Comparing DDR and durable solutions: some lessons from Ethiopia

An increasingly important area of focus for relief and development agencies relates to the demobilisation and reintegration of fighters and support for displaced people to return home or resettle elsewhere. Both groups – ex-combatants and forced migrants – are regularly implicated in war and post-conflict situations as perpetrators, victims and survivors of violence. They are widely considered to have become ‘dislocated’ from the mainstream, constitute potential ‘spoilers’ and are frequently targeted by comprehensive programmes designed to promote their sustainable ‘reintegration’ into society. But despite these and other similarities, there is comparatively little exchange between those working with soldiers and those working with forced migrants.

This article considers a number of lessons emerging from an innovative demobilisation and reintegration programme (DRP) involving more than 148,000 Ethiopian veterans. While launched specifically on behalf of veterans and not refugees or internally displaced people, the DRP offers potentially important insights for policy-makers and practitioners working to promote ‘durable solutions’ for the displaced. While the discourses and lexicons may differ, the challenges associated with ‘reintegrating’ soldiers and forced migrants ‘back’ into their former lives are broadly comparable.

A critical lesson learned is that, while reintegration is inevitably context-specific, it should nevertheless be conceived broadly. Genuinely sustainable reintegration has economic, social and political dimensions, each of which is interconnected. Another lesson is the need for humility on the part of DRP donors and planners. In the case of Ethiopia’s DRP, successful reintegration was in large part a function of the (relative) absorptive capacity of areas of return, the endowment sets of former soldiers and only lastly the quality and quantity of ‘benefits’ on offer. In other words, interventions on behalf of veterans and displaced people should accommodate all three ‘tiers’ and recognise the role of targeted assistance in relation to the whole.

Comparing DDR and durable solutions

What, if anything, do former combatants and forced migrants have in common? For one, they are frequently exposed to a bewildering array of interventions administered by a disparate collection of humanitarian, development and security agencies. Often coordinated by the UN Department for Peace-Keeping Operations (DPKO), the UN Development Programme (UNDP) and the World Bank, disarmament, demobilisation, reinsertion and reintegration programmes (DDR or DDRR) are expected to transform soldiers into productive civilians. Likewise, the UN High Commissioner for Refugees (UNHCR) and a number of UN and non-governmental agencies often assist refugees and internally displaced people to achieve a ‘durable solution’ such as repatriation, resettlement or return, and to ensure that they are adequately ‘protected’ and self-reliant.

While those agencies administering DDR and durable solutions often work autonomously from one another, there are a few instances where they actively collaborate. Their cooperation is motivated more by necessity than by design. For example, throughout Africa and the Balkans, DPKO, UNDP, UNHCR, the International Organisation for Migration (IOM) and the World Bank often collaborate (with governments) to ensure that ‘foreign’ ex-combatants residing outside of their countries of origin are separated from genuine refugee or IDP populations, and are ‘repatriated’ back home. Such activities are frequently intensely political and may entail criminal proceedings. Predictably, there is comparatively less focus on their sustainable ‘reintegration’ into civilian life.

Regardless of whether there is (inter-agency) cooperation or not, both DDR and durable solutions ultimately seek to enhance the wellbeing of potentially vulnerable populations. DDR and durable solutions generally only proceed in ‘post-conflict’ contexts. They are designed to facilitate a long-term process – often with a host government acting as the coordinating authority – integrating sequenced assistance (e.g. rations, seeds, tools and training) with predictable linkages to mainstream development. Both sets of activities often involve a complex assortment of actors with varying mandates and competencies.

In some cases, a ‘natural’ ordering of responsibilities and division of labour emerges. In the case of DDR, DPKO and to a lesser extent UNDP are frequently involved in disarmament. Owing to their comparative advantage, but also mandate constraints, the World Bank and others are often more engaged in demobilisation, reinsertion and reintegration. In the case of durable solutions, a ‘cluster approach’ recently emerged, in which different agencies assume responsibility for a range of pre-assigned priorities. UNDP, for example, often takes a key role in ‘early recovery’, while UNHCR is responsible for overseeing ‘protection’ and camp management.

