TE: Mainstreaming SLM in Rangeland Areas of Ngamiland District for improved Livelihoods

Report Cover Image
Evaluation Plan:
2017-2021, Botswana
Evaluation Type:
Final Project
Planned End Date:
11/2019
Completion Date:
08/2019
Status:
Completed
Management Response:
No
Evaluation Budget(US $):
23,400

Share

Document Type Language Size Status Downloads
Download document INDIVIDUAL_CONSULTANT_PROCUREMENT_NOTICE_SLM TERMINAL EVALUATION.pdf tor English 458.58 KB Posted 254
Download document SLM Ngamiland_TE Report 2019.pdf report English 2978.61 KB Posted 360
Download document Terminal Evaluation report clearance form PIMS 4629.pdf related-document English 219.79 KB Posted 269
Title TE: Mainstreaming SLM in Rangeland Areas of Ngamiland District for improved Livelihoods
Atlas Project Number: 00077645
Evaluation Plan: 2017-2021, Botswana
Evaluation Type: Final Project
Status: Completed
Completion Date: 08/2019
Planned End Date: 11/2019
Management Response: Yes
UNDP Signature Solution:
  • 1. Sustainable
Corporate Outcome and Output (UNDP Strategic Plan 2018-2021)
  • 1. Output 1.4.1 Solutions scaled up for sustainable management of natural resources, including sustainable commodities and green and inclusive value chains
SDG Goal
  • Goal 1. End poverty in all its forms everywhere
  • Goal 15. Protect, restore and promote sustainable use of terrestrial ecosystems, sustainably manage forests, combat desertification, and halt and reverse land degradation and halt biodiversity loss
SDG Target
  • 1.5 By 2030, build the resilience of the poor and those in vulnerable situations and reduce their exposure and vulnerability to climate-related extreme events and other economic, social and environmental shocks and disasters
  • 1.a Ensure significant mobilization of resources from a variety of sources, including through enhanced development cooperation, in order to provide adequate and predictable means for developing countries, in particular least developed countries, to implement programmes and policies to end poverty in all its dimensions
  • 15.3 By 2030, combat desertification, restore degraded land and soil, including land affected by desertification, drought and floods, and strive to achieve a land degradation-neutral world
Evaluation Budget(US $): 23,400
Source of Funding: 00077645
Evaluation Expenditure(US $): 20,511
Joint Programme: No
Joint Evaluation: No
Evaluation Team members:
Name Title Email Nationality
Vincent Lefebvre Evaluator lefebrevinc@gmail.com
GEF Evaluation: Yes
GEF Project Title: Mainstreaming SLM in Rangeland Areas of Ngamiland District Productive Landscapes for Improved livelihoods
Evaluation Type: Terminal Evaluation
Focal Area: Land Degradation
Project Type: FSP
GEF Phase: GEF-5
GEF Project ID: 4751
PIMS Number: 4629
Key Stakeholders: Department of Environmental Affairs, Department of Forestry and Range Resources, Department of Animal Production, North West District Council, NGOs, Private Sector
Countries: BOTSWANA
Lessons
Findings
1.

3. Findings

3.1 Project design / Formulation

The project’s objective was to mainstream sustainable land management (SLM) in rangeland areas of Ngamiland district productive landscapes for improved livelihoods. The project was built through consensus on a series of sectoral district priorities and key issues to be addressed such as fire control, sub-optimized livestock management, grazing land degradation and bush encroachment, low price of beef and a lack of higher value markets for livestock products as well as improved agricultural practices. This resulted in the design of a project consisting of a collection of results and brought together under the SLM concept. These would be achieved by adopting a pilot approach as it would have been difficult to mainstream all activities into a unified integrated development project. Therefore, the project opted for piloting SLM with different results in specific geographical setups(Haina veld area for commercial ranches, Lake Ngami for communal rangeland management and bush eradication, Tsodilo enclave for fire control – see Annexe 9 for project’s map) with some global results at district level (mainly from outcome 2 on beef marketing and cattle off-take). This is an effective approach for development piloting, by isolating geographically each solution to be tested as a way to assess its effectiveness. These interventions, however, do not necessarily represent real development conditions: this may be an issue as external factors not covered by the project may reduce the effects of project results. In that context, results scattering across large areas may prove also to be an issue for impact and even further for replication.

The Programme was to be implemented through UNDP’s National Execution Modality (NEX), with the MEWT serving as the designated national executing agency (“Implementing Partner”) of the project.

3.1.1 Theory of Change

The GEF IEO (2017) Guidelines for conducting terminal evaluations require that the project’s Theory of Change (ToC) should be described as part of the analysis of project design; where a project did not have an explicit ToC, the evaluator should develop one based on information provided during the evaluation. At the time of project design in 2013, the TOC approach was not a requirement for UNDP/GEF’s project formulation. Hence, the usual approach in project formulation was used based on the logical framework methodology. The logical framework analysis is an objective-oriented tool to project planning. The analysis identifies a problem then develops a "temporal logic model" that runs through a pathway to achieve an objective from inputs, activities, results to outcomes that ultimately contribute to a development objective. It also identifies risks and assumptions and indicators and targets to assess the project’s performance. A theory of change is a method9 that explains how a given intervention is expected to lead to specific development change, drawing on a causal analysis based on available evidence. It helps identify the many underlying and root causes of development issues so as to determine what priorities should be addressed to maximise a project’s contribution to achieving development change. By articulating the causes of a development issue, making assumptions explicit on how the proposed strategy is expected to yield results, and testing these assumptions against evidence, the theory of change helps ensure a sound logic for achieving project change. 

At the core of the Theory of Change is the understanding how the activities of the intervention are expected to lead to the desired results through identifying (i) the causal pathway from activities to outputs to a sequence of outcomes to impacts and (ii) the causal assumptions showing why and under what conditions the various links in the causal pathway are expected to work. Reconstructing a ToC from the project itself can be challenging, in particular when the project design is the result of the merger of different Government district priority interventions (as explained in subheading 2.1) as basically, a ToC could be formulated for each project output. In that context, the evaluator used the approach suggested in ‘Useful Theory of Change Models’ (Mayne 2015)10 under which a ToC is developed by first reconstructing the impact pathways - usually a simplified intervention logic with an emphasis on linkages (‘causal links’) between activities and results (‘impact pathways’) contributing to the overall objective, and second, adding assumptions into the impact pathways as ‘causal assumptions’. The ToC of the project is located under Annexe 8.


Tag: Natural Resouce management Global Environment Facility fund Implementation Modality Knowledge management Programme/Project Design Results-Based Management Theory of Change Jobs and Livelihoods

2.

