Somalia N.W. Demobilisation and Re-integration Programme
(DRP) Evaluation Evaluation Report

European Union
Somaliland

http://www.delken.ec.europa.eu/en/publications/Somaliland%20Demobilisation%20Evaluation%20Report.doc

(This Report is financed by the European Commission. The authors accept full responsibility for this Report that was drawn up on behalf of the European Commission. It does not necessarily reflect the views of the European Commission)

Agrisystems Ltd In association with WSP. December 2002

A. PREAMBLE

1. Principal Features of the Programme

This Evaluation Report contains the findings of the Evaluation Team on the Somalia N.W. Demobilisation and Re-integration Programme (DRP). This is response to the terms of reference issued by the European Commission entitled ‘Evaluation of Demobilisation Activities in Somalia and feasibility for future expanded activities’. The objective of the Programme is to consolidate the peace and stability of Somaliland (N.W. Somalia) – see Map at end the Report listed with Annexes – and to create favourable conditions for sustainable and long-term socio-economic development. Lack of a proper Financing Proposal makes it difficult to determine just what the components are but would seem to consist of (i) target group identification (ii) strengthening NDC, (iii) re-integration activities, and (iv) sensitisation activities. The Programme was formulated in 1999 by GTZ, funds approved after that with implementation commencing in April 2000. The approved funding by the EC amounted to €1.843 million. No indication was given of Government and local costs. The implemented Programme was started in April 2000 and was due to finish after 24 months in March 2002.

Interruptions in implementation occurred with the death of the first Programme Director (killed in Burao early June 2000) and the replacement Director was only in place by September 2000. Then the September 11, 2001 event in the United States resulted in a withdrawal of the Team until November of the same year – a delay of 2 months. As result 2 no-cost extensions were granted from the original closing date of March 2002 to December 2002. These disruptions have caused delays beyond the cumulative time lost time of 6 months since every break requires additional start up time. The case for a no-cost extension of time beyond the end of 2002 will be discussed below.

Conclusion There were a considerable number of delays to the continuity of the Programme caused by factors outside the control of the implementers.

2. Objectives and Plan of Work

The Team Leader arrived in Nairobi on October 16, 2002 for briefing on the assignment and proceeded to Hargeisa, Somaliland on October 19, 2002. Thereafter the Team conducted its assignment through review of documentation, and meetings and discussions with beneficiaries, implementers and other stakeholders. The evaluation in Hargeisa culminated in a workshop held on October 29, 2002 with the main stakeholders. Output from the evaluation included an Inception Report at the start of the assignment, an Aide Memoire presented at the end of the fieldwork and this Evaluation Report. The Team Leader returned to Nairobi on October 30, 2002 and returned to England on November 8, 2002 after submitting the Evaluation Report to the EC Delegation.

The overall terms of reference require an evaluation of the Demobilisation and Re-integration Programme in Somaliland and the pilot project in Merca in the south of Somalia (Phase 1). Phase 2 of the assignment requires an outline of lessons learnt from the two as well as those of other donors, preparation of a feasibility study and plan of action prepared as to how to engage in Southern Somalia with possible partners, and finally a guideline document on how development projects can contribute in a tangible manner to demobilisation. The scope of the evaluation is limited to the Phase 1 activities as they relate to the demobilisation Programme in Somaliland; it was agreed with the EC Delegation in Nairobi that the pilot project in Merca would be covered as part of Phase 2.

B. SUMMARY

PROGRAMME PREPARATION AND DESIGN (SEE CHAPTER C)

3. Identification and Formulation Process

Background to the Existing Programme

By 1997 the lack of demobilisation/re-integration programmes had put a lot of pressure on Government to absorb large numbers of armed militia into its combined forces. This was unsustainable because of the mounting costs and the concerns about possible revival of military power within society. Therefore the Government realised it needed a comprehensive plan for the demobilisation and re-integration of these ex-combatants.

The Proposal for the Current Programme

Formulation of a demobilisation/re-integration programme is hampered by the lack of a recognised Government. Nevertheless, a proposal was prepared in 1999 by GTZ.

Coherence and Realism of Programme Design This can be set out in the following table:

Intervention Logic

Overall Objective

The peace and stability of Somaliland is consolidated and favourable conditions for sustainable and long-term socio-economic development are created

Development Objective
Demobilised Combatants are fully re-integrated into the Somaliland regional economy and society
Immediate Objective (Programme Purpose)
Combatants have gone through the demobilisation process and are in a position to earn their own living on a sustainable basis Whether or not the peace and stability of Somaliland was a prime objective of the Government in 1999 was not clear from the Programme proposal. However, common sense would indicate that it was a priority then and still is.

The conceptual approach to the Programme set out in the Programme proposal avoids determining the numbers and background of the target group (ex-combatants in the security services), calculating the Programme costs, and a proscriptive approach to implementation. It sets out important pointers to the overall approach but many of these did not occur in practice.

The proposal avoids providing any support to those ex-combatants who are not in the security services. In mitigation, UNDP has in the past and is currently proposing to support SOOYAAL (the ex-combatants representative body). It is also evident that the Government wishes to direct demobilisation/re-integration support to its security services, primarily for budgetary reasons.

The proposal has too many results (8) where less would suffice. In addition, it does not spell out in nay detail the assumptions and risks. The overall conclusion is that the basis for evaluation is a Programme design proposal, which was tentative as to the methods and approach for achieving the main Programme Purpose. A Problem Tree, logical framework, Programme costs and financing and assumptions, pre-conditions and accompanying measures were omitted. In addition, no Financing Agreement with supporting documents has been prepared along the normal EC guidelines.

Recommendation

Any future investments along similar lines even within the overall funding umbrella should follow the guidelines and discipline of a Financing Proposal, thus containing a Problem Tree, a logical framework, total costs, assumptions, risks, pre-conditions and accompanying measures.

RELEVANCE OF THE PROGRAMME (SEE CHAPTER D)

4. General Context

The Somaliland Economy

Somaliland is classified as least developed with high unemployment, weak institutions, poor social services and a population trying to recover from years of war and turmoil.

The Political and Constitutional Context

There is rapid change in Somaliland’s governance arrangements currently being undertaken with unknown consequences. Decentralisation has started but has some way to go; local elections due for end 2002 look likely to go ahead (with EC technical support for the procedures); and multi-party democracy over a clan-based system of government is the chosen route for the country. However, economic and political concentration in and around Hargeisa is creating resentment particularly in the east.

Government Policies

With hindsight and given its pilot nature the DRP did not take into account (i) Government’s defence and security polices including desired numbers in each (there were none), (ii) Government’s demobilisation policies and what to do with the many who were unable to perform active duty (there were none), (iii) carrying out a socio-economic survey of all the security services by the Programme implementers (there was no time).

Recommendation

Agreement that one of the first activities by the Programme implementers will be the carrying out of a socio-economic survey of all the security services– a pre-condition.

EC’s Indicative Programming

The EC’s intervention strategy for Somalia has the objectives of contributing to the alleviation of poverty and supporting the sustainable improvement of the livelihoods of the Somali people.

The DRP’s Overall Objectives The overall objectives of the Programme are consistent with current EC strategy.

5. Specific Context

Problems to be Solved

The core problem identified at a workshop held at the end of the Evaluation Team’s assignment in Somaliland is that the demobilisation process is too slow, i.e. has taken too long; an effect of this is that not enough ex-combatants have been demobilised. The core problem is caused by lack of international funding, lack of income earning opportunities for beneficiaries and delays in the re-integration process.

Many of the root causes of problems recognise (commendably) shortcomings in the Government’s own institutions and own internal deficiencies in publicising and supporting the Programme.

There are several valuable pointers to any future demobilisation/re-integration Programme in Somaliland and the wider context of Somalia.

Recommendations

The findings of the workshop indicate that those actions, which can be modified in a new Programme, are:
- The benefits of a Programme should be measurable and therefore systems put in place to allow this to happen
- The weak institutional Government bodies suggests that training and capacity building should continue (even though in the opinion of the Evaluation Team this is a problem that is almost irresolvable except in the long-term)
- Demobilisation criteria must be clear and unambiguous and applied by all parts of the security services. This may involve closer involvement of the Programme in monitoring how these criteria are applied
- More attention needs to be paid to the information and awareness campaign in the security services, with greater real time monitoring of its effects as the campaign is implemented.

Scope of the Problem

The overall scope of the problem can be determined to some degree of accuracy, but is not reproduced here for reasons of confidentiality. However, the percentage of budget spent on the security services can be outlined and is set out below:

Percentage of Budget Spent on the Security Services

Year Security Services Costs Total Budget
% SS Costs to Total Budget
2001---37,511---78,026--- 48%
2002---42,785---88,852--- 48%

The numbers of the potential target group to be demobilised has been together with the desired reduction. However, the figures are at best tentative and could not be substantiated further by the Evaluation Team.

Recommendation

If the estimated figures are accepted then the likely cost of a future investment to demobilise all the ex-combatants would be manageable in a once and for all exercise (but which would have to be determined from a formulation exercise).

Recommendation

If estimates indicate a final demobilisation and re-integration as agreed with the Government then there is a reasonable basis for a once and for all investment.

Problems Resolved by the Programme

This is set out in the table below Problem Explanation
- Lack of recognition of Somaliland. .... MIS-evaluation of the country and its conditions. No direct effect from the Programme, although its (relatively) peaceful implementation provides the international community with an example of the ability to operate successfully in the area
- Difficult to measure effects of Programme.... This relates to the difficulties on measuring benefits – (reductions in Government costs on the security services, and reductions in poverty)
- The first has been a weakness which the Government recognises and have agreed to correct in any future Programme. ..... The reductions in poverty are measurable and are in the process of commencing to be done through the Programme joining FSAU in its field work
- Climatic risk........Difficult to get over this problem since this affects every farmer/herdsman in the area.
- Lack of income earning opportunities......No root cause was given for this but it concerns the overall weakness of the economy. However, the Programme by supporting and investing in the economy in the implementation of its various re-integration activities does support and contribute the gradual build up of the economy
- Institutional capacity of Government bodies (involved in implementation) are weak........ The Programme has attempted to strengthen various implementing institutions, and has trained many Government staff. Despite this the institutions remain weak because the Government pay scales do not pay a living wage and there is little the Programme can do about this
- Unrealistic demobilisation criteria........The Programme has an agreed set of demobilisation criteria with the security services. The Army in particular has chosen to ignore these in recent nominations of demobilisees putting in war widows and the inform (the demobilisees are supposed to be the physically fit). This will require a careful re-think in any future Programme but must tie in with the Government’s objective of having a lean security services composed of fully fit and able bodied personnel
- Poor health situation of beneficiaries......This is covered in the point above. The people the security services would like to re-integrate are now those such as war widows and the infirm and this is reasonable.
- Lack of dissemination and mistrust..... The Programme carried out extensive awareness and dissemination campaigns but it is fair to say that the security services were not fully supportive of these (which reflects on the weak institutional capacity of Government bodies). It is also important to note that the Programme publishes a newspaper (which has the highest circulation of any newspaper in the country)
- Speculation by the beneficiary on the benefits (offered by the Programme)....Again this relates to the awareness and dissemination campaigns and will require greater involvement and commitment by the security services in any future Programme

The Programme has attempted to address many of the root cause problems outlined in the Tree. Where the problems have not been resolved at least they are identified for inclusion in a follow on Programme (if approved). They also provide some lessons learnt.

Other Interventions in Somaliland

Other interventions in Somaliland are numerous – +/- 120, there are numerous implementing institutions – over 40, the EC is the largest donor – 35 interventions and there are +/- 40 different donors. The number of other demobilisation/re-integration programmes is limited; only 3 could be identified and these concentrate on re-integration and not demobilisation.