Neither DDR nor durable solutions have been as successful as expected or envisioned by their proponents. For a host of reasons, ranging from the extremely complex environments in which they are established to the dilemmas associated with unpredictable assistance and collection problems between agencies, they only occasionally yield sustainable outcomes. It is doubly important, then, to learn from potentially ‘successful’ cases of DDR or durable solutions. This rest of this article reflects on the outcomes of a World Bank-supported demobilisation and reintegration programme (DRP) undertaken in Ethiopia following an 18-month war with Eritrea (1998–2000). Administered rapidly over a three-year period, the DRP stands apart as a rare success story.

Ethiopia’s reintegration experience

The scale and scope of the Ethiopian demobilisation and reintegration programme was breathtaking. Between 2000 and 2003, more than 148,000 veterans – including more than 17,000 disabled soldiers – were disarmed and demobilised by the Ministry of Defence and provided with cash and non-monetary reinsertion and reintegration assistance via the Ministry of Labour and Social Services. The process was carried out efficiently and according to declared principles of transparency and equity (The total cost of the EDRP was approximately $174 million, with less than $3.1m provided by the Ethiopian government and over $170m supplied by an IDA credit).

The vast majority of these former soldiers consisted of so-called ‘new regulars’ and ‘militia’, who were rapidly recruited from predominantly rural areas by the armed forces shortly before the war. The only other DRPs that remotely compare include the demobilisation of more than 350,000 Ethiopian soldiers (and rebels) following the collapse of the Derg in 1991, and activities supported by the Multi-Country Demobilization and Reintegration Programme (MDRP) in nine countries of the Great Lakes of Africa.

The political, security and economic context in which demobilisation and reintegration occur invariably influences the outcomes. A few key characteristics of the Ethiopian DRP stand out:

• It was undertaken following a vicious cross-border war with Ethiopia’s avowed enemy (and former ally), Eritrea.
• The demobilisation entailed a controlled ‘reduction in force’ of a standing national army and disarmament was not contested.
• Most returning veterans were at first treated as heroes by public authorities and Ethiopian civilians.
• With the exception of a few border regions, communities of return were not devastated by years of conflict.
• The war was comparatively short, so recruits were not away from their homes and fields for prolonged periods.

Taken together, these characteristics stand in contrast to other post-conflict situations in Africa, which tend to be marked by protracted civil wars and lingering political and criminal violence in their aftermath.

One recent assessment found that, while generally a positive intervention, the Ethiopian DRP yielded differentiated reintegration outcomes. The participatory beneficiary assessment adopted a proportional size sampling strategy among 15 woredasin Tigray, Amhara, Oromia and SNNP. Overall, the study revealed that, while income and asset holdings partially deteriorated four years after the demobilisation process, many veterans considered themselves to be empowered and accepted by their families and communities, and not unduly discriminated against by political authorities. Whilst the proportion of rural and urban veterans slipping into lower income quintiles increased overall, there is comparatively little evidence of social pathologies or dysfunction.

It should be recalled that the absorptive capacities of areas of return were heavily influenced by a combination of macro-economic and environmental factors. Although measures were adopted by government officials and the World Bank to minimise the stresses of reintegration – including the phased provision of monetised assistance to veterans to avoid distortions to the local economy – overall national inflation rose during the period of the DRP. While it is likely that certain inputs stimulated local-level commercial activity, no evidence of this was found during the assessment. Ethiopia was also affected by a severe drought from 2001 to 2003, which disrupted cereal and pulse crops and intensified food insecurity in certain areas, including those in which veterans sought to integrate.

The origins of veterans and their endowment sets also shaped reintegration outcomes. The assessment found that there were pronounced differences in reintegration between veterans from rural areas and those from urban areas: more than 80% of all participants heralded from the hinterland. While there is ethnic and cultural variation between Ethiopia’s regions, there is also a significant level of ethnic homogeneity in predominantly agricultural and pastoral areas, from where most recruits originated and reintegrated (This is especially the case in Tigray: more than 40% of all regular veterans, ‘new regulars’ and ‘militia’ participating in the EDRP were from this region. Amhara and Oromia are less ethnically homogeneous, followed by Addis and SNNP, where there is a high level of ethnic heterogeneity). Endogenous factors such as kinship networks and respect for authority played a decisive role in facilitating reintegration outcomes. Also, because rural veterans were often away for only short periods, social capital remained largely intact. This was less the case for those originally recruited from urban areas characterised by more ethnic heterogeneity, less visible solidarity and higher labour inflexibility and standards of living.