3.1.3 Assumptions and risks

An analysis of the risks and their mitigation measures is presented in Table 4. The risks identified in the ProDoc are mostly relevant, but there are some overlooked assumptions that significantly affected the project delivery; partly because they were not addressed in the project design:

- Poor Trust governance: the project relied extensively on existing Trusts so as to target communal farmers; these trusts originate from the CBNRM policy that provides guidance on CBOs; however, it lacks an act and regulations that would professionalise management and minimise governance issues; this was not addressed in the project under component 2 and is more evidence on the lack of an ‘enabling environment’ component. The poor management and governance of collaborating Trusts proved a significant issue for project delivery and above all the impact and sustainability of several results (e.g. fire management mechanism, charcoal production). - Most cofinancing was in-kind but not necessarily output-based and it was assumed that Government counterparts would have their own budget funds to fully participate in the project implementation. This proved to be an issue with limited funding resulting in limited participation in project activities for some district offices. - As mentioned before, a significant risk for the project would have been the human/wildlife conflict; indirectly the project addressed this issue in securing the Government quarantine fence in the wider context of the CBT introduction in the district.

3.1.4 Lessons learned from other projects incorporated into project design

Very little information is available in the PRODOC on lessons learned from other projects feeding into the intervention’s design; all the same for Government policy on SLM. The main sector policy documents do not address SLM as such. The SLM concept seems to be new in Botswana and a number of new policies and interventions are building upon this particular project (see the chapter on ‘Potential impact’). Furthermore, the review of GEF projects, worldwide, showed that while there are numerous examples of SLM mainstreaming projects, few if any actually focus on rangeland and livestock. Most address crop agriculture. Nonetheless, for output 1.1, the project has built upon experience gained from the pre-existing Southern Africa Regional Environment Programme (SAREP) on zoning using a participatory land- use planning process.

3.1.5 Planned stakeholders’ participation

The core stakeholders of the project in addition to the final beneficiaries (rangeland farmers and villages’ communities) were DFRR, DEA, DNMM, several community trusts and farmers’ associations, DCP, DVS, DAP, DWNP, ONMM, TLB, NCONGO, ORI, BMC. MEWT was responsible for the overall project implementation; the actual project execution was led by DFRR from MEWT in close collaboration with DAP of the MoA. The project was very slow in securing stakeholders buy-in and numerous visits and meetings were conducted at project’s start-up and further afterwards to ensure adequate participation and collaboration in project’s activities. The planned stakeholders and an estimate of their actual contribution to the project were very comprehensive as per PRODOC’s stakeholders’ analysis (see Table 5 below).


Tag: Natural Resouce management Global Environment Facility fund Government Cost-sharing Programme/Project Design Results-Based Management Risk Management Strategic Positioning

3.

3.1.8 Linkages between the project and interventions within the sector

The project initially had planned to seek partnerships with stakeholders that would bring added value to the results; these included the following:

- The Rural Innovation Centre identifying the appropriate technology and possibly training users on such technology for processing wood products into briquettes

- The Local Enterprise Authority (LEA) training community groups on basic business management, marketing and book-keeping.

- The Social and Community Development Council being involved to mobilize the participating community group to form and empower leadership members who would be trained on basic organizational leadership such as conducting meetings, record keeping and reporting as well as conflict resolution.

There were a number of projects addressing key natural resource management challenges in Ngamiland District, possibly providing opportunities for complementarities and building of synergies. These included the following:

- The Human-Wildlife-Coexistence Management Project in Northern Botswana implemented by the Department of Wildlife and National Parks, in partnership with the World Bank with the Seronga area within Ngamiland with the prospects of piloting conservation agriculture and open game farming at the community level

- The USAID- f u n d e d SAREP aiming at assisti n g the Countries of Botswana, Namibia and Angola to effectively manage the resources of the Okavango River Basin with the prospects of (i) facilitating the implementation of the Ngamiland Integrated Land Use Plan and (ii) assisting in the formulation of a Strategic Environment Assessment for Ngamiland that takes into account aspects of SLM; it was also anticipated that SAREP would further work with the Ministry of Agriculture to explore alternative investments for SLM such as REDD+.

- The GEF-funded Biokavango project on building local capacity for the conservation of biodiversity in the Okavango Delta and focusing on strengthening tourism, fisheries and sustainability of veld products as livelihood support systems. The project’s Sustainable Land Management initiatives would utilize the systems and processes initiated by Biokavango project.

- The REDD+ pilot project taking place in NG 8 controlled hunting area within Ngamiland District complementing the SLM project through the protection of rangeland areas, monitoring and releasing benefits from such resources. Through this initiative, the SLM project benefitted from ORI’s expertise in establishing a resource monitoring system; it also contributed indirectly through a study on ecosystem services around Lake Ngami (book published in 201812).


Tag: Oversight Partnership Programme Synergy Project and Programme management Country Government Capacity Building

4.

3.2 Project implementation 

3.2.1 Adaptive management

The project under the NIM modality was due to be implemented from March 2014 to March 2019: while the initial PMU recruitment was swift (around 6 months), there were issues that impeded significantly the actual operationalization of the project back at project’s start-up: 

(i) There was no office space within DFRR Maun and a compound had to be built from scratch (inaugurated in 2015) (ii) The Project Coordinator previously employed by UNDP Gaborone remained working from the capital for over 6 months with regular commuting to Maun (from 09/2014 to 02/2015); this raised project management costs. (iii) AWP focussed nearly on consultancies and studies but little groundwork with communities These two points (ii) & (iii) did not facilitate stakeholders’ confidence-building into the project, that in turn resulted in slow implementation (iv) Procurement centralised from UNDP Gaborone, that made the acquisition process cumbersome (v) Early mismanagement claims involving the Project Coordinator resulted in setting up a procedure for his removal, culminating with his departure in 11/2015. By the time of the MTR (mid-2016), a substantial amount of funds had been devoted to project management, consultancies but little on-site activities as per 2016 PSC meeting.