There is little attempt by the DRP to try and integrate the beneficiaries into these other programmes, although this is as much caused by the limitations of time and the total lack of success to do so in the micro-credit scheme.

Recommendation

It does not appear the Programme used other projects and programmes to bring in suitable ex-combatants as beneficiaries and in a future demobilisation and re-integration programme this should be explored further.

Relevance of the Strategy and Overall Approach This has been evaluated by comparison with the similar exercise in Uganda.

Uganda ................ Somaliland Demobilisation........... Demobilisation was involuntary
Demobilisation is voluntary and an ex-combatant can decide to back out at any time before final re-integration
Demobilisation was carried out in a short period of time and before re-integration
Demobilisation does not occur until the ex-combatant is re-integrated It is not clear that there were any demobilisation criteria There are demobilisation criteria, although the Army in particular has chosen to renege on them
The vulnerable and infirm were a main target group for demobilisation
The fit and healthy are to be demobilised and this is spelt out in the criteria. The Army, however, has recently ignored this For the majority of veterans there was no awareness and sensitisation prior to demobilisation
There are always awareness and sensitisation workshops prior to demobilisation. This is used as a method to attract demobilisees (however, note the criticisms spelt out in the problem definitions above)
Demobilisation was a top down approach with the Army deciding and leaving it to the UVAB to take over once a veteran is delivered to their place of re-integration
Demobilisation is led by the security services since they nominate the demobilisee. However, the process is voluntary and the ex-combatants stays in his/her unit until fully re-integrated

Re-integration

Re-integration consisted in the main of a cash package of around US$ 560/veteran paid in 3 instalments Re-integration consists of economic re-integration through a series of economic options. None of these involve a straight cash handout but are nearly all in kind The amount spent/demobilisee varies depending on the re-integration package the demobilisee moves into – from US$ 600 (agriculture) to US$ 1,200/demobilisee (SME)
Re-integration involved an immediate return to the area of integration
Re-integration is carried out in stages with training an important element
Vocational training was extremely limited Training is an important element of all ex-combatants
Vocational training was applied after demobilisation Training is applied during the demobilisation process and before final re-integration
In terms of re-integration in the community there was little attempt at civic education Civic education is an important element that is included with the general training
Once demobilised the veteran is left very much on his/her own Because the demobilisation process is much extended re-integration occurs before demobilisation is complete
There was little attempt to work with the community where the veteran was moving to (this was needed since family ties are weaker in Uganda than Somaliland and the veteran had very often gone off fighting as a personal decision)
Community awareness is an important element of the overall package, although in the Somaliland context this appears to need less emphasis (the strong extended family, ex-combatant was asked to fight and in any case was nominated by the family. Therefore returning to the family is easier)
The strategy in Somaliland has been fairly slow reflecting the unforced, voluntary nature of demobilisation and the need to socially and economically reintegrate a demobilisee before full demobilisation occurs. As a result the numbers of demobilisees will be from 550 to 770 over an effective 2-year period
As an alternative the Uganda process has been compared. The approach was top down led by the Army and was weakest in the area of re-integration, which was effectively ignored. Nonetheless over 36,300 veterans were demobilised over a 5-year period.

EFFICIENCY (SEE CHAPTER E)

6. Means and Costs

Programme Costs

These are set out in the table below:

Description	Total	Total Spent	Estimated	Unspent Budget to end 9/0
4th Qur 2002	12/02
1	Works	91,200	87,781	9,600	      -6,181
2	Remuneration	651,380	599,084	73,500  -21,204
3	Travel costs	9,377	      5,158	      1,000	   3,219
4	Supplies/equipment 67,169	60,432	2,000	   4,737
5	Local transport	104,503	99,776	3,000	   1,727
6	Backstopping	150,750	153,000	15,488   -17,738
7	Job creation	630,000	261,871	120,000  248,129
8	Management fees	138,361	80,973	15,965   41,423
	TOTALS	      €1,842,740	€1,349,696	€240,553 €252,491

On the assumption that the average levels of monthly expenditure amount to €60,000 to € 80,000 the Programme’s no-cost extension should be to end April, 2003.

Financing There is no calculation of Government contribution to the Programme. Estimates indicate out of pocket cost of €320,000 giving a total Programme cost of €2.16 million.

Recommendation

In any future Programme design Government and other local contributions should be included in Programme costs.

Value for Money

The unit costs of the Programme are high – depending on final numbers from €2,400 to €3,350 – and this reflects the shortfall of the Programme in achieving the original 900 demobilisees. The reasons for this are primarily the interruptions and difficulties faced by the implementers, in particular the late reluctance of the Army to provide names of more demobilisees.

7. Organisation, Management and Monitoring

General Organisational Arrangements NDC is the owner of the Programme at the policy making and major implementation levels; the day-to-day running of the Programme is effectively in the hands of the EC/GTZ Team. Given this and the analysed weaknesses of NDC there is a real question as to the sustainability of the Commission except working in partnership with an implementer.

Recommendation

The organisational arrangements in place in the Programme reflect the activities carried out, and that the staffing in general is appropriate for the implementation of such activities. In addition, the quality of the staff is adequate to carry out the required functions. The staff seconded part time from NDC are also performing well and have adapted to a working regime, which is not evident in Government.

It is not clear that NDC, established in 1993, has any written legal status. Nonetheless, it does have a Minister to report to – RRR – and this is reasonably well established.

Recommendation

It will be a concern for the future that there is some formalisation as to NDC’s establishment and this should be implemented. In general the terms set out in the MOU between Somaliland’s Ministry of Planning and GTZ have been met. Exceptions include the lack of staffing of an office in Burao, no monitoring and evaluation system in the security services, and no defined numbers for the security services.

Plan of Operations and Timetable

Under the present arrangements the Programme is due to close at the end of December 2002.

Recommendation

A third no-cost extension should be granted with a scheduled closing date of end April 2003.
Financial Management Matters and Procurement Procedures
The Evaluation Team has not reviewed financial matters in any detail.

However, it is noted that the last financial reconciliation by the EC was at the end of 2001. There do not appear to be any problems in the application of procurement procedures, which appear to follow EDF procedures.

Recommendation

There is now an urgent need bring the overall financial reconciliation up to date in order to agree on the remaining funds available to the Programme. This will require the earliest agreement of the EC Delegation.

Technical Assistance

The Evaluation Team feels that the technical assistance has performed satisfactorily. In addition the size of the Team has varied as needs arose and the numbers are currently adequate.

The unit rates for the technical assistance are low in comparison with other projects but the add-ons bring it to a level that is more in line with monthly rates (but still low). It is noted that the add-ons are also supposed to cover subsistence costs in Hargeisa.

8. Intervention Methods

Result – The Target Group

This is described in the logical framework as ‘the Target Group is identified and categorised according to the different options of assistance’. In simplest terms these are the demobilisation procedures. The activities carried out, particularly in relation to the database, have resulted in the attainment of the output.

Recommendation

The DDMS should be made freely available by the EC to other projects.

Recommendation

On a wider perspective there should be established and agreed criteria on which to base the selection of ex-combatants defined along the lines of the existing criteria but with more focus than is included at present. This will require inter alia investments in a database and monitoring system for the security services, established and operated by the Programme.

The Demobilisation Recommendation Commission has been replaced by a smaller Demobilisation Recommendation Committee. Given its remit and lack of problems the decision to take this route is reasonable.

Result 2 – NDC

This is described in the logical framework as ‘the institutional capacity of NDC is strengthened’.

The institutional capacity of NDC has been strengthened but has not achieved, nor is it likely to achieve the desired goal of putting the Commission on a sustainable basis to carry on the Programme. It is concluded that given the present circumstances this is almost impossible to achieve. Instead, it is concluded that a once-for-all investment should be made to meet the desired reduction and NDC should then be closed or its functions re-defined. This is set out in an earlier recommendation.

Result 3 – Provision of Assistance

This is described in the logical framework as ‘assistance is identified and made accessible to the target group’. These are the steps taken to re-integrate demobilisees’.

The rural re-integration component has gone well, although there are questions as to whether the remaining animals can be delivered on time. Part of the success of the component relates to its implementation at the time of good rains.

The electrical skills training component has gone well, even though the commitment by the Hargeisa Power Company to employ 70% has been reneged on for the time being. Despite this most seem to have obtained alternative employment.

The micro-credit scheme has been a failure as to repayments although most of the businesses set up can be considered successful. It has been difficult to conclude as to why there has been so little sense of responsibility as regards repayment. The scheme has been discontinued.

Recommendation

In view of the likely nature and background of the beneficiaries it is recommended that in any future programme there should be no credit extended but the support should be in the form of grants as is set out in the small and medium size enterprise scheme set out below. Thus, the grants should be to existing businesses as was envisaged here and also to new businesses. The size of the grant should vary depending on whether the business exists (a lower amount) or is a new business (a higher amount)

The SME component has only just started but seems to got off to a good start. It is too early to determine if the businesses will survive. However, its establishment has further undermined the micro-credit scheme and the latter is effectively dead.

The employment creation scheme is considered to have been a success. In conclusion the various interventions outlined above have been successful in ensuring increases in incomes for demobilisees. This includes the micro-credit scheme, which has proved a failure in terms of repayments by beneficiaries into a revolving fund for the benefit of others but which has nonetheless resulted in a number of thriving businesses.

Additional Area Related to Output 3 – Trauma Victims

The Post Traumatic Stress Disorder (PTSD) component is a very challenging and worthwhile project in a domain too long neglected in Somaliland. However, the preliminary work accomplished so far is an insufficient basis upon which to establish a psychotherapeutic treatment programme. In addition, reporting consistently overstates the achievements and effectiveness of the Programme interventions.

Recommendation

Additional preliminary research and training will be required before the team is ready to implement a treatment programme. Specific recommendations are that greater consideration is given to the vital role of culture in the definition, classification, management, and treatment of mental disorders, and that systemic, family, and group considerations should be incorporated in the history-taking, diagnosis, management, and treatment of Somalis with psychological/psychiatric disorders.

Result 4 – Sensitisation

This is described in the logical framework as ‘both the communities and the ex-combatants are sensitised for demobilisation and re-integration related matters with special attention to the role of gender’.

The sensitisation process appears, to have been implemented satisfactorily reflecting the needs of the ex-combatants and the communities into which they have re-integrated.

Gender Considerations

During the initial design phase of the DRP, it appears that the gender issue was considered neither by GTZ, nor by the Somaliland government, mainly because serving members of the security forces are almost exclusively male and the Programme objectives target reducing insecurity and factional strife and reduction of Government payroll.

Thus, of the income generating activities only the SME scheme directly involved women (war widows and caretakers of trauma victims), whilst others may indirectly benefit women. In mitigation, the employment of a gender officer has gone some way to providing gender sensitivity within the Programme. However, on a wider scale many (if not most) war-affected women are either single mothers or the principle breadwinners in their families and hence outside the ambit of the Programme.

Recommendation

Aspects of the gender dimension should be revisited, especially the contention that those that support male beneficiaries will benefit female relatives, e.g. in the revolving funds component almost all the beneficiaries were men who are more likely to spend a significant portion of their income on qat and other unproductive expenses.

Recommendation

The pilot nature of this phase of the Programme provides opportunities to profit from lessons learned and to make adjustments in the second phase. The gender officer’s position should be strengthened and given a greater role in planning and design of future activities. There is also a clear need for better empirical data on the status of women within the security services (war-widows and female care-givers on the payroll) since this would presumably be a priority group for demobilisation, in line with the Government’s desire to reduce defence expenditure. This fits in with a parallel recommendation that the carrying out of a socio-economic of all members of the security services would be a pre-condition for a future Programme.

