SIPED_Final_Evaluation.pdf

April 2013

This publication was produced at the request of the United States Agency for International Development by

International Business and Technical Consultants, Inc. (IBTCI).

EVALUATION

Strengthening Institutions for Peace and Development Program in Ethiopia: “Unfinished Business”

This report was prepared by:

Simon Richards Kassaye Hadgu Kelemework Tafere

Front Photo - Camels approaching Liben – Filtu Water system at Siminto – taken by Simon Richards (All photos in this report taken by the Lead Evaluator, Simon Richards)

Evaluation: Strengthening Institutions for Peace and Development Program in Ethiopia:

“Unfinished Business”

April 30, 2013

This publication was produced for review by the United States Agency for International Development. It was prepared under Contract Number AID-RAN-I-00-09-00016, Task Order Number AID-623-TO-13- 00003 by International Business & Technical Consultants, Inc. (IBTCI)

DISCLAIMER

The authors’ views expressed in this publication do not necessarily reflect the views of the United States Agency for International Development or the United States Government.

CONTENTS

Acronyms ....................................................................................................................................................... i

Executive Summary ....................................................................................................................................... i

Evaluation Purpose & Evaluation Questions

Evaluation Purpose

Evaluation Questions

Project Background

Evaluation Methods & Limitations

Methods

Limitations

Findings, Conclusions & Recommendations

Key Contextual Challenges for the Project (and evaluation):

Findings

Best Practices and Lessons learned

Conclusions

Recommendations

Annexes

Annex I: Evaluation Statement of Work

Annex II: Data Collection Instruments

Annex III: Sources of Information

Annex VI: Disclosure of any Conflicts of Interest

Annex VI: Proposed Approach for South Cluster NRM activities

ACRONYMS

CHSA Charities and Societies Agency (introduced to differentiate it from another acronym) CoP Chief of Party CPR Conflict Prevention and Resolution CPS Civil Peace Service CSA Charities and Societies Agency CS Conflict Sensitivity DCoP Deputy Chief of party DEC Development Experience Clearinghouse DfID Department for International Development (British) DNH Do No Harm EPRDF Ethiopian People’s Revolutionary Democratic Front EW Early Warning FY Fiscal Year GIZ Deutsche Gesellschaft fur Internationale Zuzammanarbeit GoE Government of Ethiopia IGR Inter-governmental Relations MoFA Ministry of Federal Affairs NGO Non-Government Organisation OBAS Oromia Bureau for Security and Administration OPA Oromia Pastoralist Association RCCSGA Resource Center for Civil Society Groups Association SAB Security and Administration Bureau SIPED Strengthening Institutions for Peace and Development SNNPR Southern Nations Nationalities and Peoples’ Region SOW Statement of Work USAID U.S Agency for International Development i

EXECUTIVE SUMMARY

EVALUATION PURPOSE AND EVALUATION QUESTIONS

International Business and Technical Consultants, Inc. (IBTCI) was contracted to perform the final performance evaluation of the Strengthening Institutions for Peace and Development (SIPED) program in Ethiopia. This evaluation took place in February to April 2013 following the closure of the SIPED program in September 2012. The evaluation is intended primarily to:

• Assess strengths and weaknesses of program activities and approaches

• Identify ‘Lessons Learned’ and best practices

• Provide specific recommendations on how these might be incorporated into future Democracy and

Governance program design and implementation and to

• Gauge the effectiveness of Monitoring and Evaluation systems and performance indicators that effectively capture progress and results

The evaluation questions outlined in the Statement of Work (SOW) are based on consideration of four main areas:

1. Design: Assessing the effectiveness of the design, appropriateness of objectives and M&E systems in addressing the conflict risks of the project areas and capturing change.

2. Stakeholders: Did program activities target the most appropriate stakeholder groups such as youth, women and civil society with the right balance and support to, traditional community and local government structures and institutions linking them to the federal level?

3. Effectiveness: Are there examples of “best practice” and lessons learned emerging? Are there missing gaps and if so how might these be addressed? What has been the effect of introducing Do No Harm (DNH) principles and practices in to program activity planning and implementation and are there recommendations on how to utilize them more effectively?

4. Sustainability: Did the programs support and strengthen structures and processes which reinforce peaceful coexistence and decrease the likelihood of violent conflict? Are local peace structures interacting with all levels of government and functioning effectively? Are these structures likely to remain in place and effective after the conclusion of the program?