With respect to DRP benefits, the assessment found that the predictability of assistance was almost as important as the amount received. Veterans were promised a host of entitlements by the Ethiopian government in the wake of the conflict – contributing to rising expectations. In some instances veterans borrowed from local money-lenders against the promise of future income and earnings. But a significant proportion of rural and urban veterans spent their initial cash assistance – the so-called transitional subsistence support – on consumer goods and debt repayment, rather than on ‘productive’ assets as anticipated by programme planners. This constitutes a form of moral hazard. When promised assistance failed to materialise in a consistent or regular fashion, veterans found their financial situation deteriorated (A household survey administered by an evaluation team (AGEG) found that 56% of respondents answered negatively to the question ‘do you believe your reintegration into civil society has been successful?’). Meanwhile, when training in financial management or enterprise development was provided ‘after’ they had spent their entitlements, they had little incentive to continue the course without the means to invest.

Conclusions

The DRP in Ethiopia was broadly successful when compared to other countries in Africa. The country did not relapse into conflict following the programme – although the recent Ethiopian-led intervention in Somalia is arguably a continuation of the previous conflict with Eritrea by proxy. Equally, the process was accompanied by comparatively little social unrest, and there is no evidence that criminality increased in areas of integration. A compelling feature of the DRP was its ‘reintegration’ strategy – an approach that consciously adopted international and domestic best practice and borrowed from contemporary development thinking. While the reintegration outcomes were mixed, they nevertheless yield important lessons for forced migration specialists.

First, the EDRP reminds us that reintegration has economic, social and political aspects. Genuinely successful reintegration requires improvements in all three areas and not exclusively in relation to economic livelihoods. Moreover, there are important linkages between various elements of reintegration. For example, real and relative declines in income and asset distribution can lead to changes in social status and relative acceptance at the family and community levels. It is therefore important to stress both ‘economic’ and ‘social’ aspects of the reintegration process – and the relationships between them.

Second, the absorptive capacity in areas of return invariably influences reintegration outcomes. The macro-economic climate – inflation, commodity prices and the like – will invariably influence the behaviour and decisions taken by returning veterans. But other factors at the micro-level also matter fundamentally. For example, in rural areas there were comparatively more robust social and market networks, and the economy was less affected by labour migration, housing pressure and unemployment than was the case in urban areas. Sustainable reintegration thus requires a concerted focus on enhancing/strengthening the absorptive capacity of areas of return.

Third, the endowment set matters. The majority of participating veterans came from income-poor households and had comparatively low levels of employment, education and marketable skills prior to their recruitment into the armed forces. While most ‘new’ soldiers were originally recruited from jobs as farmers and agricultural labourers, urban recruits frequently ‘self-selected’ and often had few alternatives to joining the army. In other words, the endowment sets of most participating veterans were comparatively low well before the DRP was initiated. The expectations of donors and government planners of what can realistically be achieved in such contexts must be commensurate with the capacities and endowments of the target group.

Finally, if reintegration entitlements are to be made more effective, it is important to ensure the proper sequencing of financial benefits in line with training and vocational support. In many cases, cash assistance was provided to veterans before enrolment in training/vocational courses. Many veterans complained that they spent these funds ‘improperly’. Few had any experience of managing finances on the scale provided, or expertise in making long-term investment decisions. The risks and trade-offs associated with providing lump sum payments to veterans without sufficient financial management skills are well known. Indeed, it is expected that the provision of assistance in phases allows for learning from mistakes.

Robert Muggah, Small Arms Survey, the Graduate Institute of International Studies, University of Geneva, and Queen Elizabeth House, University of Oxford. Robert’s email address is: muggah@hei.unige.ch. Thanks to the Ethiopian Ministry of Social Affairs and Labour and the Addis Ababa World Bank office for their inputs.