By the end of 2016, decisive actions had been made to speed up implementation: - The Project Coordinator had been replaced (from 09/2016 onwards), hence a 9 months gap without project coordination - Procurement was decentralised to the Admin & Finance Officer, as per 2016 audit recommendations - A plan for implementation acceleration had been formulated (already by 2015) - The Technical Reference Group with members from all major stakeholders, acting as an advisory board, reviewing AWP and settling outstanding technical issues, was reinstated and fully-functional. Adaptive management was used wisely for activities not necessarily in the PRODOC but contributing to SLM and modernisation of rangeland management: - With the introduction of BAITS in Ngamiland in 2016, the project assisted Government in demonstration and training for ear tags in replacement of cattle branding. - Tablets were provided to the population during a major beef measles outbreak in the region of Shakawe. - Assistance with chemicals was provided during staple fly and tick outbreaks. - Communication and social events focussing on activities in the project area (e.g. lake Ngami boat competition and Tsodilo Hills bushwalk and cultural night) boosted the project foothold in the surrounding communities and overall in the district. - The project accompanied the introduction of the CBT mechanism in Ngamiland with workshop participation and the partial funding of the rehabilitation of the Government quarantine (buffalo cables and water tanks) at Makalamabedi. - When the baseline was established in 2017, the optimum livestock stocking rate was changed from 1 head/16ha to 1 head/ha (with HLMA) - ‘Control’ farms (under output 1.2) would not be set up from scratch but collaboration sought with existing (commercial) ranchers instead .

This resulted in a very effective project delivery with over 95% delivery rate by March 2019 (project closure).The overall focus of the project (project goal, objective, and outcomes) remained unchanged over the whole project period.


Tag: Communication Human and Financial resources Implementation Modality Monitoring and Evaluation Partnership Project and Programme management

5.

3.2.5 Project finance

As per CDRs and UNDP’s co-financing estimate, the total cost of the project (including Q1 2019) from 2014 to 2019 is explained under Table 6, evidencing an approximate 400% co-financing ratio. The co-financing of two NGOs as well as SAREP’s never materialised while cofinancing from other stakeholders was one or two orders of magnitude lower except for MoA 3 times higher. Table 7 shows that the project initial operationalization was spread over 3 years (2014/5/6) (very low delivery rate) which is unusually long for GEF projects. Despite these low delivery rates, there was no request of extension at MTR stage. The increase in the delivery rate coincided with the arrival of the 2nd Project Coordinator in Q4 2016.

The analysis of the cumulative delivery rate (see Figure 2) show an S-shaped curve (‘effective’) against an asymptotic curve (‘planned’) for the cumulative spending as anticipated at formulation stage; this is evidence for the need to take into account an extended inception phase to resolve operationalization difficulties like recruitment and initial involvement of all stakeholders, and to lengthen substantially the project cycle to ensure a smoother implementation. 

No information was available on planned and committed budgets per outcome, hence it is not possible to estimate the project management budget share. The MTR had evidenced very high project management cost overruns for 2014, 2015 and 2016 (up to 300%). Overall, the project management budget was supposedly contained within the planned envelope despite its very low amount (0.145M$ or less than 5%); more common values for similar projects have a much higher project management budget (10-15%). 


Tag: Efficiency Global Environment Facility fund Monitoring and Evaluation Procurement Project and Programme management Civil Societies and NGOs

6.

3.2.7 UNDP and Implementing Partner implementation/execution coordination and operational issues

Both UNDP and the implementing partner (DFRR under MEWT) were involved in project implementation with UNDP having an advisory role with the provision of technical advice and monitoring. 

Implementing Partner:

The project was supervised by MEWT. DFRR hosted the project team (PMU) throughout the duration of the project and showed a strong commitment to and ownership of the project throughout its lifespan. It is worth noting that the MEWT Permanent Secretary was committed to this project and ensured that lessons learned would be passed at the national level (resulting in the adoption of some project results and the formulation of new interventions [see impact]). The GEF operational team under MEWT was quick to build on strengths and weaknesses of the project to review future GEF interventions design – in particular, related to governance (see recommendations) -. Quality of implementing partner execution RATING: Highly Satisfactory (HS)


Tag: Global Environment Facility fund Human and Financial resources Implementation Modality Monitoring and Evaluation Ownership Technical Support

7.

3.3 Project results

3.3.1 Overall results

A brief assessment of the project overall results (as per PIR), is presented in the following paragraphs. Project Objective: mainstream SLM in rangeland areas of Ngamiland District productive landscapes for improved livelihoods. Progress at project’s end: as the objective is relatively vague, one can assume that it is somewhat achieved: both final and institutional beneficiaries were exposed to a wide range of SLM practices. In relation to the actual quantitative target, 1.000.000 ha under SLM, 839.000 ha were covered by the project. This may be impressive but does not mean that all the area is now under SLM, just that the project did target this global area for SLM practices. Out of that result, 88% is actually referring to improved bush fire management (area under community surveillance). The remaining 12% covers livestock grazing improvement. In any case, the target was too ambitious in relation to the budget and remaining time, given that the project did really accelerate implementation only by MTR time, 2 ½ years after PRODOC signature.

3.3.1.1 Outcome 1: Effective range management improves range condition and flow of ecosystem services to support the livelihoods of local communities

Progress at project’s end: partially achieved. There is overall increased awareness about new rangeland management approaches and much interest in income-generating activities through charcoal production. 


Tag: Natural Resouce management Women's Empowerment Results-Based Management Capacity Building Jobs and Livelihoods

8.

3.3 Project results

Output 1.3: Bush-control programme is piloted and provides financial incentives for controlled bush clearance

With livestock overgrazing, grass is substituted by bush with a resulting reduction of quantity and quality of grazing area; bush removal is a solution to restoring quality grazing land combined with charcoal production: training was delivered on charcoal production, bush clearance and wood cutting machines benefitting mostly Lake Ngami community trust members (and two other trusts with more limited results) with benchmarking to Namibia on biomass production. Equipment was provided for cutting wood and charcoal production (drum kilns) to 15 villages. All these were centralised by the Lake Ngami Trust. Livestock feed from the bush was also planned but cancelled: the hammer mill for feed production was expensive and additional funding would have been required to analyse a future feed value chain in the district.  


Tag: Natural Resouce management Technical Support

9.

3.3.1.2 Outcome 2: Effective resource governance frameworks and markets provide incentives for livestock off-take and compliance with SLM

Progress by project’s end: achieved. The project contributed directly to establishing an effective governance framework for SLM; as for markets and increased livestock off-take, the project contribution was real but somewhat small in relation to the effects of BMC’s cofinancing (slaughter increased from 19.000 to 25.000 cattle in 2016 – source: BMC) and the current drought that has significantly increased off-take, offset by FMD outbreaks. Overall, the project developed quite a number of initiatives that did contribute or have the potential to increasing off-take as well as raise livestock farmers’ income.