PROJECT BACKGROUND

Strengthening Institutions for Peace and Development (SIPED) in Ethiopia was a project initiated in October 2009 as a three year cost-extension of the 5-year Conflict Prevention and Resolution (CPR) Project funded by U.S Agency for International Development (USAID) beginning in December 2004, with the overarching program to ‘Reduce inter-community tensions and create an environment for sustainable peace’. In October 2011, SIPED also received two additional and complementary cost extensions to develop inter-community water development initiatives and youth programming specifically in the Somali Region. The program ended in September 2012. While the legislative context has changed and direct peace-building interventions conducted by NGOs are no longer permissible, nevertheless the program has journeyed a long way from its inception. The program goal has been addressed through four initial Peace objectives: (1) Building institutional capacity and partnerships to prevent, manage and respond to conflicts, and promote peace at local, regional, state and national levels (2) To improve the understanding and application of “Do No Harm” principles at all levels; (3) To support initiatives promoting improved collaboration, peace and equitable development across state boundaries and in border areas and (4) to reduce tensions and violence by supporting peace initiatives ii that respond to conflict situations and address root causes of conflict through development interventions.

Ethiopia is situated in the volatile and unstable Horn of Africa where it has been affected by a long history of regional conflict as well as internal political struggles. The SIPED project was designed to address the prevalence of internal inter-communal violence, primarily due to a paucity of natural resources coupled with their poor or weak governance. In recent years there has also been an increase in ethnic based conflict particularly around borders as a consequence of the 1995 Ethno-federalist constitution which has inadvertently encouraged groups to try and establish identity- based administrative and governance units as well as secure access to and ‘ownership’ of scarce resources including land, water and other natural resources. The Ethiopian government has increasingly acknowledged its own issues of capacity in the field of conflict management and peace-building and the USAID supported SIPED program has been viewed as a welcome support in addressing this need.

EVALUATION QUESTIONS, DESIGN, METHODS AND LIMITATIONS

The evaluation employed a mixed methodologies approach designed to collect data that can inform a rich response to each of the evaluation questions. Four main methods were used;

Firstly, a review of both project literature as well as academic literature in the public domain providing background analysis on the project as well as the context over the past ten years or so.

Secondly, some limited analysis of quantitative data using both project monitoring and evaluation data, where available.

Thirdly, analysis of qualitative data obtained through focus group discussions with stakeholders and groups from specific demographic sections of the community (i.e., women, youth and elders) involved in the project (from all sides of any particular conflict), supplemented by participatory methods where appropriate and key informant interviews (for example key project staff personnel as well as external experts) using semi-structured interviews.

Fourthly, the team undertook field site visits and directly observe project infrastructure including water systems and natural resource management sites to ascertain the technical quality and appropriateness of the structures as well the degree to which community agreements are being upheld (or not).

There have been a number of limitations affecting the evaluation design, the collection of data and the quality of data obtained and therefore potentially the findings.

The evaluation design and management has been constrained by the sensitivity of the project as well as the registration status of the implementing partner Mercy Corps. Recent changes to the legislation affecting NGOs, has meant that direct NGO engagement in the peace-building, conflict sector is much more complex. The government perspective on what is considered the legitimate role of NGOs now demands a different approach, focusing on fundamental development and utilising the language of community development around service delivery and shared resources. The heightened sensitivity of all stakeholders to this changed NGO role exacerbated Mercy Corps’ desire to avoid drawing unwanted attention to the evaluation process as this was considered to indirectly highlight the peace-building work that they have been engaged in at a critical time in the relationship between government and Mercy Corps.

iii

The collection of accurate, objective, triangulated data collection in conflict programming is always challenging and has been exacerbated by time constraints, the difficulties of accessing program participants in remote areas and most importantly issues of trust and frankness from respondents. The quality of the data has been influenced by a number of factors that potentially present them in a more positive light than perhaps is the reality. These include the limited opportunity for triangulation of self-reported change (particularly for government), and the Hawthorn effect in a resource poor environment where participants attempt to please the interviewer perhaps in the hope of being able to continue accessing resources or influencing the allocation of development resources.

FINDINGS AND CONCLUSIONS

In conclusion the SIPED program evolved positively and travelled a substantial ‘distance’ from its starting point. It made a considerable contribution to conflict management and peace in some of the clusters being tackled (those visited) and evidence from stakeholders across all sides of conflict systems suggest that it has achieved significant impact, both economic and social, at the community level. There is an expression from all stakeholders though, including government, that the peace work of the project has ‘unfinished business’ and needs further support to increase the opportunity for sustainable peace in the project areas. The project has also developed a number of ‘best practice’ programming features that are worth scaling up carefully and that shift the emphasis from peace-building interventions concentrating on dialogue to development oriented interventions that consolidate and reinforce peace.

