressive change barometer
Comments/observations:
72
Domain: Poverty relevance
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
R&D processes ensure inclusion of the poor
Poor (or their representatives) are actively involved in the processes
Relevance of process to the poor actively assessed by different sources
The progressive change barometer
output targets consulted stakeholders coalition partners
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
R&D processes ensure inclusion of the poor
Poor (or their representatives) are actively involved in the processes
Relevance of process to the poor actively assessed by different sources
The progressive change barometer
Comments/observations:
73
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
R&D processes ensure inclusion of the poor
Poor (or their representatives) are actively involved in the processes
Relevance of process to the poor actively assessed by different sources
The progressive change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
R&D processes ensure inclusion of the poor
Poor (or their representatives) are actively involved in the processes
Relevance of process to the poor actively assessed by different sources
The progressive change barometer
Comments/observations:
74
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
R&D processes ensure inclusion of the poor
Poor (or their representatives) are actively involved in the processes
Relevance of process to the poor actively assessed by different sources
The progressive
change barometer
Comments/observations:
Date:
Critical questions Strongly disagree
Disagree
Neutral
Agree
Strongly agree
R&D processes ensure inclusion of the poor
Poor (or their representatives) are actively involved in the processes
Relevance of process to the poor actively assessed by different sources
The progressive change barometer