Output 2.1: A regional multi-stakeholder forum for facilitating a dialogue on SLM and mainstreaming

SLM into regional and national policy programs and processes is created and empowered. Instead of creating a new forum (the PRODOC had explicitly reported that there is already a large number of coordination initiatives led main many Government agencies), the project opted for mainstreaming SLM into the preexisting Okavango Delta Wetland Management Committee that was reconfigured and empowered to serve as the multi-stakeholder forum for SLM. This proved to be the right decision with the forum now serving as a platform for discussing SLM mainstreaming and divulgation into both Government institutions and the civil society. Sessions held have resulted in better inter-ministerial dialogue, coordination and action on SLM (e.g. lobbying for reducing human-wildlife conflict through hunting ban removal, a renewed collaboration between DAP and the Ministry of Health on livestock measles campaigns [awareness campaigns on tablets, pit latrines…], increased involvement of DVS with selected project communities and commercial ranchers, enhanced/increased M&E by DCP on conservation agriculture farmers and by DFRR on HLMA, routine communication from MoA to DEA in case of potential environmental issues…).

However, the effective bargaining power of the forum remained quite limited with the ability to conduct dialogue, discussions or even report on SLM mainstreaming options but little leverage to officially mainstream SLM into existing policies. Notwithstanding these limits, there is evidence that the central Government has been thoroughly monitoring the project with a number of new national initiatives based on the project’s results (see impact). It remains to be seen whether the forum itself or a lead institution member will continue advocating for SLM through it by project’s end.


Tag: Natural Resouce management Effectiveness Women's Empowerment Integration Civil Societies and NGOs Country Government

10.

3.3.2 Relevance

The project is highly relevant to the main objectives of the national environmental and development priorities, UNDP priorities and global goals and GEF 5 Land Degradation Focal Area. Relevance to national environment and development priorities: informants indicated that land degradation due to overgrazing and overall poor management practices is a recurrent topic of discussion at Government level and by most sector stakeholders, together with the FMD and human-wildlife conflict issues. The project was aligned with (i) the NDP 10 (2009-2016), in particular, the pillar on sustainable use of natural (rangeland) resources including several strategies such as beef industry VC approach, small livestock development, wider use of ear tags, improving range management and adoption of good practices, (ii) the NDP 11 (2017-2023) with a focus on emerging issues such as land management and unsustainable agricultural practices, (iii) Vision 2036 under pillar 3 on sustainable environment, in particular ecosystem functions and services and pillar 1 on sustainable economic development in the agricultural sector with VC development and a disease-free agricultural sector.


Tag: Natural Resouce management Relevance Global Environment Facility fund Country Government

11.

3.3.3 Effectiveness and efficiency

Effectiveness (contribution of the actual outcomes to the project objective): The project objective was to mainstream SLM in rangeland areas.

Two main outcomes (components) were formulated:  (i) Outcome 1: effective range management in over one million ha improving range condition and flow of ecosystem services to support the livelihoods of local communities in Ngamiland (ii) Outcome 2: effective resource governance frameworks and markets providing incentives for livestock off-take and compliance with SLM

Outcome 1 results: the actual contribution of the outcome depends on each particular outputs as the project was geographically segmented; overall the project has very successfully created awareness on SLM amongst all stakeholders

Rangeland management: the efforts made by the project on improving rangeland management practices have little contributed to the outcome: while a number of commercial ranchers participated in the project, several techniques from demonstrated HLMA were abandoned and there is little evidence of widespread adoption by other farmers according to most informants. Although trainings were provided to communal livestock farmers, no actual HLMA demonstration activities were conducted on communal lands; this is partly due to the insufficient time before project closure and to the investments necessary to set up a participative communal ranch that would require extensive and lengthy dialogues between community members on costs, organisation, and responsibilities. At this stage and despite high enthusiasm, HLMA is not prevailing amongst exposed farmers. At institutional level, the situation is different with HLMA appropriation both within MEWT and MoA and further discussions on how to switch from demonstration to replication. This seems to be a bit hasty as final beneficiaries insisted during interviews on proper demonstration in different environmental and social contexts (commercial, communal conditions, fenced, not fenced, nearby game reserves or not…)

Conservation agriculture: the adoption of conservation agriculture, despite failures because of recurrent droughts, seems to be high as it generates both environmental benefits and economic interest. It remains to be seen whether there is an economic case for conservation agriculture as it is currently heavily Government-subsidized. 

Fire management: the project support in fire management has been very successful and the decrease in the extent of fires is spectacular, directly contributing to land degradation reduction in the Tsodilo area. Awareness-raising has been very effective as the reduction in induced fires extent results more from behavioural community members change than extensive use of donated fire equipment.

Bush eradication and charcoal production: despite the production issues, the activity has the potential to contribute both to land degradation reduction and to community income generation. However, so far, community members haveshown little interest, possibly because the Trust in charge of charcoal production has little capability to switch from a development project support approach to a commercial approach involving capital investment and overall a business approach. On-site, there is little evidence that grass regeneration is occurring as yet: the extent of bush removal is small (tens of ha) and it was not accompanied by reseeding and/or fencing for grass regeneration because of associated costs and livestock management issues as it was in communal land. Reseeding was tested unsuccessfully on a few ha in one pilot ranch (drought issue).

Woodland preservation/veld resources harvesting: the project supported a women-led initiative on Mongongo processing: at this stage, it seems that fruit collection would be based on an individual harvesting mechanism with little impact on the resource. Indirectly, this activity contributes to woodland preservation and can be considered as a smart SLM endeavour.


Tag: Natural Resouce management Effectiveness Communication Policies & Procedures Advocacy

12.

3.3.3 Effectiveness and efficiency

Efficiency (project costs):

The five-year-long project spent in total around 2.7M$ (over 95% of the budget) before the project’s closure. Looking back at the project’s history, very few activities on the ground were committed prior to the mid-term review with a very low delivery rate; there were also suspicions of mismanagement including grants provided to TOCADI. This situation changed with the arrival of the second Project Coordinator: there was a sudden acceleration in both on-the-ground results and resources spending (over 60% of the budget was spent in less than 30 months). This approach was also the consequence of not requesting a no-cost extension at the mid-term review stage. Paradoxically, this has had negative consequences on project delivery, in particular when it comes to on-the-ground empowerment of project results by the final beneficiaries: for many activities, there was insufficient time to ensure project results appropriation through regular PMU/ Government monitoring, follow-up and close advice to the beneficiaries. RATING for Effectiveness: Moderately Satisfactory (MS) RATING for Efficiency: Highly Satisfactory (HS) Overall project outcome RATING: Satisfactory (S) 


Tag: Climate Change Adaptation Climate change governance Efficiency Sustainability Local Governance Human and Financial resources Monitoring and Evaluation Ownership Project and Programme management Country Government

13.