These include inter-community resource sharing approaches and models that often span administrative borders as well as conflict lines. While it is early to judge impact, sound, iterative approaches and processes founded on strong, peace-building principles appear to have been successful to date in the development of mixed ethnic and cross-border committees and mechanisms to manage potential conflicts over and use of shared community resources. To date these initiatives have focused mainly on water and natural resource management. The project has fostered increasing trust and confidence between stakeholders, as well as increased horizontal communication and interactions across borders described by one key informant as “...effective state-building”. This has been at both the individual and institutional level through formal government, traditional, semi-formal institutions such as peace committees, as well as newly developed mechanisms such as the management committees for shared resource use. Similarly capacity to manage conflict and build peace has been enhanced at all levels as evidenced by changed attitudes and behaviors that have shifted from cycles of violence to peace. The sustainability of the changes though, is uncertain with some indicators pointing towards reasonable sustainability and others suggesting progress is fragile without additional support.

EVALUATION PURPOSE & EVALUATION

QUESTIONS

EVALUATION PURPOSE

International Business and Technical Consultants, Inc. (IBTCI) was contracted to perform a performance evaluation of the Strengthening Institutions for Peace and Development (SIPED) program in Ethiopia.

This evaluation took place in February to April 2013 following the scheduled conclusion SIPED program in September 2012. SIPED in Ethiopia was a project initiated in October 2009 as a three year cost-extension of the 5-year Conflict Prevention and Resolution (CPR) Project funded by USAID beginning in December 2004. In October 2011, SIPED also received two additional and complementary cost extensions to develop inter-community water development initiatives and some youth programming specific to the Somali Region. The program ended in September 2012. While the legislative context has changed and direct peace-building interventions conducted by NGOs are no longer permissible the program has progressed a long way from its inception and therefore the purpose of the evaluation of the SIPED Evaluation is to:

• Assess strengths and weaknesses of program activities and approaches

• Identify ‘Lessons Learned’ and best practices

• Provide specific recommendations on how these might be incorporated into future Democracy and

Governance program design and implementation and to

• Gauge the effectiveness of Monitoring and Evaluation systems and performance indicators that effectively capture progress and results

EVALUATION QUESTIONS

The evaluation questions as described in the original scope of work (Annex I) are listed below:

1. Design

a. Assess the effectiveness of the design (strengths and weaknesses) and discuss the appropriateness of the objectives in addressing the conflict risks of the project areas.

b. Were Monitoring and Evaluation systems appropriate? Do indicators capture outcome and impact in addition to input and output? Did they effectively capture progress and results? How could monitoring and evaluation systems be improved for future programming interventions?

2. Stakeholders

a. Did program activities target the right partners and stakeholders groups?

b. Did activities strike the right balance between and provide support to, traditional community and local government structures and institutions and linked to federal actors and institutions?

c. How effectively has civil society been involved in peace building processes?

d. How effectively have youth and women been involved?

3. Effectiveness

a. Are there any examples of success stories, “best practice” and lessons learned emerging that can be documented?

b. What missing gaps exist and how might these be addressed

c. What has been the effect of introducing Do No Harm principles and practices in to program activity planning and implementation?

d. What are the most salient recommendations for how to most effectively utilize Do No Harm principles and practices?

4. Sustainability

a. Did the programs work with, support and or strengthen those structures and processes which reinforce the prospects for peaceful coexistence and decrease the likelihood of the outbreak, reoccurrence or continuation of violent conflict?

b. Are these structures/mechanisms functioning in an effective manner? If ye, why and if no, why not?

c. Are these structures likely to remain in place and effective after the conclusion of the program?

If yes, why and if no, why not?

d. How and to what degree are local peace structures interacting with local government, regional government and federal government actors?

PROJECT BACKGROUND

Ethiopia is located in the heart of a historically volatile Horn of Africa. Ethiopia has been inundated by internal conflict and wars with neighbors for much of its history. In the past 30 years this has taken the form of a 17 year-long civil war that brought the current Ethiopian People's Revolutionary Democratic Front (EPRDF) government to power in 1991, inter-state wars with Somalia in the 70s and with Eritrea in the 90s. Internal violence is primarily inter communal and is predominantly due to limited or inappropriately governed natural resources (land, water, forests etc...) as well as weak or nascent governance and rule of law institutions. In addition, over the past several years, a marked rise in inter-ethnic conflict has occurred, particularly in Southern Nations, Nationalities and Peoples (SNNP) Regional State (RS) and along the common border of Oromia and Somali regional states, as groups sought to establish identity-based administrative units and thus gain control of state, economic and natural resources under the 1995 ethno-federalist constitution. In 2005, disputes over the results of the National Elections exploded in two bouts of street violence. Now extremist influences, worrisome signs of increasing inter-religious tension, continued fighting in Somalia and increased insurgent and counter-insurgent operations in Somali RS threaten to increase conflict and instability in Ethiopia.