Monitoring and Evaluation

There is internal monitoring by the Programme and by the EC Delegation in Nairobi. There is no evident internal monitoring by the Somaliland authorities.

Two previous monitoring exercises have been carried out but because of their limited time inputs cannot be considered full evaluations. Also, the first was carried out too early in the life of the Programme before much had been achieved, whilst the second had recommendations of a general nature.

EFFECTIVENESS (SEE CHAPTER F)

9. Extent to which the Results have Contributed to the Programme Achievement of Programme Purpose OVI

The Programme Purpose is consistent with the overall objectives and with the PCM guidelines.

There has been a reduction of 200 in the numbers of ex-combatants included in the Programme Purpose caused principally by delays in nominating demobilisees from the Army.

Recommendation

The difficulties outlined above would suggest improved communications for any future Programme from the beginning with the agencies from which the beneficiaries are to be selected. This is first achieved in being involved in establishing systems for determining numbers of the target group and the desired end results. It would also involve a more proactive and continuous approach at the earliest opportunity with the security forces involved.

An estimate of the final number of beneficiaries is set out in the table below:

Description	Male 	Female Total

Potential Beneficiaries1/			
	Employment	                  69	3	72
	Micro-credit                  141	1	142
	SME	                        35	54	89
	Rural Re-integration 	      302	0	302
	Skills Learning (Electricity)	167	0	167
	TOTAL	714	58	772 	%	92% 	8%	100%

Active Beneficiaries2/ 			
	Employment	                  54	0	54
	Micro-credit 	            59	1	60
	SME	                        12	2	14
	Rural Re-integration 	      177	0	177
	Skills Learning (Electricity)	144	0	144
	TOTAL	      446	3	449 	%	99%	1%	100%

Notes: 1/ Ex-combatants who have been screened, interviewed, given an ID and potentially are beneficiaries
2/ Beneficiaries who are receiving all their benefits
Of the 772 potential beneficiaries the number of active beneficiaries is only likely to reach around 550 to 600, the main reason being the difficulty of buying animals for the rural re-integration beneficiaries.

Recommendation

Given the difficulties in buying animals at this time the Programme should be extended on a no-cost basis as long as possible.

Other relevant information is as set out in the table below:
Description ----Female--Male--Total
Custodian corps---6---165 ---171
Army--------------26---316---342
Police------------26---233---259
TOTALS------------58--714---772

It is apparent that the custodial corps has demobilised the greatest percentage, followed by the Police and lastly the Army. Yet it is the Army where the over-manning is the greatest and where there is the need for the largest reductions.

Recommendation

In any future exercise it will be necessary to concentrate the information and awareness campaign and also to ensure that Government applies pressure on the Army Command structure if the agreed support is not forthcoming.

Programme Flexibility

The original Programme proposal built in a high degree of flexibility, because much of the Programme design, implementation methods and direction was left to the implementers. This has been taken up by them during implementation and they have adapted as needs arose.

Geographical Focus of the Programme Analysis shows that the predominant area of activity under the Programme is the west of Somaliland (96% of potential beneficiaries), with little support given to the east, even though a Programme office was supposed to be set up there (in Burao).

However, the death of the previous Director in Burao (the east) has had negative effects on involvement there. Also, there are good reasons for the preponderance of beneficiaries in the west, given the short time frame of the Programme, and it was not possible during the evaluation exercise to determine whether or not this reflects the make up of the security services.

Recommendation

The east of Somaliland should be a target area for future involvement, working on the assumption that those who are demobilised have some connection with the area.

Summary of the Situation Towards the End of the Programme

The Programme has accomplished much in terms of a sound starting point for a Phase 2 if it is decided to provide further support. Any delays in follow on support will have negative repercussions on realising the considerable infrastructural investments and contacts made, and methodologies successfully formulated under this Phase.

Recommendation

If the donors wish to reap the benefits of the sound start made under the existing Programme then decisions should be made in the shortest possible time as to future funding along the lines suggested. If, in turn, this is accepted then a formulation exercise should take place as soon as possible; in any case before the end of 2002. Finally, consideration has also to be given as to an approach to those veterans promised support in the 1993 Declaration but who have yet to receive anything. The consideration should involve other donors.

Unforeseen Results/Beneficiaries

There have been a few unforeseen results and beneficiaries, which include the possible wider application of the information system, the progress made by staff seconded from NDC to the Programme, and the unexpected direction of the employment assistance component.

Possible Accompanying Measures

There are several areas where conditionality/assurances should have been included in the Programme – government policies on defence and demobilisation, measuring Programme benefits through the security services, and carrying out a socio-economic survey on the security services at the start of the Programme.

IMPACT (SEE CHAPTER G)

10. Contribution of Programme Purpose to Overall Objectives

General Context

The accomplishment of the Programme in relation to the overall objective is modest, even excluding those ex-combatants registered with SOOYAAL. It has not been possible to measure the effects of the Programme in terms of the first two OVIs (reductions in insecurity and factional strife) as stated in the logical framework, although indications are that peace and security is evident in the country. It has also not been possible to determine whether individual beneficiaries have been dropped from the security services payroll sheets. However, Government budgets for 2001 and 2002 indicate that the amounts spent on the security services to total Government budget has remained constant at 48% for the two years.

General Development Objectives

Given a general development objective of reductions in poverty then the total estimated number of beneficiaries (3,300 to 4,600) are insignificant in relation to the estimated population living in absolute poverty (around ½ million). However, this should be put in the context of the Programme, which is only a pilot one.

However, a demobilisation/re-integration programme such as the DRP is not a development programme. It has clear objectives set out in its Programme Purpose and these do not include the wider development considerations of the various sectors in the economy.

In the Context of the EU’s General Development Objectives

The DRP fits into the support provisions of the Cotonou Agreement for countries without established governments. It is involved in institution building (NDC), economic and social development activities (its re-integration activities) and is focussed on the needs of some the most vulnerable sections of the population (people in the security services who are not on a living wage, some of whom are vulnerable – the old, war widows and the physically and mentally sick).

Recommendation

If it is decided to invest in another demobilisation/re-integration programme (in the Evaluation Team’s view this should be a once and for all investment) then this should not be seen in isolation of investments in the main sectors of the economy (rural development, education, health, infrastructure etc.). These should also be developed and undertaken by the donors reflecting cohesion and critical mass, which the current small isolated donor interventions do not. This will require a re-evaluation by the donors in the context of Somaliland’s political and constitutional development.

ECONOMIC & FINANCIAL ANALYSIS (SEE CHAPTER H)

11. Economic Analysis

No economic analysis has been carried out since the nature of the Programme – demobilisation/re-integration – does not lend itself to evaluation of this kind. However, an indication of likely returns can be concluded from the financial analysis below.

12. Financial Analysis

On the basis of the assumptions made the rate or return is calculated as 2%. However, if the full number of the beneficiaries had been achieved (900) the rate of return would be 12%. The returns are considered acceptable given that this is a pilot programme and the short time and nature of the investments would give rise to larger benefits in a follow on programme.

SUSTAINABILITY/REPLICABILITY (SEE CHAPTER I)

13. Policy Support Measures

Support from the Government

The Government has given general policy and administrative support to the Programme, but has not formulated any policies on defence, security and demobilisation.

Financial Sustainability

This can be considered for the various income earning activities carried out under the Programme.

The irrigated farming investments are financially sustainable.

For the agro-pastoralists whilst the inputs are sufficient for a start there is some concern as to the financial sustainability of this group when the first drought occurs, although this applies to all farmers irrespective of whether or not they are demobilisees. Whilst the animals provide some security, the sustainability of the agro-pastoralist, if the next season is a drought one, is questionable; the strong point here is that the re-integration is a family based one and therefore in times of stress the extended family assists to the extent possible. The closure of the Programme in the near future will not assist in monitoring and assessing the situation.

There is concern by the Team as to the financial sustainability of the nomadic pastoralists especially when the first drought occurs. Their financial base and income earning capacity is limited in the first place (around US$ 350 per year). Whilst the animals themselves provide some security the sustainability of the nomadic-pastoralist, if the next season is a drought one, is questionable. Much will depend on the ability of the owner to move the animals to where there is pasture and the ability of the extended owner to assist in times of stress. The closure of the Programme in the near future will not assist in monitoring and assessing the situation.

The overall conclusion for the rural re-integration component indicates that this is the largest in numbers of beneficiaries, but at the same time the most vulnerable as to sustainability, with the nomadic pastoralist most under pressure. However, the Programme has pointed out that the approach has been family based (not individual) and therefore the family acts as the safety net in the event of adverse circumstances.

The prospects for financial sustainability of those who have passed the electrical skills training are good.

The micro-credit scheme itself is not sustainable and has been discontinued. However, of the 60 businesses that started 49 are still in operation and appear to be financially sustainable.

Recommendation

Ongoing training should be provided to beneficiaries who have repaid; the topics should include business practices and record keeping. It is too early to tell whether or not the SME scheme is financially sustainable but the prospects are promising.

Recommendation

As for the micro-credit scheme ongoing training should be provided to beneficiaries in business practices and record keeping.

In general the prospects for sustainability of the employment creation investments are good with satisfaction expressed by both beneficiary and owner. One area of minor complaint by beneficiaries was the lack of tools provided by the owner on occasion.

Recommendation

An area to be looked at in the future is the limited provision of tools to owners in lieu of or in addition to other benefits that are provided. Training is an additional area of support to businesses. This would also assist the business in ensuring its sustainability.

14. Appropriate Technology

In general, technology has been appropriate to the Programme, except in the case of the micro-credit scheme where one reason for non-repayment was the inability of the Programme to ensure that the funds were controlled by women, who traditionally are seen as more responsible in credit situations.

15. Environmental Protection

Whilst there are negative environmental effects of the Programme these are minimal when looked at the country as a whole.

16. Institutional and Management Capacity

The institutional and management capacity of the Programme as a whole is strong when combining the planning and implementation strengths of the EC/GTZ Team with the political strengths and influence of NDC management. However, without this combination it is unlikely that the trained staff would stay with NDC, and hence the conclusion that the institutional and management capacity is vulnerable and most probably unsustainable.