3.3.5 Mainstreaming

Gender mainstreaming:

The consultant found that gender considerations were somewhat insufficiently taken into consideration in the PRODOC despite the fact that it is an important factor for success given the differentiated roles of men and women in rangeland management, in particular when it comes to beef (mostly men-run) and small livestock (women-run). No resources had been allocated for gender mainstreaming of project activities. Although not gender-specific, the selection of pilot farms (for HLMA demonstration - output 1.2 -) took into consideration vulnerable groups (youth, female-run homestead and senior farmer). Little attention had been made to gender prior to the MTR but following up on the MTR’s recommendations, a series of activities were implemented that came at a very late stage during implementation (2017/2018). These included at least the following: (i) gender analysis produced in 07/2018 (8 months before project closure), (ii) project support to a women group focussing on piggery (from 11/2018 onwards), (iii) support to a women group engaged in mongongo agro-processing (2018 and 2019). As for (ii) and (iii), support continued after the official project end date in 03/2019. Albeit locally impacting, these activities came too little too late and could not be fully mainstreamed into the logic of the initial project design (scattering effect). The project did not produce gender-disaggregated data.

Project linkages to SDG targets:

The project is having direct positive effects on several SDGs: this is particularly the case for: 

- SDG 1 “No poverty” with Targets 4 and 5 on ‘ensuring that all men and women, in particular the poor and the vulnerable, have equal rights to economic resources’ (small livestock opportunities), as well as ‘ownership and control over land and natural resources’ (rangeland restoration), ‘appropriate new technology’ (Mongongo women group), on ‘building the resilience of the poor and those in vulnerable situations and reducing their exposure and vulnerability to climate-related extreme events’ (e.g. humaninduced / wildfires), ‘and other economic, social and environmental shocks and disturbances’ (e.g. grazing land degradation impacting livestock rearing in particular on communal lands)

- SDG 15 “Life on Land”, Targets 1 to 4 on ‘’ensuring the conservation, restoration and sustainable use of terrestrial ecosystems and their services’, ‘promoting the implementation of sustainable management of all types of forests (savannah), halting deforestation’ (restoring savannah environment through de-bushing), ‘restoring degraded land and soil’ (improving pasture quality through HLMA)

- SDG 17 (Partnerships for the Goals), Targets 6 and 9 on ‘Enhancing South-South and triangular regional cooperation’ (charcoal production with expertise from Namibia, HLMA with support from Zimbabwe) and related capacity building. 


Tag: Vulnerable Gender Equality Gender Mainstreaming Women's Empowerment Theory of Change Poverty Reduction South-South Cooperation Data and Statistics SDG Integration

14.

3.3.6 Elements of Sustainability

Sustainability is the likelihood of continued benefits after the project ends. Overall project sustainability RATING: Moderately Likely (ML)

3.3.6.1 Social & cultural risks to sustainability: Extensive efforts were undertaken to enhance (outcome 1) project’s results ownership - especially at community level through Trusts with awareness-raising activities. Overall, there is a wide acceptation of most project results but the actual empowerment remains low, especially for Trusts: ex.1 the Tsodilo Trust has no solution to compensating community members putting out fires despite some income generation from the World Heritage site, ex.2 the Lake Ngami Trust is relying nearly exclusively on Government support for charcoal production despite very positive economic prospects. The situation is somewhat different with (farmers’) associations and women groups: more proactivity and entrepreneurial spirit result in activities moving forward by project’s end (e.g. NAMA’s multispecies abattoir, women-run mongongo and piggery ventures). Conservation agriculture farmers (>60) were requested to form an association to benefit from the project support – in particular, for equipment grants -. This association never materialised and all the equipment was kept under the care of DCP. Overall though, there has been a wider acceptance of project’s activities and the type of assistance provided to communities. Socio-cultural sustainability RATING: Likely (L)


Tag: Sustainability Women's Empowerment Ownership Risk Management Advocacy

15.

3.3.7 Potential impact

In this terminal evaluation, the impact of the project has been assessed in terms of changes or benefits achieved in social, economic, institutional, environmental areas as well as the changes achieved in terms of gender. Impact RATING: Significant (S)

3.3.7.1 Social Impact: The social impact of the project can be assessed through behavioural change: the key question is whether there was enough project time to initiate sufficient behaviour change by final beneficiaries. This is not the case for most HLMA practices with little evidence of widespread adoption due to insufficient testing on commercial ranches and aborted demonstration on communal lands. However, interviews showed that all the activities that encouraged HLMA have increased farmers’ interest (commercial ranchers in Haina veld, community members from involved villages, even conservation agriculture farmers that were not associated with HLMA) while these were very reluctant at project’s start to even consider HLMA in their rangeland activities. This is a significant first step that the Government should build on with further divulgation of HLMA in the district (through routine activities). The awareness-raising activities on fire management have had a very positive impact on fire incidence with community members’ limiting fire extent early on. Social impact RATING: Significant (S)

3.3.7.2 Economic Impact: It is too early to evidence the economic impact of the project but the prospects are high under outcome 2 thanks to CBT and less so for market development (regional exporting). CBT combined with quarantine refurbishing is resulting in a recently increasing livestock off-take thanks to a higher beef price. Under outcome 1, the potential economic impact for charcoal production is also very high but has yet to materialise due to the lack of capital of Lake Ngami Trust and insufficient entrepreneurial skills and /or management capability. The situation is similar for activities led by women; however, interviews showed that these groups develop a strong entrepreneurial spirit but still lack entrepreneurial skills. There is also an indirect economic impact with the reduction of fires, in particular around Tsodilo Hills World Heritage site with tourism development that is affected negatively with fires. Under the current subsidizing scheme, conservation agriculture is more efficient than conventional one with participating farmers reporting higher yields (2 to 3 fold with conservation agriculture). It remains to be seen whether non-subsidized inputs will not erode farmers’ earnings and turn conservation agriculture unprofitable. Economic impact RATING: Significant (S)


Tag: Impact Gender Equality Gender Mainstreaming Women's Empowerment Local Governance Communication Monitoring and Evaluation Programme Synergy Capacity Building Advocacy

Recommendations
1

Project design and budget allocation

There is a need at the formulation stage to reflect better actual development project implementation with the inclusion of an extensive inception period to allow for initial project operationalization. This can have significant positive consequences as it will allow the project team to follow better the PRODOC framework with more logical activity sequencing and allow progressive delivery more in tune with reality.

2

The PRODOC did not include formally any exit strategy.

It was assumed that stakeholders would be “empowered through activity delivery, in particular, trainings and resource mobilisation. Experience has shown that appropriation and empowerment are not systematic through planned delivery. There is a need to accompany this process. In this particular project, there was no direct linkage between the project activities and the involved department’s routine programming. Projects need to ensure results integration into Government routine activities by allocating specific resources for that purpose; this includes support to update and/or reformulate periodic programming documents to mainstream project results (e.g. one or two years before project closure).