The Government of Ethiopia (GoE) recognizes that violent conflict and fear of renewed violence or lack of safety constrains development, undermines efforts towards poverty reduction and weakens the social fabric along with physical, economic and human capital. Over the past several years, the Government of Ethiopia (GoE) has come to acknowledge the challenges associated with its own capacity in managing local conflict and has welcomed international donor supported conflict management activities.

The overarching SIPED goal was to ‘Reduce inter-community tensions and create an environment for sustainable peace’ through the following four Peace objectives:

1. Build institutional capacity and collaborative partnerships to prevent, manage and respond to conflicts, and promote peace at local, regional, state and national levels;

2. Improve understanding and application of conflict sensitive approaches and “Do No Harm” principles at the national, regional and local levels;

3. Support initiatives that promote improved relationships and collaboration across regional state boundaries, to enable improved peace and equitable development in border areas; and

4. Reduce local tensions and violence through support to initiatives which build peace and respond to conflict situations and which address root causes of conflict through developmental and livelihood interventions.

A further set of objectives was added with the incremental youth and water funding in October 2011.

Objectives for the Water component

1. Secure long-term, sustainable access to potable water in areas where water shortage results or exacerbates violent conflict.

2. Reduce pejorative consequences of extreme water shortage, especially violent conflict, spread of water borne diseases and increased burden on women.

3. Strengthen relationship among communities who have experienced violent conflict through increased interaction and co-management around newly constructed and rehabilitated water points.

Objectives for the Youth component

1. Contribute to the creation of economic opportunities for peri urban youth through the provision of education and livelihood support;

2. Improve the social opportunities for disadvantaged youth resulting in strengthened relationships and positive engagements within youth groups and their communities;

3. Support educational, business and youth institutions to assist and mentor youth, and

4. Conflict sensitivity and Do No Harm principles incorporated into assessment, planning, implementation and monitoring protocols and activities

The project has applied three main strategies1 to achieve these objectives through holistic integrated programming linking peace and livelihoods programming:

1. Strengthen institutions: Working with a range of existing institutions within government, and community to strengthen abilities to address difference and tension by peaceful means. Where appropriate it will also support the establishment of new institutions. Support will provide new skills and knowledge as well as opportunities to model approaches to address tensions and conflict.

2. Support cultures of peace: The SIPED program also seeks to work directly with stakeholders within local communities, to promote peaceful coexistence. This work will build on the cultural values within societies such as tolerance and cooperation, it will also encourage communication and dialogue to deal with change, and provide new ideas and knowledge to younger generations to help develop more positive and constructive attitudes and behaviors

3. Address multiple drivers of conflict: Mercy Corps also recognizes the need to tackle the multiple underlying causes of tension to ensure sustained peace. This requires a more holistic, integrated style of programming blending peace building with development interventions, and ensuring conflict sensitivity. It includes the design of processes which enable communities to identify issues of mutual

These strategies have been adapted from the SIPED Technical Narrative cost extension proposal.

concern and benefit, and ensures activities are conducted in ways which promote coexistence and collaboration.

The project’s theories of change that inform the programming are as follows:

Change in individual attitudes and behaviors requires more than just learning but also the opportunity to practice, and experience new knowledge and skills.

Capacity building in institutions requires both learning by individuals within the institution as well as support to new processes and systems within the institution.

Effective promotion of peaceful resolution of conflicts requires both changes to attitudes and behaviors, as well as credible, legitimate institutions or systems to support dialogue and communication.

Regular, sustained, proactive communication can have significant positive impacts on relationships between conflicting groups.

Combined approaches which address relationships, attitudes and behaviors as well as the economic, social and political drivers of conflict will help to ensure sustainable improvement in peace and security.

Linking traditional and ‘formal’ institutions and processes is necessary to address the complex and interlinked drivers of conflict.

The project areas and conflict clusters under the program are illustrated in the map below:

SIPED Project Areas and Conflict Clusters

EVALUATION METHODS & LIMITATIONS

METHODS

The evaluation employed a mixed methodologies approach designed to collect data that can inform a rich response to each of the evaluation questions. Four main methods were used:

1. A review of both project literature as well as academic literature in the public domain providing background analysis on the project as well as the context over the past ten years or so.

2. Analysis of quantitative data using both project monitoring and evaluation data, where available.

3. Analysis of the qualitative data obtained through focus group discussions with community groups involved in the project (from all sides of any particular conflict), supplemented by participatory methods where appropriate and key informant interviews (for example key project staff personnel as well as external experts) using semi-structured interviews.

4. Field site visits and direct observation of project infrastructure including water systems and natural resource management sites to ascertain the technical quality and appropriateness of the structures as well the degree to which community agreements are being upheld (or not).