CONCLUSIONS & RECOMMENDATIONS (SEE CHAPTER J)

17. Overall Outcome

A summary of the successes and failures of the Programme are set out below: Description
SUCCESSES (at all levels)

- A structure and modus operandi has been built up for a future programme
- An approach to re-integration has been developed, which appears to work in the Somaliland context
- Confidence has been built up with businesses and communities that the Programme can deliver
- A workable and highly useable information system has been developed, which is an asset to be used in any future Programme and could be used in other EC projects and programmes
- The re-integration activities appear to have given rise to sustainable livelihoods for the beneficiaries concerned
- Re-integration with communities went smoothly and did not appear to face the problems encountered in other countries
- Procurement appears to have gone smoothly (this is not the case in many EC projects/programmes evaluated by the Evaluation Team)
- There has been a high degree of flexibility in Programme implementation
- The organisational arrangements in the EC/GTZ Team were appropriate and adaptable enough to respond to changing circumstances
- The FIRR is positive if low at 2%. If the full 900 beneficiaries had been achieved the rate would have been 12% - an acceptable rate for the Horn of Africa FAILURES (at all levels)
- The original proposal was incomplete and did not include a logical framework and a Problem Tree
- The time frame for implementation was too short, although this has been affected by unforeseen disruptions in implementation
- The number of beneficiaries will be no more than 770 maximum and most likely around 600. This is below the original Programme Purpose number of 900
- The number of likely beneficiaries are far below those that require to be demobilised, caused by the limited available funding
- As a result of the limited number of beneficiaries (see above) the cost/beneficiary will be high when compared with those implemented with other countries
- The desired co-operation with the Army was not forthcoming towards the end of the Programme. This occurred with the change in the Army Command
- The demobilisation criteria, whilst clear, were inappropriate and would likely to be so in any future Programme
- The inability to strengthen Government institutions on a sustainable basis
- The inadequate funding by Government of the main implementing body – the NDC
- Concentration of benefits has been overwhelmingly in the west and there have been few in the eastern part of Somaliland. To this end the Programme did not renovate and staff the office in Burao as the MOU indicated should have happened
- The information and awareness campaigns with the security services did not achieve all of its objectives and there was uncertainty and misconstrued benefits of the Programme amongst ex-combatants
- There was a marked gender imbalance in favour of males (although this was likely given that the target group is employed in the security services where the majority are males)
- Benefits/beneficiary provided by the Programme were limited by funding constraints
- More has to be carried out in trauma activities for a workable programme to be implemented
- It has been impossible to measure whether or not the Programme has provided the intended benefit of reducing the percentage spent in Government’s budget on the security services. In addition, the level of troops was never defined in the MOU as should have been done
- No conditionality was developed in Programme formulation on measurement and verification of benefits
- No clear Government policies on defence and demobilisation have been developed
- The Programme avoided considering what to do with ex-combatants expecting benefits, but who are not in the security services – this is a matter which will have to be considered by donors in the future (note that this was never in the scope of the Programme)
- The micro-credit scheme was a failure in that the repayment rate was extremely low (estimated at 12%)

Achievement of Objectives in Relation to Time and Budget

The estimated costs of €1.59 million at the scheduled closing date of end December 2002, are some €252,000 less than the budget of €1.84 million. Hence the decision to recommend an extension in view of the shortfall in beneficiaries.

18. Sustainability

In general the main weaknesses identified by the Team on sustainability are those related to the Government institutions involved in the Programme, who are unable to keep qualified and trained staff, the risks faced by agro- and nomadic-pastoralists in the event of prolonged drought (although this is faced by all farmers), and the extremely poor payment record of micro-credit beneficiaries, which has caused that intervention to be discontinued.

19. Alternatives

No radically different approaches to demobilisation/re-integration were identified as being workable in the Somaliland context.

LESSONS LEARNED (SEE CHAPTER K)

20. Summary

This can be set out in table form: Lessons Learned -- Comment

POLICY, ORGANISATIONAL & OPERATIONAL LESSONS

- A carefully prepared and reviewed Financing Proposal or its equivalent should be prepared. This should include inter alia a logical framework, a Problem Tree, total costs, assumptions, risks, pre-conditions and accompanying measures This should assist in start up, even though later events may require flexibility
- In a demobilisation/re-integration programme a socio-economic survey of the target group should be carried out early on in the Programme and from this demobilisation criteria established
Although demobilisation criteria were established these proved to be unworkable at the end. In part this was because it was unclear as to the likely demobilisees
- There should be a realistic assessment as to whether or not Government institutions can be made sustainable, and if not an alternative approach adopted It is the conclusion of the Evaluation Team that in the context of Somaliland the sustainability of Government institutions is open to question for the foreseeable future, and it is unrealistic to assume otherwise
- Where possible, efforts should be made to include ex-combatants as beneficiaries in other donors projects & programmes
In the context of the DRP, time limited these efforts
- Evaluation should be planned early enough to ensure continuation of a Programme into a follow on phase, especially if the benefits of the investments in the earlier phase are to be realised
Even with an extension of time to end April 2003 it is doubtful if a follow on programme can be approved without a halt in operations

POSSIBLE PRE-CONDITIONS FOR A SIMILAR FUTURE PROGRAMME

- The policy framework by Government should either be in place or its preparation and acceptance by the donor included as a pre-condition
In the context of the DRP the Government had no policy as to defence or demobilisation
- The benefits of a Programme should be agreed, measurable and procedures put in place to allow such measurement to take place.
This will include the establishment and operation of a viable monitoring and evaluation system A negative effect of this is that a lack of definitions and measurement is not conducive to further support by donors

GENERAL DEVELOPMENT LESSONS

- Re-integration strategy such as that developed for the DRP must be carefully thought through and made to fit in the social and cultural context of the country
Experience shows that in strategies for demobilisation in other countries would be unlikely to succeed in Somaliland
- A demobilisation/re-integration programme should be devised with an overall exit strategy based on the completion of demobilisation – it is not an open-ended activity. It should also include a strategy for the end result of the government organisation charged with this process
It cam as a surprise to NDC when it was pointed out that the organisation should have a finite life and that there was an end view in sight towards completion of demobilisation in Somaliland
- Political and geographic considerations should be taken into account in developing strategies in a demobilisation/re-integration exercise
Lack of time has meant that the Programme has concentrated on the western part of Somaliland, with relatively few of the benefits accruing to the east
- It should be made clear that a demobilisation/re-integration programme is not a development programme since the Purpose is different from sector specific interventions In the context of the DRP several sectors have been affected. It is concluded that in the context of Somaliland cohesive sectoral strategies should be developed and investments proposed of a critical mass
- Sustainable livelihoods are a keynote theme of re-integration activities This has been reinforced by the DRP and should not be forgotten in the future

C. PROGRAMME PREPARATION AND DESIGN

IDENTIFICATION AND FORMULATION PROCESS

21. Background to the Existing Programme

The starting point to the Programme is the end of the long liberation struggle against the regime of Somalia’s former President Mohamed Siad Barre with the victory of the Somali National Movement (SNM) in N.W. Somalia. This occurred in January 1991. The SNM called for peace and cessation of all hostilities within the territories of the former British Somaliland. A subsequent conference in May 1991 held in Burao issued 3 declarations – Somaliland to regain its independence, peaceful coexistence and reconciliation between the Somaliland clans, and rewards for SNM veterans with special benefits in recognition of struggle and sacrifices for the liberation of Somaliland. This was adopted by the country’s then legislative body in May 1993. At the same time the SNM War Veterans Association (SOOYAAL) was legalised as a private voluntary organisation representing the interests of the fighters and war widow headed families.

The process of demobilisation followed the 1991 fighting with many fighters demobilising themselves. However, lack of demobilisation assistance meant many had little success in building up another livelihood. At the same time many of the disbanded fighters re-grouped as armed clan militia; by the end of 1991 there were more than 40 such militia in addition to dozens of irregular forces scattered over the area. The subsequent disturbances were resolved through a series of community based peace conferences but the disruption continued in sporadic outbursts in 1994 and then in 1995 to 1996. Peace was only restored in early 1997.

Throughout this period the lack of demobilisation/re-integration programmes had put a lot of pressure on Government to absorb large numbers of tribesmen into its combined forces to satisfy every clan in the country. This was unsustainable because of the mounting costs and the concerns about possible revival of military power within society. Therefore the Government realised it needed a comprehensive plan for the demobilisation and re-integration of these ex-combatants. At the time the estimated number of ex-combatants on the Government payroll to be demobilised was 10,000. It was also estimated that a further 50,000 had demobilised themselves and were waiting for support. However, it was determined that the priority for Government was to reduce the payroll.

Conclusion By 1997 the lack of demobilisation/re-integration programmes had put a lot of pressure on Government to absorb large numbers of armed militia into its combined forces. This was unsustainable because of the mounting costs and the concerns about possible revival of military power within society. Therefore the Government realised it needed a comprehensive plan for the demobilisation and re-integration of these ex-combatants.

22. Previous Interventions and the Current Proposal

Previous Interventions The process of support started in August 1994 with a German Government financed demobilisation project in Somaliland implemented by GTZ. The total allocation amounted to DM 5.4 million (or around ECU 3 million). The main objective was to create employment opportunities for ex-combatants. Social and economic re-integration measures were to target the 10,000 ex-combatants and their families as well as war widows and their dependents. Before the Project was able to achieve its goals GTZ had to discontinue its activities because of an incident in another GTZ run project, which resulted in the cessation of all co-operation in Somaliland. It is not clear how successful the Programme was but the continued disruptions in Somaliland at the time made implementation difficult.

The Proposal for the Current Programme

With peace established from 1997 on it took some time for the demobilisation/re-integration process to resurface. Part of the problem was and still is the difficulty in working with a Government that is not recognised internationally. However, in 1999 GTZ prepared a draft proposal ‘Demobilisation in North-West Somalia’, which formed the basis of the ongoing EC-funded Somalia N.W. Demobilisation and Re-integration Programme (DRP). On this basis the GTZ proposal dated June 1999 represents the definitive document on which to carry out the evaluation. It covers the demobilisation of up to 10,000 demobilised combatants and recognised it did nothing for the remaining ‘50,000 to 100,000’ armed population.

Conclusion

Formulation of a demobilisation/re-integration programme is hampered by the lack of a recognised Government. Nevertheless, a proposal was prepared in 1999 by GTZ.

23. Coherence and Realism of Project Design

In Terms of Objectives and Purposes

The Evaluation Team has not had the benefit of a normal Financing Proposal and supporting documents on which to base its evaluation. This reflects the unusual position of Somalia as a whole where there is an absence of a functioning national government and administration, and where two distinct self-administering regional entities have emerged (Somaliland and Puntland). Without an internationally recognised government Somalia has not signed the 7th and 8th EDFs, and has been without a National Authorising Officer. As a consequence Rehabilitation Programmes have been approved, which constitute overall funding umbrellas covering defined timeframes. Funding for the DRP has been provided from the 2nd Rehabilitation Programme for Somalia.

The GTZ proposal did not contain a logical framework or a Problem Tree, although the descriptive format follows the logical framework format. A detailed timetable of the Proposed Work Program is included as well as terms of reference of many of the proposed technical assistance positions. The proposal determined its main objective as an evolution from the aspects of security (disarmament), political (national unity), fiscal (reduction in public expenditures) and economic (resumption of economic activities). From this an immediate objective (Programme Purpose), development objective and an Overall Objective (logical framework terminology) was determined. These are as set out below:

Description Intervention Logic

Overall Objective

The peace and stability of Somaliland is consolidated and favourable conditions for sustainable and long-term socio-economic development are created

Development Objective

Demobilised Combatants are fully re-integrated into the Somaliland regional economy and society

Immediate Objective (Programme Purpose)

Combatants have gone through the demobilisation process and are in a position to earn their own living on a sustainable basis. In terms of whether or not the peace and stability of Somaliland was a prime objective of the Government in 1999 was not clear. However, common sense would indicate that it was a priority then and still is. The history and background to the proposal set out above in Sub-section 21 gives a clear indication that the years of war and turmoil from the 1980s to the cessation of hostilities in 1997 should not be repeated. Further that any Government has as a main objective the socio-economic development of its people is a reasonable assumption. To achieve that a semblance of law and order is required within a framework of security services that are manageable (not a threat to the government) and sustainable (affordable). Thus, it is fair to conclude that the overall objective is appropriate.

It is unclear why the proposal contained a development objective since the development and immediate objective can be put together as a Programme Purpose. A suggested wording could be that ‘combatants are demobilised and fully re-integrated into the Somaliland regional economy and society’. The two go together – demobilisation will not work without re-integration.

Conclusion

Whether or not the peace and stability of Somaliland was a prime objective of the Government in 1999 was not clear from the Programme proposal. However, common sense would indicate that it was a priority then and still is.