3

Evaluation Recommendation 3: Implementation

Project AWP did include information on each activity implementing partner but no indication as to who was leading the activity when two or more stakeholders were involved. Interviews showed that there was insufficient responsibility-taking by stakeholders and often, most work rested with the PMU. AWP needs to be much more specific in defining roles and responsibilities for activity delivery; in particular when two or more parties are implementing activities, one must be designated as the lead responsible party; otherwise, no one is responsible, and PMU will need to intervene and take over.

4

Communication

Despite no explicit communication strategy, the PMU made extensive efforts to successfully communicate on the project’s results through various means (press releases, events, UNDP web site…). Interviews have shown that this has led beneficiaries to view the project as UNDP-led with limited Government support despite being under NEX modality. More attention on how to communicate is necessary to place Government more at the forefront of projects’ information; this can be done through communicating on project results via Government channels, in particular through the information divisions of ministries, possibly reserving financial resources for that purpose at formulation stage (including through regular ministries web pages).

5

Government empowerment at district level

 The project had planned for the designation of counterparts within MoA and MEWT to ensure adequate coordination between the PMU and on-site technical staff. This approach is insufficient to ensure Government empowerment as discussed with many departments’ heads: much closer involvement is necessary, in particular, through the designation of focal points at department level who work on a nearly daily basis with the PMU, taking over the actual operationalization of project activities. In that context, Government staff is an intermediary between the PMU and the final beneficiaries. This is a necessary approach for (i) ‘branding’ (beneficiaries see the intervention as Government- / not UN-run) and (ii) for sustainability as Government staff will be the sole interlocutor for follow-up and advice of final beneficiaries, once the project closes. This may be a missing link to ensure Government empowerment, but it requires extensive commitment from Government – possibly to be secured at formulation stage; in particular dedicated staff and time (cofinancing) to work on activities as per AWP in close collaboration with PMU.  

6

Income generation and VC approach

 The project design has not adopted a VC approach for income-generating activities; this is because the focus was on some but not all stakeholders of value chains related to livestock production, veld product harvesting and value addition through de-bushing activities (charcoal, cattle feed). It resulted in partial project support, mostly at production level, with additional externalised/coordinated support whenever possible (e.g. LEA on entrepreneurial skills). While this may be effective for minimising required resources, assistance to other parts of the VC might be necessary to effectively operationalize the VC and ensure adequate commercialisation. Project support to the development of income-generating activities - in particular new ones like piggery, mongongo processing, charcoal development - needs to adopt a more systematic VC approach at design stage to ensure that VC stakeholders’ technical and entrepreneurial gaps are filled from production to commercialisation. This can be primarily accomplished through external co-financing as was done with BMC for the beef VC

7

Lack of ‘enabling environment’

The project relied on a series of on-site stakeholders as beneficiaries to mainstream several SLM practices (e.g. fire management, de-bushing on communal lands through community trusts). Trusts, in particular, were created following the drafting of the CBNRM policy. As a policy, there are no specifically defined quality standards for management. The project has suffered from the lack of professionalization of Trusts to ensure activity development. Projects need to include resources for ensuring that a conducive policy governs stakeholders; in the case of Trusts, an act and regulations should be amongst the highest priorities of project support or alternatively, PMUs should resolve themselves in not collaborating with stakeholders in an unclear institutional framework. At a higher level, it is necessary to include an ‘enabling environment’ or policy component in projects ensuring that legal instruments and project activities are in line for smooth project implementation. On the one hand, it is somewhat surprising that a policy component was missing in the PRODOC as the objective of the project was to mainstream SLM. On the other hand, there are socio-economic/political pressure groups that wish a status quo on TGLP and other SLM-related policies. Nonetheless, future interventions should open up the debate for policy reforms

8

Delivery acceleration versus project extension

 One of the criteria for success is the budget delivery rate by project’s end. In the SLM project, it is nearly 100% with no requested project extension. A lot of emphasis by donors is put on this particular criterion and it is often viewed as success if a project delivered within the originally planned timeframe. This, however, does not account for the quality of results – in particular, the level of appropriation of results and the empowerment by stakeholders. These are complex processes that need time. The SLM project did complete (nearly) all activities as per PRODOC but lacked time to ensure adequate ownership and empowerment: this is the case for many activities that were initiated very late during implementation (2017/8) and for which support just stopped with the project’s end (not quite with additional UNDP support after project’s end though): HLMA was both tested and demonstrated on selected (commercial) farms at the same time, meaning participating farmers were both assessing new management practices and expected to adopt those most promising ; for other activities, trainings were delivered but no activity implementation carried out; this raises final beneficiaries expectations only to experience project closure before actual activity implementation or visualise activity effects. With low delivery at project start-up, it is necessary to systematically request a project extension even if the PMU expects to accelerate budget delivery. Otherwise, this can be done at the expense of output quality. This is particularly the case when behavioural change is sought as project outcomes, such as in the SLM project with improved rangeland management.

9

Project governance

 The SLM project governance mechanism through TRG and PSC was very participatory, ensuring that all stakeholders were aware of progress and could contribute both at planning and technical levels (TRG) and in the decision-making process (PSC). The analysis of minutes showed that technical and senior department staff were often participating in both structures. This may be an issue for ensuring adequate responsibility-taking by participating departments. TRG and PSC ToRs need to be more precise on their composition; PSC structure should be streamlined with only senior staff present (e.g. one per ministry) and for TRG, only technical staff (one per department). This would ensure (i) more balanced representativeness of Government stakeholders in project governance structures (the same would be expected from Non-Governmental Organisations participation) and (ii) a more proactive role of heads of departments in briefing their hierarchy on project progress and issues prior to PSC

10

Adaptive management

 The PMU has shown strong adaptive management with procurement process streamlining, additional activities (measles/ticks/Tsodilo & Lake Ngami events) when required and always in line with the project objective, flexibility in amending activities for efficiency purpose (e.g. association with volunteer commercial ranchers on control/pilot farms). The governance structures were however shy in curtailing quantitative objectives (e.g. acreage under SLM) although it was clear these would not be met. Decision making by the project governance structures should be based on advice from as many relevant stakeholders but primarily take into consideration the opinion of PMU, project stakeholders and the MTR. Under the SLM project, reducing the scope at MTR level would have been very beneficial to the objective with resource consolidation to ensure closer follow-up and more intense advice to final beneficiaries with a view to empowerment and ownership.