The final set of data collected in the above methods consisted of:

23 government official interviews at Federal, Region, Zone and Woreda levels in 19 offices

6 individual in-depth senior project staff interviews and two field staff group discussions

20 community focus groups discussions (Total of 186 people)

11 Mercy Corps Staff members (not including senior staff)

3 Senior management (Chief of Party (COP), Deputy Chief of Party (DCoP), Project Manager)

8 key informant interviews (other stakeholders including 3 donors)

LIMITATIONS

Limitations concerning sampling and choice of locations visited

There were a number of limitations to the data collection; trade-offs and constraints affecting the way in which the data was collected. The duration of the evaluation as per the SOW provided for a condensed time frame resulting in the decision that only two of the major project areas could be visited and to omit a visit to the Somali region around Dire Dawa, Shinele and Gode. The decisions were made in consultation with USAID, and were based on the following pragmatic reasons: primarily to ensure that visits were made to geographic areas where a wide array of activities had been implemented for a significant period of time. Secondarily, geographically visiting all of the major areas would have added a significant amount of travel on to the itinerary or required splitting into two teams. The latter option was considered but rejected as it would also have meant that the number of consultations undertaken by the two teams individually would have been limited. Effectively, with USAID’s concurrence, the evaluation team decided that to achieve the objective of establishing lessons learned and investigating best practices from the project, a smaller geographic area explored in greater depth was a better option than a broader area considered more ‘thinly’.

Within the very large geographic area covered, (despite cutting the sample area by a third), the Evaluation team used a purposive sampling method to choose the sites that best matched the major programmatic themes being investigated and that would be most likely to yield lessons’ learned and best practice. The type of sampling was also adapted to better suit the situation and constraints. For instance, one of the initial intended approaches considered was to attempt a household survey across the project beneficiaries to obtain a random sample of those affected, while simultaneously trying to obtain a balance between those who benefited from the project and those who were not direct beneficiaries. In practice this was impractical given the constraining time, security and political parameters. Instead small groups from a purposive sample based on demographics of program participants were chosen (i.e.

women, youth, elders, etc.) for semi-structured, in-depth interviews.

Limitations concerning potential data quality

There are also a number of factors that have potentially shaped the type of data obtained through the sampling methodologies:

Conflict environments - When collecting data in conflict zones and about conflicts themselves, data may not be accurate. It is expected that experienced interviewers will pick up trends and will be able to interpret information provided but even with strong triangulation from many different viewpoints the following elements need to be accepted:

Information provided will be influenced by the informant’s perspective and where they sit in the conflict.

On the one hand information will be deliberately shaped in order to favor interpretations to their advantage and to elicit the listeners’ empathy; for instance, numbers of victims, casualties or lost and stolen live-stock will be exaggerated significantly in their favor. On the other hand, while not necessarily deliberate, reporting will also be colored by the communal ‘clan narrative’ through this filter and lens.

This means that objective reporting is the exception. This effect is exacerbated in oppressive environments, where information is at a premium and politicized. People may be vulnerable if they report openly and this information deviates or contradicts the formal government line. Generally this is not such an issue in pastoralist conflicts unless there is an overt political dimension.

Government - Government perspectives on conflicts often have multiple facets that shape informants’ ability to talk about issues openly. This depends on a number of factors; for instance, how sensitive the conflict is, how politicized, to what extent the government itself may be complicit or involved in it, whether it is acknowledged that conflict is considered a government failure, whether it is acknowledged that other actors may be helpful in resolving issues (i.e. whether or not government feels able to accept assistance from others), and so on. In the case of Ethiopia, all of these factors are at play, their strength depending on the particular locality and context as well as the individual office-holder’s interpretation of the situation.

Institutional culture - Secondly government agencies, particularly those associated with security often have an institutional culture of secrecy, and suspicion, which means that it is often hard for them to share information and discuss issues openly.

‘Hawthorn Effect’ - In conjunction with the above points, there are also milder issues associated with the ‘Hawthorn Effect’. This is where respondents’ behavior and responses during the course of the exercise as a result of the social situation and context itself. In this case a person being interviewed is more likely to say something positive about the program to please the interviewer or because they think that the interviewer wants to hear this. In resource-poor environments especially, this can be the case where respondents might consider that if they report in a certain way this may help in continuing to be able to access resources (e.g., by reporting the situation as worse than it is - a typical scenario in the Somali region of Ethiopia).

Limitations of the overall scope and design of the evaluation

Evaluations of conflict mitigation programs are notoriously difficult. This is for a number of reasons and these should be kept in mind when considering the findings to date.