Conceptual Approach to the Proposed Programme

The proposal makes clear that it is not based on any socio-economic survey as to the numbers and type of demobilisees but uses the figure determined some years earlier. It is not clear that this is a reasonable basis for formulation. However, the proposal avoids this by not setting out any costs although a timeframe of 2 years is proposed. It also avoids a proscriptive approach to implementation by stating that ‘due to the complex set of problems of the conditions of the ex-combatants, only a highly flexible approach and a combination of various activities will be successful’. The proposal then suggests a ‘process oriented approach’ involving a ‘permanent learning process and dialogue amongst all partners concerned’.

Other important points relevant to the overall approach include:
- Beneficiaries and communities to be involved as much as possible in the identification, planning, and implementation of project activities
- Destitute groups to be included
- A group approach whereby the groups will participate in the planning and execution of interventions
- Promotion of activities, which generate employment and income activities after a short period
- Use of labour intensive public works
- Information and sensitisation awareness will be undertaken
- Implementation by various groups at the grass roots who may put forward their own proposals.

As will be seen later many of these have not happened in practice, due to non-acceptance, the traits of Somalis, which promotes individualism and the non-sustainability of for example public works. Conclusion The conceptual approach to the Programme set out in the proposal avoids determining the numbers and background of the target group (ex-combatants in the security services), calculating the Programme costs, and a proscriptive approach to implementation. It sets out important pointers to the overall approach but many of these did not occur in practice.

Consideration of Other Ex-combatants

An important consideration of the proposal is its avoidance of any support to those ex-combatants outside the security services, namely those registered with SOOYAAL who are not paid by the Government and are in need of support (estimates are that they number around 17,000 to 20,000 ex-combatants plus their dependents). This is an issue that surfaces from time to time and which arose at the time of the Evaluation Team’s visit to Hargeisa. It appears that this could be a powerful political and destabilising force, but one that the Government appears not to wish to deal with at present. Consideration of its inclusion in any future investments was not raised by Ministers at the final Workshop. A prime consideration on Government’s part is the reduction of its budget, and hence its concentration on those in the security services.

Another point is that UNDP has made approaches to fund this group over the years and has supported SOOYAAL in improving its administration but has not yet funded a demobilisation/re-integration programme. Discussion with UNDP indicate that they wish and have wished in the past to do this but their record in translating this into action has not been good.

Conclusion

Results and Activities

The proposal sets out in some detail the outputs (results) and activities. The results are set out in the logical framework (Annex 1) prepared by the Evaluation Team from the proposal document. It can be seen that the results (8 in number) are too many and would indicate that there should be 8 components in terms of implementation. Given the limited funds that would likely be spent this is overly complicated and many of the results can be merged together. It is probably reasonable that the number of results should be 4 to 5, and this, the Programme implementers has done in its logical framework.

The activities set out in the proposal are numerous and relate to the 8 results. These are not set out in Annex 1 but a review of them indicates a close correlation with the individual results. The proposal does not set out any definable assumptions and risks. A paragraph in the Report sets out the key factors for Programme success and sustainability outside the control of the implementers – commitment and participation of the beneficiaries and the population. Regional authorities are also mentioned with the closing remark that failure in commitment and participation of these 3 groups would be a ‘killer’ assumption. Assumptions are made by the Programme implementers and these will be considered below in the relevant sub-section Conclusion The proposal has too many results (8) where less would suffice. In addition, it does not spell out in nay detail the assumptions and risks.

Overall Conclusion

The basis for evaluation is a Programme design proposal, which was tentative as to the methods and approach for achieving the main Programme Purpose. A Problem Tree, logical framework, Programme costs and assumptions and pre-conditions were omitted. In addition, no Financing Agreement with supporting documents has been prepared along the normal EC guidelines.

Recommendation

D. RELEVANCE OF THE PROGRAMME

GENERAL CONTEXT

24. The Somaliland Economy

After years of conflict, turmoil and destruction, movements of large numbers of people (principally returnees and internally displaced persons), under-investment in the area and the recent ban by Saudi Arabia on the export of the area’s livestock it is fair to conclude that the overall economic situation of Somaliland is poor. However, it is difficult to establish the state of the Somaliland economy. The Ministry of Planning was unable to provide any overall figures, instead indicating pointers such as relevant but often outdated (1999) figures on livestock, agriculture, Government revenue and expenditure, main exports and imports, transport and communications, education, health and returnees. No real GDP per capita figures were available.

The UNDP’s Human Development Report 2001 is more comprehensive but does not break down the situation in Somaliland from the rest of Somalia. By any measure Somaliland is classified as a least-developed country and near the bottom of the international league of human development; for Somalia as a whole its ranking was 172 out of 174 countries in 1996.

Comparatively over time the figures for Somalia as a whole is indicated as:

 
Indices	1989/90 (base)	1995/97 (median)	2001
Life expectancy at birth	      46.0	42.0	47.0
Gross enrolment ratio (education)	12.9	6.6	7.5
Real GDP/capita (PPP US$)	      $836	$650	$795
GNP/capita (US$)	                  $170	-	$200 
Human Development Index (HDI)      	0.303	0.240	0.284

The figures are low by any measure but it is not clear how relevant these are to Somaliland. The GNP/capita figures in absolute terms are depressing since they indicate a daily average income of less than US$ 1/day. This is below the averages for surrounding countries, with the possible exception of Ethiopia.

However, the relative political and security stability of Somaliland compared to the rest of Somalia has given it an impetus for inward flows of money, primarily from the Somali diaspora and from the international community (EC statistics indicate that Somaliland is the largest recipient of EC funds in Somalia as a whole). There is a building boom in the main urban areas and the general mood is optimistic. Good rains in 2002 have further boosted this optimism particularly in the rural areas.

Nevertheless, the economic situation in Somaliland is dire. There is high unemployment, institutions are extremely weak, social services are basic at best weak or non-existent, and many of those previously involved in the fighting are either physically wounded and/or traumatised from their experiences.

Conclusion

Somaliland is classified as least developed with high unemployment, weak institutions, poor social services and a population trying to recover from years of war and turmoil.

25. The Political and Constitutional Context

Without repeating too much history the significant dates for Somaliland peace are the conferences of 1993 in Boorama and 1997 in Hargeisa and more importantly their outcomes. The 1993 Government differed from previous ones in that it possessed a broad base in terms of political participation and representation and the role of traditional clan leaders was institutionalised. Its task was to implement the National Charter agreed at Boorama, which defined a hybrid system of government (bicameral legislature – House of Representatives and Council of Elders, a Presidential Executive and an independent judiciary). Various other measures followed the installation of the Government including the introduction of a Somaliland currency. However, whilst the Administration was more inclusive than its predecessors not all grievances had been settled at Boorama as well as the formula for representation. In addition the peace accords at Boorama gave control of public infrastructure to the Government. However, the clan running Hargeisa airport refused to give it up. Fighting ensued in 1994, which widened and spread to Burao in 1995. There was no conclusive result but a series of inter-clan conferences culminating in a national conference in Hargeisa between December 1996 and February 1997 signalled the conclusion of the war.

The Hargeisa conference differed from Boorama in that the Hargeisa conference was organised and funded by the Government, it was larger and the existing Government was re-instated. Also a provisional constitution replaced the National Charter and stipulated the validation of the constitution by referendum, the introduction of a multi-party system and the holding of elections. Whilst the conference had its critics as an undemocratic exercise dominated by one circle in power, meanwhile the political, economic and social reconstruction of Somaliland continued, including donor support from the ‘peace dividend’.

The promised referendum on the constitution was held on May 31, 2001. The delay was caused by disagreements over the powers of the legislature and the executive, which was resolved only in the year 2000. The result was overwhelming approval. Approval of the constitution paved the way for local government elections (to be held December 15, 2002 ) to be followed by general parliamentary and presidential elections. There have been delays in this process caused by the need for laws and regulations on the formation of political organisations, electoral laws and so on. In the meanwhile the death of the President on May 5, 2002 led to the peaceful succession of the Vice President.

The present system of governance is a hybrid system that combines a clan-based system of representation with western style institutions of government. A workshop in Hargeisa in November 2000 revealed 2 schools of thought – the hybrid system should continue until Somaliland matures versus an early move to modern forms of government. The current government has moved to the latter and has begun the transition to a multi-party system with regular elections. Despite this the clan system still plays an important part in the running of government; the large executive is explained by the need for a balanced clan representation, a bloated (and ineffective) bureaucracy for the same reason. It was concluded at the 2000 workshop that public opinion in Somaliland is highly attuned to clan representation, yet often assigns the blame for the corruption of the appointments system on clan elders, rather than government officials. Blame was also put on the clan system for the discrimination against women in public life.

The current concerns can be summarised as follows:

- The unequal distribution of national resources, prevalent pre 1991, continues and the clan system is widely perceived as serving the interests of a narrow, urban based male political elite
- Many people in the non-majority (the non Isaaqs) generally feel they are inadequately represented politically
- Even within clans there is a feeling of inequitable representation, particularly those in the rural areas
- The exclusion of women from political life in the clan system, even though they are probably the majority of the population, and a strengthening and important economic force At the same time in terms of balance of government there is an unchallenged consensus on the need for decentralisation in Somaliland.

The previous experience of Somalia’s highly centralised and authoritarian regime pre 1991 has persuaded Somalis everywhere of the advantages of a more diffuse political authority and control, nowhere more so than in the North West. The authors of the 1993 National Charter and 1997 Constitution mandated the Somaliland leadership to decentralise. However, there has been remarkably little progress and many perceive the current arrangements as a perpetuation of the old; presently although there are local officials the Executive appoints them. They are assessed as performing poorly, with little independence from central government, ignoring rural areas and long absences from their duty stations.

Under the proposed changes there will be more local representation at the district and village levels but regional governors will be appointed by the Executive and will be under the jurisdiction of the Ministry of Interior. It is not clear what the financial support arrangements are going to be, how much they will rely on central government allocations, and how enshrined in law such allocations will be.

A common theme amongst Somalilanders is that the former arrangements, where Mogadishu was the centre of power, should not be substituted by Hargeisa. However, the town has rapidly emerged as the de facto hub of Somaliland – a situation that is already creating some resentment in other regions, particularly the east. Nevertheless, decentralisation has its limits since it is often seen as promoting competition between clan leaders as is the clamour for the creation of new districts.

Conclusion

The outline above gives a brief overview of the Somaliland’s political and constitutional context. There is rapid change in its governance arrangements currently being undertaken with unknown consequences. Decentralisation has started but has some way to go; local elections due for end 2002 look likely to go ahead (with EC technical support for the procedures); and multi-party democracy over a clan-based system of government is the chosen route for the country. However, economic and political concentration in and around Hargeisa is creating resentment particularly in the east.

26. Government Policies

There are no clear Government policies in the main economic areas, beyond the general objective of reducing poverty and improving the socio-economic situation of the population. There is no overall policy framework within which the Programme operates and it is difficult to identify it as a priority. This was dealt with above. At the time of Programme formulation the proposal did not attempt to justify the Programme in this context, but rather in terms of an identified problem based on a known series of historical events and commitments to secure both immediate and long-term peace. Thus, the plans of the Government are concerned mainly with combating the common threat of insecurity and pays inadequate attention to the foundations of sustainable service delivery and the productive sector.