11

Final beneficiaries’ commitment

The low commitment of final beneficiaries and difficulty in achieving behaviour change in SLM and in particular for rangeland management comes from (i) weekend farmers whose primary income is no longer livestock-related (mostly urban-generated) and often unable to fully participate in and oversee project activities and (ii) farmers entirely relying on livestock for a living and resisting change for fear of losing income. There is little evidence that the project adopted a differentiated approach for collaboration with the intent of establishing control farms (in commercial ranches and on communal rangeland). For (i), participation is mostly seen as an add-on on the conditions of little extra effort required; they have a tendency to move over/abandon if there is limited project advice; emphasis must be put on accompanying closely these farmers that often rely on poorly trained workers. The approach must be sensibly different for (ii): it should be based on the actual beneficiary needs that may be revisited during the inception workshop, and require a more holistic approach with different issues that are critical for new rangeland practices adoption (e.g. grazing rotation requiring additional boreholes, 24/7 herding requiring professional herders…). In this particular case, commitment may also be secured at design stage through farmers’ / community cofinancing (most often in-kind such as labour e.g. 5-10% of activity budget.

12

Evaluation Recommendation 12: Usefulness of baseline studies

Experience has shown that most baseline studies requiring international procurement usually results in a long process often not finalised by MTR. Yet, these exercises have value only if integrated at project start-up (under the SLM project, the rangeland assessment was available in 2017, 3 years after project start-up). UNDP needs to revisit the mechanism for initiating baseline studies, alternatively: (i) baseline studies should be integrated into PPG, with TRAC funds if necessary, meaning they would be ready at project signature, TORs drafted by the PPG team during formulation, (ii) baselines studies procurement should be initiated immediately after Government project signature through direct UNDP procurement (without PSC approval / before PMU contracting), and be ready ideally 1 year after project signature, when the PMU is in place. This would require TORs integrated into the PRODOC beforehand.

13

Evaluation Recommendation 13: Project area and logistical costs

The SLM project covered huge sways of land resulting in extensive PMU travel. For piloting, it is advisable to seek close-by project sites for the sake of costs and ease of organising training sessions. Combining outputs in limited areas may turn the project more into an integrated intervention under which results can benefit one another. For demonstration, scattering is preferable to increase beneficiary outreach.

14

Evaluation Recommendation 14: Need for non-governmental beneficiary’s follow-up

The project remaining timeframe was insufficient to ensure adequate appropriation of several project results; there is a need to continue accompanying the mongongo processing project, Lake Ngami Trust charcoal production, piggery women group, NAMA abattoir by relevant sectoral departments to support their activity development ; they also lack entrepreneurial skills and should be redirected to LEA (or similar) for closer follow-up (e.g. assignation of a LEA Business Advisor), relevant trainings in entrepreneurial skills (planning, business plan, management, accountancy…) and exploring organisation’s capitalisation (loans through CEDA or grants through Gender A and Youth Independent Office), with a view on customised support.

15

Evaluation Recommendation 15: Charcoal production

The Ministry of Energy should be involved in supporting Ngamiland charcoal production as part of its Green Energy strategy and explore the potential VC; there are indications of local interest for charcoal through the high-end tourism sector, for low volumes, and from Namibia’s charcoal industry for high volumes intended for export.

16

Evaluation Recommendation 16: HLMA further analysis and piloting/demonstrating

With new project proposals/concept notes currently being drafted, the Government has clearly adopted HLMA as one of the tools to ensure more sustainable rangeland management and livestock rearing modernisation. However, it should be reminded that HLMA has not been thoroughly tested – in particular when it comes to costs-; amongst the priorities in HLMA would be a comprehensive CBA of HLMA techniques and their potential impact on rangeland regeneration within the Ngamiland context. This might require the collaboration of academia.

17

Evaluation Recommendation 17: De-bushing and animal feed

There are prospects for cattle feed from de-bushing activities. However, the project did not purchase the required equipment (shredder/hammer mill). A VC analysis should be commissioned to assess feasibility and economic viability (in particular on potential / actual demand in the district and beyond.)

18

Evaluation Recommendation 18: TAC / LEA support to Lake Ngami Trust

TAC was not fully following-up the Trust and now it requires technical advice to continue charcoal production that has stalled altogether. The Trust needs to be advised on how best to streamline the production process and involve Lake Ngami community members in production. The trust lacks entrepreneurial spirit and skills with the imminent collapse of charcoal production at the end of the project support. LEA should be involved in the professionalization of management and support the trust in seeking capital investment.

19

Evaluation Recommendation 19: CBT and quarantine

There is a current spike of cattle off-take with CBT up and running and the current drought. With this trend in mind, there is a risk that grass will be rapidly exhausted before the end of the dry season: the project support to the quarantine enabled the opening of 6 paddocks out of 16; however the required investments vastly surpass the project support and one should reconsider how to effectively run the quarantine on a continuous basis (100% Government run, farmers’ small/large contribution to maintenance…).

20

Evaluation Recommendation 20: SLM mainstreaming in district institutions

With the project’s end, it is not ensured that the Okavango Delta Wetland Management Committee will pursue dialogues on SLM mainstreaming; it is entirely dependent on the members to put forward SLM on the forum’s agenda. It may be relevant to seek TNC support to follow-up on SLM mainstreaming.

21

Trusts strengthening:

The lack of legislation on trusts’ operations has been an impediment for project delivery: there is insufficient accountability of trust management to community members; this was evident with most trusts the SLM project interacted with: there was insufficient commitment, poor management, incomplete revenue reporting to DWNP, a lack of professionalism and overall a lack of vision for trust development. To ensure improved management of trusts, all stakeholders should lobby for the endorsement of the CBNRM act, complete with regulations on trusts. With rules and regulations, it may be easier for projects to secure commitment and professional involvement in project activities

22

Evaluation Recommendation 22: SLM policy changes and mainstreaming into interventions:

UNDP should lobby Government together with civil society on policy reform to integrate SLM into the main policies on land use, agriculture and wildlife, and related economic development. SLM should also be systematically integrated at the design stage into interventions focusing on rangeland and agriculture development.

23

Evaluation Recommendation 23: Involvement of the private sector in charcoal production:

 The project has been constrained by the lack of proactivity of Lake Ngami Trust despite extensive support; furthermore, the trust is not reporting to revenue to DWNP as advised and has neither voluntarily sought assistance from LEA on capital investment nor formulated any business plan. As an alternative, the private sector interest should be probed to speed up charcoal VC development.

24

Evaluation Recommendation 24: Export market development

The project support on export markets was significantly constrained by beef quality (e.g. importers for Mozambique estimate that Ngamiland meat is not tender enough); this was an external factor not accounted for by the project. Interviews confirmed that Ngamiland cattle genetics is poor and needs improvement. While artificial insemination is available in Ngamiland, divulgation remains low and costs may be an issue for farmers on communal lands. In any case, improvement should be associated with quality feed, hence the need for rangeland management practices maximising grass availability and possibly higher quality feed through feed lotting. These should be tested at some point in the future through an integrated livestock development project, targeting livestock farmers on communal lands.