The multi-causal and multi-stakeholder nature of peace-building and conflict work make both programmatic collection of data difficult during the project as well as creating difficulties with degrees of attribution during the evaluation itself. Each set of conflict dynamics is unique and will have differing degrees of ‘difficulty’ in resolving them depending on a large variety of variables and factors (see discussion later).

While this assignment was a performance evaluation, nevertheless, the non-linear nature of complex social change means that an evaluation will provide a snapshot at the particular time of undertaking, but still may not necessarily adequately reflect the impact of the program as this may emerge (both positively or negatively) later. Indeed stakeholder perceptions of the impact of program activities may also fluctuate depending on when they are questioned or when they are reflecting on the situation. For instance, opinions immediately after a community peace conference will probably be different if canvassed six months later (which may depend on how many follow-up activities or commitments have been fulfilled), which in turn may be different a year later as well in the light of further evolution. This is because the importance and role that an event may play, changes with hindsight, time and reflection and indeed how the person’s perspective and attitudes may have changed during this period. In addition, the timing of outside factors may skew data results (e.g., an annual survey which happens to be scheduled just after a drought or during the dry season may produce results not necessarily indicative of program success and progress as there are ‘masking’ factors).

As a result of these perspectives, the evaluation methodology was more strongly shaped to emphasize a qualitative approach aimed at lesson learning, on the ground. This was to ensure that USG captures the threads of nuance in change that deserve to be aired if articulated by participants and observers and which risk being lost in the search for more rigor given the way that the project has evolved and its final form.

FINDINGS, CONCLUSIONS &

RECOMMENDATIONS

KEY CONTEXTUAL CHALLENGES FOR THE PROJECT (AND EVALUATION):

As noted in the project background above, many of the conflicts encountered in the project areas are a reflection of the current ‘popular’ interpretation and understanding of Ethnic Federalism in Ethiopia at the moment. There is a disturbing trend towards dividing up the southern regions of the country into districts under the premise that each ethnic group should have its own geographically defined area and be self-governing. This is potentially divisive as competition for land, resources and political representation becomes intense and exclusive in nature whereby at its worst groups perceive that they ‘own’ an area or resource and that others should be cast out. While it is clear that large tracts of land might well be better administered with an increase in the number of Woredas, thus bringing government closer to people, nevertheless the proliferation of division is creating new boundaries and new conflicts.

For instance Liben zone in the Somali region has now been divided into six districts or Woredas which effectively has created three ‘Garre’ Woredas, two ‘Degodia’ Woredas, and one ‘Marehan’ Woreda. This is an area that was broadly accepted to be Degodia land. The consequence of this is that the Degodia now feel that they are being pushed off and marginalized with the Garre taking the upper hand. This further increases the sensitivities associated with where boundaries are placed, always a sensitive subject, as where resources are situated with respect to the boundary ‘defines’ ownership. It is within this context that the Mercy Corps project has been operating. In fact, one key informant described this project as being the only “State-building project addressing how Ethnic Federalism is manifested” because it is assisting people consider how they share resources in the classic ‘win-win’ approach pioneered by Roger Fisher and William Ury.2

On occasion this form and manifestation of ethnic federalism becomes further politicized at the federal level where States are jockeying for political positions and power. Regional and national dynamics also have an influence at this level where politics and concerns regarding relationships with surrounding countries such as Somalia, Djibouti, Kenya, South Sudan and Eritrea all play a role in defining the sensitivities and involvement of the federal government in issues being played out at the local level. This can mean that traditional grass-roots peace-building or conflict management may not be as effective due to the effect of other interests that influence the behaviors. It also means that a project might ‘touch’ an issue that is seemingly local in nature and find that it is intractable due to the influence and involvement of regional dynamics.

The volatile nature of politics in conflict areas can also play a role in many small ways that have a disproportionate impact on the ground such as in the turnover of government staff and the way that office holders behave as a consequence. For instance this high turnover may be reflected in a variety of personal behaviors ranging from risk-averse behaviors from staff who do not want to find themselves in a worse posting. So rather than do something ‘wrong’ or take an initiative that may not result in positive impact (or perceived positive impact by one’s seniors – as perception is the key feature) a person does nothing that might jeopardize their short tenure. Equally this vulnerability reinforces a culture where the individual relationship is key. If you want something done or an issue resolved then better to rely on friends where reciprocity and trust are more certain rather than on institutions where change and uncertainty are constant. This further encourages short–term thinking and emphasizes the loss of institutional memories. These features are particularly relevant in conflict programming because it is possible on the one hand to get things achieved through personal relationships and if there are office holders with political will (and authority); on the other hand, the longer term goals of institutional capacity building and long-term societal change are vulnerable and tenuous.