At the time of this evaluation there are no clear Government policies relevant to the Programme as a whole. The following considerations are important:
- A Programme such as this should start on the basis of a clear outline from Government on its defence and security policies, which should lead to indicative levels of numbers of able bodied personnel to be included
- Following on from the above the Government should also have a clear policy on demobilisation, in particular its objectives as well as a set of guiding principles (following the example of the Rwanda demobilisation exercise). Of importance here is a clear set of guidelines on what to do with security services’ personnel who are unlikely to serve in an active capacity, namely veterans incapable of active duty because they are overage, war widows, and those who are physically and mentally handicapped (this latter group is a real and growing problem faced unexpectedly by the current Programme)
- Based on the above it is then necessary to determine the scope and direction of a demobilisation Programme. A primary activity to accomplish this is the carrying out of a socio-economic survey of all the security services as part of start up activities (as happened in the demobilisation project in Uganda). This will give an early indication of the direction of the re-integration support that is appropriate , subject of course to the wishes of the potential beneficiary. It should be noted that the 1999 proposal did indicate that a socio-economic survey should be carried out.

Following on from that then the other activities, which follow – awareness, sensitisation, community involvement, local capacity building and so on – are familiar as has been implemented in the present Programme.

Conclusion

Recommendation

In any future Programme there should be a clear approach and sequence of activities:
- Statements from Government as to its:
- Defence and security polices including desired numbers in each of the services – a pre-condition
- Demobilisation policies and what to do with the many who are unable to perform active duty – a pre-condition
- Agreement that one of the first activities by the Programme implementers will be the carrying out of a socio-economic survey of all the security services– a pre-condition.

27. EC’s Indicative Programming

The EC’s indicative programming can be determined from its 2002 Annual Operational Review and its Strategy for the Implementation of Special Aid to Somalia 2002 to 2007. The guiding principles for EC assistance to Somalia consists of:
- Complementing the peace dividend approach by peace building, whereby EC (and other donors) intend to support geographic areas, which achieve basic standards of governance and provision of security and to encourage similar patterns in less fortunate areas. The evolution is from humanitarian assistance towards rehabilitation which includes some development focus
- Defining long-term intervention objectives with a dual objective:
- Contribute to alleviation of poverty and support the sustainable improvement of the livelihoods of the Somali people (which the donors can influence and support)
- The promotion of a more peaceful, equitable and democratic society (which is entirely in the hands of the Somali people).

Conclusion

The EC’s intervention strategy for Somalia has the objectives of contributing to the alleviation of poverty and supporting the sustainable improvement of the livelihoods of the Somali people.

28. The DRP’s Overall Objectives

It has been stated above that there was no logical framework set out in the original Programme proposal. However, the Programme implementers prepared a logical framework early on in the implementation process, which has been amended as implementation progressed. This is set out in Annex 2. In this the overall objectives are set out as ‘peace and stability of Somaliland are consolidated and favourable conditions for sustainable and long term socio-economic development are created’. The objective is reasonably consistent with the EC strategy for the implementation of Special Aid to Somalia 2002 to 2007 (developed in 2001), which states that the overall long-term objective of the Commission strategy for Somalia is to ‘contribute to the alleviation of poverty, support the sustainable improvement of the livelihoods of the Somali people and promote a more, peaceful, equitable and democratic society’ (see sub-section 27 above). It can thus be concluded that the overall objectives of the Programme are consistent with current EC strategy.

Conclusion

The overall objectives of the Programme are consistent with current EC strategy.

SPECIFIC CONTEXT

29. Problems to be Solved

No problem tree had been prepared for the DRP and therefore it is not entirely clear that demobilisation and re-integration are priority areas for the Government. However, common sense would indicate that they are; the EC’s 2001 Annual Operational Review of the Co-operation between the People of Somalia and the EC states that the public finances of Somaliland for the year indicates a total budget of US$ 37 million of which 70% is to be spent on security. This high level of expenditure reflects the approach of the Administration to re-establishing security in the past by incorporating uncontrolled militias into formal military structures and was the answer to the conflict, which erupted in the country most recently from 1994 to 1996. Its effect is to limit the ability of the Administration to invest in anything other than salaries of the security forces. As a result it would appear that material reductions in security forces’ numbers would be a major step in improving the Government’s capability to finance other important activities for the benefit of the population.

Conclusion

It appears that no Problem Tree exercise was ever prepared. However, given that statistics indicate that around 70% of the Government’s budget is spent on the security forces it is reasonable to assume that this is a main problem for the country as a whole. In order to determine the problems related to the DRP a workshop was held on October 29, 2002 attended by senior members of the Government. The findings of the Workshop in the form of a Problem Tree are set out in Annex 3 to this Report. The Tree shows the core problem as ‘the demobilisation process is too slow’. The 3 main causes of the core problem are defined as (comments in italics):
- ‘Lack of sufficient grants (external funds) to make sustainable re-integration of the beneficiaries’ {root causes being lack of recognition of Somaliland, misevaluation of the country and its conditions and difficulties to measure the effects of (a demobilisation) project}
- ‘Lack of income earning opportunities (for beneficiaries)’ (this was not elaborated further but by imputation referred to the lack of development of the Somaliland economy and hence high unemployment thereby concluding that it was difficult to find income earning opportunities where ex-combatants can be integrated)
- ‘Delays in the re-integration process’ (see below). The last one was elaborated in some detail and the root causes defined as:
- ‘Weak institutional capacities of Government bodies’ (this is elaborated further in this Report and will not be expanded here)
- ‘Unrealistic demobilisation criteria’ (this is also outlined below, but can be summarised as criteria that were agreed between the security services and the Programme, which subsequently proved unworkable. Thus, the original intention was to demobilise people who were fit and able to work. However, this rather defeated the object of having fit and active personnel in the security services and hence contrary to the agreement subsequent nominees for demobilisation were those who were unfit for active duty – the aged, war widows, and the physically and mentally sick)
- Poor health situation of beneficiaries (this is related to the above point in that many of those in the security services are unfit for duty)
- Lack of dissemination of information and mistrust (there was a feeling at the workshop that the dissemination of information at the beginning was insufficient and that ex-combatants really didn’t understand what was being offered and the opportunities they would have under the Programme. Although the indications are that a considerable effort was put into this it is quite possible that still more was required and is a lesson for the future)
- Speculation by beneficiaries on the benefits of the DRP (this relates to the point above and reflects the perceived lack of understanding by the prospective beneficiaries, which in turn would indicate more is required to disseminate and inform them of the Programme).

It is concluded that this fairly identifies many of the problems that the Team identified independently and it is commendable that Government recognised many of the problems relate to weaknesses in its own institutions and own internal deficiencies in publicising and supporting the Programme. It also identifies lack of ability to access external funds to finance the Programme caused by the difficulties of being an unrecognised Government, and also the lack of transparency (in the security services) in being able to measure the benefits of the existing Programme. These are all valuable pointers to a future Programme if the donors agree (which itself is an identified problem).

Conclusion

The core problem identified at the workshop is that the demobilisation process is too slow, i.e. has taken too long; an effect of this is that not enough ex-combatants have been demobilised. The core problem is caused by lack of international funding, lack of income earning opportunities for beneficiaries and delays in the re-integration process.

Conclusion

Many of the root causes of problems recognise (commendably) shortcomings in the Government’s own institutions and own internal deficiencies in publicising and supporting the Programme. Conclusion There are several valuable pointers to any future demobilisation/re-integration Programme in Somaliland and the wider context of Somalia.

Recommendations

The findings of the workshop indicate that those actions, which can be modified in a new Programme are:
- The benefits of a Programme should be measurable and therefore systems put in place to allow this to happen
- The weak institutional Government bodies suggests that training and capacity building should continue (even though in the opinion of the Evaluation Team this is a problem that is almost irresolvable except in the long-term)
- Demobilisation criteria must be clear and unambiguous and applied by all parts of the security services. This may involve closer involvement of the Programme in monitoring how these criteria are applied
- More attention needs to be paid to the information and awareness campaign in the security services, with greater real time monitoring of its effects as the campaign is implemented.

30. Scope of the Problem

Numbers in the Security Services

The Evaluation Mission attempted to define the overall magnitude of the numbers in the security services. This followed a meeting with the concerned Ministries followed by written submissions as to numbers, and review of the Government’s budgets at the Ministry of Finance. The estimated numbers were subsequently discussed with government and agreed as being reasonable. However, it should be remembered that the figures are approximations only and the Evaluation Team recognises that the totals will need adjustment. However, they appear to be a reasonable working start.

Percentage of Budget Spent on the Security Services As regards the percentage of the budget spent on the security services this is set out in the table below (currency amounts in Sl Sh. millions):

Year	Security Services Costs	Total Budget - % SS Costs to Total Budget
2001	37,511	78,026	48%
2002	42,785	88,852	48%

Conclusion and Lessons from the Above Figures

A lesson to be learnt for the future would be a more proactive approach in Programme implementation as to numbers in the security services. This would involve, where circumstances require, conditionality to ensure that support is provided by the Programme in establishing the numbers of the potential target group in the security forces, and also agreement as to their desired future levels. This would give indications as to the size and scope of a proposed Programme. Another important point is to ensure that if support is given to a future investment then there should be Government commitments to abide by agreements to live within the reduced numbers of the security forces within a reasonable time frame and not to increase above such levels.

Desired Staffing Levels in the Security Services

The Evaluation Team has not attempted to request the security services their desired staffing levels at this time. However, general discussions indicated that a once and for all demobilisation could be incorporated in one investment The total costs in this case would likely be in the region of around €11 million.

Conclusion

The numbers of the potential target group in the security services have been determined but it must be stressed that the figures are at best tentative and could not be substantiated further by the Evaluation Team.

Recommendation

In any future Programme conditionality should ensure that support is provided by the Programme in establishing the numbers of the potential target group in the security forces, and also agreement as to their desired future levels. This would give an indication of the size and scope of a proposed Programme. In addition, if support is given to a future investment then there should be Government commitments to abide by agreements to live within the reduced numbers of the security forces within a reasonable time frame and not to increase above such levels. These pre-conditions are set out in Sub-section 25 above. Conclusion If the above figures are accepted then the likely cost of a future investment to demobilise all the ex-combatants would amount to a fundable amount, which is not very large (but which would have to be determined from a formulation exercise).

Recommendation

If the estimates for a final demobilisation and re-integration are close to being correct then there is a reasonable basis for a once and for all investment.

31. Problems Resolved by the Programme

The Programme has attempted to address some the problems outlined above, but must be viewed in the perspective (outlined in more detail below) of the number of beneficiaries and the type of the intervention. The likely numbers of beneficiaries will number from between 550 (lower Evaluation Team estimate) and 770 (the present number of potential beneficiaries). This is not particularly immaterial in the overall numbers to be demobilised and re-integrated in Somaliland as a whole. Whilst numbers are difficult to determine it appears that around a further several thousand security services personnel need to be demobilised/re-integrated plus a further 17,000 to 20,000 people who are members of SOOYAAL but are not in the security services.

The funding by the EC of €1.843 million was unrealistic in terms of the proposed 10,000-target set out in the GTZ proposal – the unit cost per demobilisee of €184 is far too low by around a factor of 10. Also the 2-year time frame for the Programme meant that the period was too short for the attainment of the outlined figure, even though that was the proposed timeframe set out in GTZ proposal. The result was that the Programme has become a pilot project. However, there are many positive aspects to the DRP, which need to be highlighted. The root cause (lowest levels) problems are set out below together with explanations as to if and how the Programme has affected them. They are elaborated, where relevant in the later sub-sections of the Report:

32. Beneficiaries and Actors Involved

Annex 4 sets out the organisation chart of the Programme, which helps to identify many of the actors involved. This is supported by the representatives of the many groups met by the Evaluation Team (people met and their organisations are set out in a separate Appendix but are not included in this version for the website). Thus, the actors and beneficiaries involved are as set out below:
- The beneficiaries (the ex-combatants and others)
- Those who have demobilised themselves outside of the Programme
- Those who do not want to be demobilised
- The implementers including:
- The EC offices in Nairobi and the Liaison Unit in Hargeisa
- The GTZ Team in Nairobi
- The SACB Team in Nairobi and in Hargeisa
- The NDC and EC/GTZ Team in Hargheisa
- The relevant Ministries in Hargheisa (RRR, Interior, Justice, Finance and Planning)
- The security services in Somaliland (Police, Custodial Corps and the Army)
- NGOs and operational agencies in Hargeisa (UNDP, CARITAS, SOOYAAL)
- Business community representatives (those who have benefited from the Programme)
- Local training institutions (Hargeisa Training Institute as a rehabilitated beneficiary of the Programme)
- Other interested and supporting organisations, e.g. FSAU (Hargeisa).

The above are the main actors and beneficiaries and represent those met by the Evaluation Team.

33. Other Interventions

The other donors and implementers in Somaliland are numerous. A list of these have been prepared but due to the large numbers it is probably not complete. SACB is the focal point of co-ordination. The listing indicates that the sectoral distribution in Somaliland of support is as set out below:

Sector	                         Percentage
Governance	                             36%
Education	                             24%
Health and nutrition	                 15%
Food security and rural development      10%
Water sanitation and infrastructure      10%
Others	                             5%

A notable feature of the listing in the Appendix (not included in this version) is the large number of implementers involved. The pertinent information is as below:
- +/- 120 separate interventions
- +/- 40 different donors
- 18 of the total, represent interventions that were due to finish in 2002 (it is not known if extensions were granted)
- Of the donors, the EC is the largest with 26 interventions as sole funder and a further 9 as co-funders with other donors
- There are over 40 different implementing institutions (IIs).

There is obviously a potential for duplication although examination of the each II’s write up on what they are doing seems to indicated that where interventions are location specific then there is little duplication. Where the intervention is more general then there do appear to be cases where there is overlap.

In terms of interventions that are in the same field as the DRP and are ongoing then these are extremely limited and there does not appear to be any overlap. The main other programmes are:
- Re-integration of trainees through vocational training – CARE but due to cease 8/2002
- Vocational training and small business support – CARITAS ongoing but due to cease end 2003
- Revolving loan funds to war widows – IRC through a local MFI (CSS) due to end in 11/2002

These are the direct interventions and it can be seen that they concentrate on re-integration and not demobilisation. Many of the others such as those supporting agriculture, vocational training, community development etc. are complementary to the DRP. However, there seems to be little overlap and joint action in trying to place ex-combatants in these other programmes, although to be fair this is a function of time. In any future programme this should be an area to concentrate on during implementation. It should also be noted that attempts to get the beneficiaries included in the loan schemes of CSS and another MFI (KRep) were rejected by the organisations concerned because they did not consider (in retrospect correctly) the beneficiaries likely to repay.

Conclusion

Other interventions in Somaliland are numerous – +/- 120, there are numerous IIs – over 40, the EC is the largest donor – 35 interventions and there are +/- 40 different donors.

Conclusion

The number of other demobilisation/re-integration programmes is limited; only 3 could be identified and these concentrate on re-integration and not demobilisation.

Conclusion

Recommendation

34. Relevance of the Strategy and Overall Approach

The strategic approach to the DRP encompasses both demobilisation and re-integration. There could be a third area, which is not relevant in Somaliland, and this relates to reconciliation. The country has been at relative peace for the last 6 years and it is reasonable to assume that the reconciliation process has already been implemented.

The two areas go together – a Programme that concentrates on demobilisation and ignores re-integration is storing up problems for the future. Whilst it may be relatively easy to demobilise parts of the security services, the end result will be a large number of disaffected and displaced people who may resort to creating turmoil and disturbances, when they realise that they have no visible means of supporting themselves and their dependents.

The Experience of Uganda

One approach to re-integration has been studied as a comparison to Somaliland and that is the experience of Uganda. In that programme over 36,300 veterans were demobilised over a 5-year period. The total cost was US$ 44.8 million and the unit cost per demobilisee was around US$ 1,230. The UVAB employed over 222 people at its peak (51 headquarters 171 in the Districts). The demobilisation process, handled over 3 phases, was essentially in the hands of the Army whose objective was to discharge as many soldiers as possible. The preparation time under Phase 1 was too short and at the beginning occurred without co-ordination with the UVAB. It was concluded that the co-operation aspects by the Army was poor. Most of the veterans were discharged due to old age, poor health, voluntary retrenchment, non-citizenship, and general reductions in force (which were not voluntary). Thus, selection was based on a military rationale, but was inconsistent amongst various areas. Notice to veterans was often only 1 to 2 days. The conclusion was that there was a complete lack of preparation for veterans to return to civilian life and therefore made re-integration that much more difficult. Later phases addressed some of these problems but the numbers being demobilised were considerably less than under Phase 1. Also the logistics of demobilisation were poorly planned.

Re-integration was defined in the Uganda context of transformation of soldiers to civilians and economic self-reliance of the veterans and their families (defined as providing a cash income for food, shelter, education costs of children and medical services). There is also the concept of integration into a community. The package for the veterans amounted to US$ 560 equivalent of which US$ 150 was paid as a 1st instalment, US$ 190 as a 2nd and US$ 220 for the 3rd. Some of the benefit was in kind – agricultural tools and seeds and building materials. However, a revolving fund scheme never got off the ground and therefore UVAB’s support was confined to counselling activities. Vocational training was limited and covered only a small percentage of the veterans (11%); parallel schemes from donors were also introduced on a geographic basis but these also did not cover many veterans. It is fair to conclude that this aspect of the Programme was not a success. Despite this, and it says a lot for the tolerance of the average Ugandan, there was no civil disobedience or unrest, which might have occurred in another country. On balance it is very unlikely that this approach would work and be acceptable in Somaliland.

The Somaliland Approach The overall approach and strategy in Somaliland has been different and reflects the nature of the society and a slower approach time-wise to demobilisation and re-integration. The contrast with Uganda may be best illustrated in a table:

Uganda................. Somaliland

Demobilisation
Demobilisation was involuntary
Demobilisation is voluntary and an ex-combatant can decide to back out at any time before final re-integration
Demobilisation was carried out in a short period of time and before re-integration
Demobilisation does not occur until the ex-combatant is re-integrated
It is not clear that there were any demobilisation criteria There are demobilisation criteria, although the Army in particular has chosen to renege on them
The vulnerable and infirm were a main target group for demobilisation
The fit and healthy are to be demobilised and this is spelt out in the criteria. The Army, however, has recently ignored this
For the majority of veterans there was no awareness and sensitisation prior to demobilisation
There are always awareness and sensitisation workshops prior to demobilisation. This is used as a method to attract demobilisees (however, note the criticisms spelt out in the problem definitions above)
Demobilisation was a top down approach with the Army deciding and leaving it to the UVAB to take over once a veteran is delivered to their place of re-integration
Demobilisation is led by the security services since they nominate the demobilisee. However, the process is voluntary and the ex-combatants stays in his/her unit until fully re-integrated

Re-integration

Re-integration consisted in the main of a cash package of around US$ 560/veteran paid in 3 instalments
Re-integration consists of economic re-integration through a series of economic options. None of these involve a straight cash handout but are nearly all in kind
The amount spent/demobilisee varies depending on the re-integration package the demobilisee moves into – from US$ 600 (agriculture) to US$ 1,200/demobilisee (SME)
Re-integration involved an immediate return to the area of integration
Re-integration is carried out in stages with training an important element
Vocational training was extremely limited Training is an important element of all ex-combatants
Vocational training was applied after demobilisation
Training is applied during the demobilisation process and before final re-integration
In terms of re-integration in the community there was little attempt at civic education
Civic education is an important element that is included with the general training
Once demobilised the veteran is left very much on his/her own
Because the demobilisation process is much extended re-integration occurs before demobilisation is complete
There was little attempt to work with the community where the veteran was moving to (this was needed since family ties are weaker in Uganda than Somaliland and the veteran had very often gone off fighting as a personal decision)
Community awareness is an important element of the overall package, although in the Somaliland context this appears to need less emphasis (the strong extended family, ex-combatant was asked to fight and in any case was nominated by the family. Therefore returning to the family is easier)
It is apparent from the above that the approaches taken in the 2 countries are significantly different. The Uganda process was very much a top down approach led by the desire of the Army to reduce its numbers. As a result over 36,000 veterans were demobilised over a 5-year period. In Somaliland whilst the security services nominate the demobilisees the approach is voluntary and supported by the National Demobilisation Commission. The process has been much slower and only 550 to 770 will have been demobilised over an effective 2-year period (although the available funds could not really have led to a much greater number).

Conclusion

The strategy in Somaliland has been fairly slow reflecting the unforced, voluntary nature of demobilisation and the need to socially and economically reintegrate a demobilisee before full demobilisation occurs. As a result the numbers of demobilisees will be from 550 to 770 over an effective 2-year period

Conclusion

As an alternative the Uganda process has been compared. The approach was top down led by the Army and was weakest in the area of re-integration, which was effectively ignored. Nonetheless over 36,300 veterans were demobilised over a 5-year period.

E. EFFICIENCY MEANS AND COSTS

35. Programme Costs

Programme costs are based on the figures to the end of September 2002, reflecting the agreed costs with the EC Delegation to the end of December 2001, and the incurred (to end September 2002) and estimated costs (October to December 2002) by the Programme for the year 2002. It is assumed that the costs incurred by the Programme for the year are accepted by the EC. The figures indicate the following (all amounts in €):

Description	Total	Total Spent	Estimated	Unspent  Budget	to         end 9/02	4th Qur 2002 12/02
1	Works	91,200	87,781	9,600	-6,181
2	Remuneration	651,380	599,084	73,500	-21,204
3	Travel costs	9,377	      5,158	      1,000	       3,219
4	Supplies/equipment67,169	60,432	2,000	       4,737
5	Local transport	104,503	99,776	3,000        1,727
6	Backstopping	150,750	153,000	15,488	-17,738
7	Job creation	630,000	261,871	120,000	248,129
8	Management fees	138,361	80,973	15,965	41,423
	TOTALS	      €1,842,740	€1,349,696	€240,553	€252,491

The figures indicate that at the end of December 2002, the current closing date, the unspent balance will amount to €250,000 as a minimum. On the assumption that the average levels of monthly expenditure amount to €60,000 to € 80,000 then the Programme will exhaust the EC funds in early to end April, 2003, which would include time for liquidation. It would also provide some additional time for securing funds for the follow on Programme if that was agreed.

Conclusion

On the assumption that the average levels of monthly expenditure amount to €60,000 to € 80,000 the Programme’s no-cost extension should be to end April, 2003.

36. Financing

The figures available to the Evaluation Team indicated a total cost, which is entirely financed by the EC. No indication is given of the Somaliland Government contribution. This will consist of the costs of running NDC, and salaries paid to ex-combatants while they are in the process of demobilising and re-integrating. In terms of out of pocket costs, this amounts to around €320,000. It could be argued that a large part of the Government costs (those relating to the salaries of ex-combatants) are non-incremental since they would be incurred in any case. However, if this is not assumed, then the total cost of the Programme is estimated at around €2.16 million.

Conclusion There is no calculation of Government contribution to the Programme. Estimates indicate out of pocket cost of €320,000 giving a total Programme cost of €2.16 million.

Recommendation

In any future Programme design Government and other local contributions should be included in Programme costs.