A further difficulty encountered by the project has been the contradictory messages and views of the government towards peace-building. On one hand the government at every level from the Federal down to the Kebele has been extremely supportive and appreciative of the work that Mercy Corps has been

Fisher, Roger and William Ury. Getting to YES: Negotiating Agreement Without Giving In. Penguin Books. New

York, New York, 1981.

undertaking. On the other the hand, the new Charities and Societies Agency (CHSA) legislation has effectively outlawed any work in this sector from being undertaken by NGOs. Unfortunately this valuable work has been a victim of the political environment where the climate and space for civil society activities in Ethiopia has been severely and deliberately constrained, possibly to prevent any opportunities for potential political opposition activity.

Under the new law governing charities and civil society organizations, these entities must seek re-registration every three years. Mercy Corps has obliged but it has been a protracted process and they have apparently been under a special investigation likely to determine whether their peace work has contravened the provisions of the legislation. This has meant that Mercy Corps has been had to provide a large amount of additional information (on top of what they already provide for registration) and has also induced a great degree of anxiety among staff, given the risks associated with potentially having to halt the large portfolio Mercy Corps implements in Ethiopia. As a result of this situation and the interpretation that Mercy Corps brings to bear on the issue, Mercy Corps has been sensitive and cautious vis-à-vis the attention this evaluation might bring to their peace work, which they are deliberately trying to down play at this moment in time.

It is also worth reiterating the following truths when working on conflict in complex political environments. Conflict is multi-causal and multidimensional and each conflict and context is unique and demands a different rate of progress, different amounts of time for processes, and possibly alternative approaches albeit with fundamental principles as a foundation. This is very demanding on program staff and creates intense pressure on a project. As noted, social change is not linear in nature which means at its most simplistic interpretation one cannot apply a cookie-cutter approach to the project and assume that a process that is effective in one region can immediately be applied in another with similar success.

This also applies to time and unfortunately conflict and peace work does not fit strict project timelines.

The SIPED project has applied a classic iterative approach that focuses on process rather than simple outputs. This means that the process itself is a fundamental component of development and has the dual nature of being an end in itself as well as proposing to be a transformational process that both models and reinforces the behaviors that the program is seeking to instill in partners (citizens and government).

The project has also evolved significantly both as a consequence of the changes in the operating environment but also as part of a natural progression towards consolidating peace as stability and initial inroads are made.

The simple diagram below encapsulates the spectrum of approaches that can be undertaken when implementing development programs in conflict environments, with the two more polarized approaches at either end of the diagram. At one end of the spectrum, programs operate on the principle that development will bring peace to a certain context. This is often the classic poverty alleviation approach where it is assumes that the major driver of conflict is poverty and competition over resources. This fundamental thesis suggests that if one alleviates the poverty then peace will prevail as people’s livelihoods improve. At the other end of the spectrum there are those who propose working directly on peace building processes independently of broader poverty alleviation work. In the broader field of development, practitioners take three major approaches when they find themselves in conflict environments they either:

• Work around conflict – treating it as an impediment to be avoided

• Work in conflict – attempting to use DNH and conflict sensitivity methodologies

• Work on or with conflict – designing programs that deliberately attempt to alter the dynamics of

PEACE FOR DEVELOPMENT DEVELOPMENT FOR PEACE

conflict

Increasingly development practitioners who find themselves operating in conflict environments tend to take the second approach and attempt to use Do No harm and/or conflict sensitive methodologies (with variable success). SIPED started very much at the ‘Peace for Development’ end of the spectrum at the beginning of the program, but as time has progressed they have moved further along the spectrum towards the ‘Development for Peace’ end, partly because this was a natural progression and partly due to the constraints of the environment.

Programs that are only at the extreme ends of the spectrum will not be effective, as working on peace-building without addressing the development needs of communities will just end up being a series of ‘talk fests.’ Equally approaching purely from the development for peace end of the spectrum will not address many of the fundamental structural drivers or proximate causes of the conflicts. Ignoring this complicated set of drivers will not mean they simply go away.

FINDINGS

This section is organized by the evaluation questions and sub-questions as presented in the evaluation Scope of Work (Annex I).

Design

Effectiveness of the Design

The project peace objectives were very relevant to the context and conflict situation of Ethiopia. They were deliberately broad in intent and vague in nature, which was appropriate given the difficult context and space available for undertaking governance work within a difficult operating environment. This has meant that the project has built in significant flexibility in implementation. It is always hard to undertake conflict and peace work in such environments and Mercy Corps has been able to create considerable space to do so. This has also meant that it has been able to evolve and adapt as opportunities have arisen and as confidence and trust has been built.