37. Value for Money

Working on the assumption that the full amount of the EC grant (€1.84 million) is spent then the number of beneficiaries is going to be between 550 (lower Team estimate) and 770 if the total number of potential beneficiaries is realised. On this basis the unit cost/beneficiary amounts to €3,350 at the lower beneficiary number level and €2,400 if the higher number of beneficiaries is reached. It is the Team’s conclusions that at the very least the cost/beneficiary will amount to +/- €3,000 as a minimum. It should be noted that had the original number of beneficiaries been achieved (900) then the unit cost would have been €2,000, which is a more reasonable figure of unit cost. The actual figure is high and reflects the interruptions and difficulties faced by the implementers, in particular the late reluctance of the Army to provide names of more demobilisees.

Comparisons with other programmes show that Uganda demobilisation costs were around US$ 1,230/demobilisee including administration, whilst Rwanda was around US$ 2,000 calculated on the same basis. However, neither were pilot programmes and undoubtedly there are economies of scale as numbers of demobilisees increase.

Conclusion The unit costs of the Programme are high – depending on final numbers from €2,400 to €3,350 – and this reflects the shortfall of the Programme in achieving the original 900 demobilisees. The reasons for this are primarily the interruptions and difficulties faced by the implementers, in particular the late reluctance of the Army to provide names of more demobilisees.

ORGANISATION, MANAGEMENT AND MONITORING

38. General Organisational Set Up

Organisational arrangements were specified in the original Programme proposal, although not in the required detail. No organisation chart was attached. However, the general outline indicated that the ownership of the Programme should be entrusted to the National Demobilisation Commission (NDC), with a Programme office in Hargeisa. Proposed staffing including technical assistance was modest in numbers.

NDC Staffing and Budget

Annex 4 indicates the staffing at for the Programme as a whole. There are several points of note related to that organisation:
- The organisation chart does not reflect the Departmental organisational arrangements of NDC
- Three of the eight professionals in NDC also work most of the time for the EC/GTZ Team. This is to provide on the job training and more importantly to supplement their Government salaries through separate contracts so that they do not leave. At present Government wage rates it is unlikely they would stay since Government pay rates do not provide a living wage
- A tour of the offices of NDC reveals that many people do not even bother to come to work, except to collect their monthly pay. Again this reflects the low rates of Government pay.

The organisation chart (see next page) sets out the organisational arrangements for NDC. These indicate that the organisation is heavily weighted towards administration (50% or 17 of the 34 staff) with very little staffing in the Demobilisation and Re-integration Department. (4 of the 34 staff), which surely should be larger reflecting the operational aspects of the Commission’s work.

As will be seen below NDC’s role in the Programme vis a vis the role of the EC/GTZ Team is that whilst NDC may be the nominal owner of the Programme as originally envisaged, in effect it has little involvement in day-to-day implementation. Even the senior staff, who have been seconded and shown to be the most capable spend most of their time working with the EC/GTZ Team rather than in NDC, which is physically next door. Examination of the budget for NDC indicates that Government’s actual allocation of funds in 2002 to the Commission will amount to US$ 24,000 equivalent. This is insufficient with the result that the EC makes a contribution of US$ 1,200 a month towards the Commission’s running costs. Thus, the total costs of NDC in 2002 will amount to less than US$ 40,000.

The causes of many NDC’s weaknesses relate to its budget and funding. This was briefly mentioned above but a breakdown of the budget for 2003 indicates the following:

Description	          S/l Shilling	US$ 
Staff salaries	    3,192,000	515
Staff bonuses	    8,500,000	1,371
Staff incentives	    4,800,000	774
Travelling allowances 6,000,000	968
Unpaid salaries       18,400,000	2,968
New vehicles	    100,500,000	16,210
Vehicle & office rent 11,160,000	1,800
Fuel & lubrication    30,256,000	4,880
Vehicle spare parts   18,650,000	3,008
Office stationery	    16,300,000	2,629
Printing & photocopying	10,500,000	1,694
Office equipment 	    30,375,000	4,899
Office equipment repairs 3,100,000	500
Utilities	          6,000,000	968
Communications	    3,725,000	601
Postage	          3,300,000	532
Staff training 	    1,500,000	242
Miscellaneous expenses 8,040,000	1,297
TOTAL                 284,298,000	45,855

The figures make depressing reading. The US$ equivalent is based on the free market exchange rate of Sl Sh. 6,200 to US$ 1. The figure outlined above corresponds to the US$ equivalent given to the Team by NDC management. Even if this budget represented funds received by NDC, which is most unlikely, there is no way that this could be translated into a viable implementation body. This is obviously inadequate to function at a level, which will allow effective implementation of the Programme. A main area of concern is salary levels, which are below a living wage. In discussions with the Minister of Resettlement, Rehabilitation and Reconstruction (RRR) it became apparent that top ups of any kind, let alone to a living wage (a multiple of many times the existing levels) are not acceptable to Government and therefore this is not an option.

However, on the positive side NDC’s management have provided considerable political support and back up when implementation bottlenecks have been faced, and they have been involved in decision making at the policy and major implementation levels.

Comment on NDC’s Sustainability

Given the above analysis there is a real question as to the sustainability of NDC except working in partnership with a donor such as the EC/GTZ Team. This leads to the conclusion that in any future programme either a different way or approach will have to be devised to avoid the duality of the present arrangements, or else it is accepted that this situation will continue. The latter approach is acceptable if it is considered that demobilisation is a finite activity for a government, unlike education for example, which in ongoing, and that achievement of the end result would result in the closure of the government body implementing this activity. If such a view is accepted then it is possible to adopt an implementation approach, which achieves the end result as quickly and efficiently as possible; just such an approach is recommended.

Conclusion

NDC is the owner of the Programme at the policy making and major implementation levels; the day-to-day running of the Programme is effectively in the hands of the EC/GTZ Team. Given this and the analysed weaknesses of NDC there is a real question as to the sustainability of the Commission except working in partnership with an implementer.

Recommendation

In any future programme either a different way or approach will have to be devised to avoid the duality of the present arrangements or else the status quo is accepted so as to achieve demobilisation as quickly and efficiently as possible. This would be followed by the demise or re-defining of the role of the Government organisation concerned. This approach is preferred and on the basis of the present estimates of required numbers of demobilisees this is feasible. Organisational Arrangements in the EC/GTZ Team The EC/GTZ Team is part of NDC, and it was observed that they are physically located together and it was observed that there is close co-operation. An important part of the routine is a daily staff meeting held in NDC offices in which problems, actions and goals are discussed. This brings in NDC staff (many of whom work in the EC/GTZ Team anyway) and above all NDC management, who are fully cognisant of what is going on at the operational level.

As the Programme evolved during implementation, different needs were identified and the organisational structure of the EC/GTZ Team at the time of evaluation is as set out in Annex 4. This shows the relationship amongst the various implementers and the detailed organisational arrangements within the EC/GTZ Team. The staffing numbers vary greatly from those originally envisaged, whilst the various departments reflect the activities as they have evolved from the needs of the Programme. It is the conclusion of the Team that the organisational arrangements reflect the activities carried out and that the staffing in general is appropriate for the implementation of such activities. In addition, the quality of the staff is adequate to carry out the required functions, including those seconded on a part time basis from NDC.

Conclusion

Role of NDC in Government

NDC was established in 1993 by decree of the SNM Central Committee (then the country’s legislative body) to carry out the last of the 3 declarations made at the 1991 national conference – ‘to reward SNM veterans with special benefits in recognition of their struggle and sacrifices for the liberation and independence of Somaliland’. One of its first acts was to oversee the voluntary demobilisation of the assembled motorised clan militia and heavy arms in early 1994, primarily in Hargeisa and Burao. Its primary function over the years has been in demobilisation with re-integration occurring only when supported by donors.

It is not clear that NDC has any written legal status. Many actions such as this were never properly decreed in the sense of being written down and it is believed that this applies to the Commission. Nonetheless, it does have a Minister to report to – RRR – and this is reasonably well established. It will be a concern for the future that there is some formalisation as to NDC’s establishment.

Conclusion

It is not clear that NDC, established in 1993, has any written legal status. Nonetheless, it does have a Minister to report to – RRR – and this is reasonably well established.

Recommendation

It will be a concern for the future that there is some formalisation as to NDC’s establishment and this should be implemented.

Legal Relationship for the Programme

The absence of a legally recognised Government has meant that the EC has effectively passed funds to GTZ, the implementer, who in turn has signed a Memorandum of Understanding (MOU) with the Somaliland Ministry of Planning. This document was signed on April 26, 2000. It is a short, less than 3 page document, which is remarkably short as to detail. The Project objective describes the aim as ‘assisting in the administration to demobilise enlisted soldiers, with a view to encouraging continued peace and development in Somaliland’. It is not necessary to belabour the point but the description is only about demobilisation, not re-integration. It talks about enlisted soldiers without defining which services they are in – for practical purposes the security services (Police, Army and Custodial Corps) are in fact included. There is talk about an early survey and dialogue leading to the production of a document to be used as the basis for implementation. It is not clear what this document looked like but it is assumed to be the ‘Operation Plan’ set out in the form of a logical framework.

The Specific Conditions are primarily boiler-plate clauses giving legal exemption to GTZ, provision of security for GTZ personnel etc. Those of interest to the implementation of the Programme (comments in italics) include to:

- Provide 2 Government buildings for renovation of offices – one in Hargeisa and one in Burao – it is apparent that implementation is centred in fact around Hargeisa and a recommendation above is that in the future more attention should be paid to the east of Somaliland
- Continue to pay all participating members of national forces during demobilisation – this the Government has done
- Select a contract person/co-ordinator to support the Programme Manager in dealings with Government – this is presumably the Chairman of NDC and this has been fulfilled
- Assist in the design and follow-up of an accurate and transparent monitoring and evaluation system against which the project’s progress and activities can be measured – this has not been done, working on the assumption that this refers to monitoring and evaluation in the security services. It is recommended below that such a system be a pre-condition in any future programme
- Ensure that the size of the national defence force (not defined) during the period of the programme remains unchanged at a level of ?? troops. The number is not specified and it would be hoped that there would be reduction in the national defence force. This provision has not been met
- GTZ’s commitment is that no former project manager of any previous GTZ project will play any role in the demobilisation programme. This is an interesting provision since it would be thought previous experience in any country would have been preferable. It is thought that this relates to the GTZ Project Director in place in 1994 to 1996 who it is felt conducted himself in manner, which offended the local administration.

Conclusion In general the terms set out in the MOU between Somaliland’s Ministry of Planning and GTZ have been met. Exceptions include the lack of staffing of an office in Burao, no monitoring and evaluation system in the security services, and no defined numbers for the security services.

39. Plan of Operations and Timetable

Under the present arrangements the Programme is due to close at the end of December 2002. As stated before some of the activities are unlikely to be completed and the recommendation has been made in principle that a third no-cost extension should be applied for and granted. This will also allow time for the commitment of new funds if it is decided to go ahead with a new Programme directed towards a follow on demobilisation investment in Somaliland. On the basis of the calculations set out in Sub-section 35 above the indications are that, with a re-allocation of funds, the Programme should now be scheduled to close by end April 2003.

Recommendation

A third no-cost extension should be granted with a scheduled closing date of end April 2003.

40. Financial Management Matters and Procurement Procedures

Financial Matters

Given the limi

Somalia N.W. Demobilisation and Re-integration Programme (DRP) Evaluation Evaluation Report

Somalia N.W. Demobilisation and Re-integration Programme
(DRP) Evaluation Evaluation Report