Conflict Prevention and Resolution training has proved to be the right entry point into a fundamentally politicized field that has also served to try and depoliticize it. As noted elsewhere in the report this training has also served as a means to ‘normalize’ the discussion of the topics.

In terms of design, the peace processes adopted by Mercy Corps have been excellent because they are inclusive, participatory and cautious – as they should be when lives are at stake. The success that they have achieved has been partly as a result of extra effort to include all stakeholders such as women, youth, and traditional leaders and most importantly others who if not included might become spoilers later down the track.

The change from CPR to SIPED has also marked a timely evolution in approach from more direct peace work to development work with the expectation that it would impact (reduce) conflict and consolidate peace. This lead to the second set of Youth and Water objectives which are also appropriate as the application of sound peace principles to the development of shared resources is in keeping with the broad evolutionary design to achieve the objectives.

There have been areas of weakness though, some of which touch on challenges that are not easy to address. For example, project staff report that one of their lessons learned over the course of CPR and SIPED was the importance of working on the development needs of stakeholders simultaneously while working on peace, rather than just focusing on talks or dialogue. In actuality, this ‘lesson learned’ is not new as it is expressed by peace workers across the globe time and again. Hence, the design could have benefitted significantly from a more deliberate incorporation of this element from the beginning (this has traditionally been seen since around the year 2000 in the rather dated language of ‘peace dividends’ which suggests the provision of a ‘reward’ or a visible benefit from having peace). The project did move naturally to this approach but this could have been formalized much earlier and incorporated within a peace and development Mercy Corps strategy where peace and development are ends in themselves.

While clearly tied to the funding streams, and very broadly lying within objective four of the peace objectives, the water and youth objectives (i.e. the development objectives) could have been incorporated more explicitly earlier in the project. The result of adding these objectives later on in the program created a fragmented program design; though it is noted that adding these objectives were partly driven by donor decisions rather than it being an overt strategy of the implementing partner. This in turn does suggest that, while the fundamental flexibility of USAID has been a strength of the program in allowing change and an iterative approach, it makes more sense to undertake democracy and governance programming in Ethiopia within a more overtly and deliberately flexible funding mechanism that lends itself to evolution, opportunism and responsiveness, rather than trying to fit such programs into the ‘wrong’ administrative box which risks limiting innovation and creativity. The contextual challenges associated with governance programming in Ethiopia described throughout the report, demand a light, flexible, funding mechanism (more like an OTI-type approach) that can shift rapidly to take advantage of new opportunities (or withdraw and redirect if the space is closing down or threatens unsavory repercussions) and does not have to contort itself to fulfill administrative obligations. This would be a helpful contribution to the potential success of a governance program whether it is operating at the federal or grass-roots level.

Similarly, assured funding commitments in the long-term to DG programming is critical in the Ethiopian context where a large development and poverty alleviation portfolio with significant implications for regional Horn of Africa stability is more vulnerable if not supported by continued, ongoing, robust governance programming. Good governance is the ‘oil between the cogs’ that enables sectoral and poverty alleviating programs to be effective. Continued, consistent work in the governance and conflict sectors should, at a minimum be considered as insurance for other larger development investments. The evidence for such an approach to stability is encapsulated in the large body of written work on fragile states which articulates the development cost of failure to address conflict and instability and the capacity development of institutions to manage them.

Recommendations:

USAID should ensure that its funding mechanisms for democracy and governance programs in Ethiopia institutionalize the following characteristics to maximize impact in the sector and context:

flexibility, adaptive ways that encourage innovation, and creativity to take advantage of gaps and opportunities to leverage space for improved governance

SIPED can be divided into three phases over the course of its eight years. The first phase during the initial few years, the project was characterized by a necessary emphasis on training programs before entering the next phase of trying to reach a peace agreement. The third phase has been one of consolidation around development for peace – inter-community resource sharing or cross-border livelihood options while simultaneously addressing some of the root causes of the conflicts around natural resources.

The above graphic illustrates how the program was divided. It is important to note that the complexity of the processes undertaken in phase two should not be underestimated.

With hindsight, this design was overly ambitious to try to achieve in all geographic locations at the same time, even over an eight year project direction – 2004 to 2012 – which is rather reasonable. Instead, a ‘rolling program’ approach might have been more advantageous as it would have required a commitment to achieve peace consolidation to a certain level in one area before moving to another cluster or system of conflicts. Conclusion -- this has not been a simple project to implement.

Cluster / conflict system One

Cluster / Conflict system Two

Cluster / Conflict system Three

Time

Alternative ‘Rolling’ Model in addressing conflict in Ethiopia

The gradient within the arrows illustrates the shift of project emphasis from peace work to development work. While it is certainly feasible to be working on at least two complex conflict systems at one time, the number being addressed should…