25

Evaluation Recommendation 25: CBT and rangeland management

With successful CBT combined with higher prices, there is a risk for further pasture degradation with renewed interest in livestock rearing. The government should pick up on the project’s results, in particular on HLMA and accompany the modernisation of the sector with continued divulgation of HLMA.

26

Evaluation Recommendation 26: Empowerment of HLMA by livestock farmers

Combined with successive piloting and then demonstrating the viability and cost-effectiveness of HLMA techniques (paddocking, 24/7 herding, kraaling, grazing plans…), the divulgation of HLMA should be carried out through developing a mechanism of added support and training through behaviour change milestones (the more the attitudes are changing, the more technical assistance the farmer receives, based on cost efficiency, hence the need for CBA). Overall, the project did not change fundamentally the way livestock farmers in Ngamiland manage their grazing land: first, this was a pilot project and second SLM was a new concept that require time for behavioural change. Therefore, it is recommended that Government and donors alike invest time and resources in sequentially (i) testing HLMA TNC has been involved in the ODMP resulting in graduating the most environmental-friendly and economic practices and (ii) demonstrating and divulging HLMA to district livestock farmers using voluntary farmers (using the concept of FFS for livestock).

27

Evaluation Recommendation 27: Human-wildlife conflict

Interviews showed that this is concern number one by large, far ahead of land degradation. Any follow-up of project’s results should include a component on this issue (through lobbying for renewed hunting, buffalo-proof fencing and/or trenches digging, land reallocation, improved agriculture/wildlife demarcation…) to ensure more appropriation and ownership of results.

28

4.1 Recommendations

4.1.1 Corrective actions for the design, implementation, monitoring and evaluation of the project

(i) Project design and budget allocation:

The budget allocation at project formulation stage (see Figure 2; planned budget) was typically skewed as too optimistic without any period of low delivery corresponding to the project initial operationalization period (inception workshop, purchase of initial equipment, recruitment of staff, baseline study and consultants). Most if not all projects experience an initial period of very low project activity that is not considered by project designers who plan for immediate delivery of activities; typically, the budget allocation will follow a linear or logarithmic spending curse (scenarios a. or b. in Figure 3); this is in contradiction with any real-world situation, which is why all projects experience major budget reallocations during the second half of the project and need to accelerate delivery often at the expense of quality. This puts unnecessary pressure on project teams that are unable to follow up PRODOC results framework and work plans, inevitably leading to suboptimal delivery and systematic requests of project extensions.

Under the SLM project though, the acceleration in delivery without project extension resulted in squeezing many activities under a very tight timeframe with negative consequences in appropriation. There was little time for beneficiaries to empower themselves of project results (e.g. Trusts), activities were not completed as planned/required (e.g. output 1.5 MOMPS, output 1.2 on HLMA for communal land). In real situation, projects follow more of a sigmoid delivery curve (scenarios c. or d. in Figure 3) as for the SLM project (see Figure 2; actual budget). There is a need at the formulation stage to reflect better actual development project implementation with the inclusion of an extensive inception period to allow for initial project operationalization. This can have significant positive consequences as it will allow the project team to follow better the PRODOC framework with more logical activity sequencing and allow progressive delivery more in tune with reality.

(ii) Exit strategy: The PRODOC did not include formally any exit strategy. It was assumed that stakeholders would be “empowered through activity delivery, in particular, trainings and resource mobilisation”. Experience has shown that appropriation and empowerment are not systematic through planned delivery. There is a need to accompany this process. In this particular project, there was no direct linkage between the project activities and the involved department’s routine programming. Projects need to ensure results integration into Government routine activities by allocating specific resources for that purpose; this includes support to update and/or reformulate periodic programming documents to mainstream project results (e.g. one or two years before project closure).

(iii) Implementation Project AWP did include information on each activity implementing partner but no indication as to who was leading the activity when two or more stakeholders were involved. Interviews showed that there was insufficient responsibility-taking by stakeholders and often, most work rested with the PMU. AWP needs to be much more specific in defining roles and responsibilities for activity delivery; in particular when two or more parties are implementing activities, one must be designated as the lead responsible party; otherwise, no one is responsible and PMU will need to intervene and take over.

(iv) Communication Despite no explicit communication strategy, the PMU made extensive efforts to successfully communicate on the project’s results through various means (press releases, events, UNDP web site…). Interviews have shown that this has led beneficiaries to view the project as UNDP-led with limited Government support despite being under NEX modality. More attention on how to communicate is necessary to place Government more at the forefront of projects’ information; this can be done through communicating on project results via Government channels, in particular through the information divisions of ministries, possibly reserving financial resources for that purpose at formulation stage (including through regular ministries web pages). 

(v) Government empowerment at district level The project had planned for the designation of counterparts within MoA and MEWT to ensure adequate coordination between the PMU and on-site technical staff. This approach is insufficient to ensure Government empowerment as discussed with many departments’ heads: much closer involvement is necessary, in particular, through the designation of focal points at department level who work on a nearly daily basis with the PMU, taking over the actual operationalization of project activities. In that context, Government staff is an intermediary between the PMU and the final beneficiaries. This is a necessary approach for (i) ‘branding’ (beneficiaries see the intervention as Government- / not UN-run) and (ii) for sustainability as Government staff will be the sole interlocutor for follow-up and advice of final beneficiaries, once the project closes. This may be a missing link to ensure Government empowerment but it requires extensive commitment from Government – possibly to be secured at formulation stage- ; in particular dedicated staff and time (cofinancing) to work on activities as per AWP in close collaboration with PMU.

(vi) Income generation and VC approach The project design has not adopted a VC approach for income-generating activities; this is because the focus was on some but not all stakeholders of value chains related to livestock production, veld product harvesting and value addition through de-bushing activities (charcoal, cattle feed). It resulted in partial project support, mostly at production level, with additional externalised/coordinated support whenever possible (e.g. LEA on entrepreneurial skills). While this may be effective for minimising required resources, assistance to other parts of the VC might be necessary to effectively operationalize the VC and ensure adequate commercialisation. Project support to the development of income-generating activities - in particular new ones like piggery, mongongo processing, charcoal development - needs to adopt a more systematic VC approach at design stage to ensure that VC stakeholders’ technical and entrepreneurial gaps are filled from production to commercialisation. This can be primarily accomplished through external cofinancing as was done with BMC for the beef VC.

(vii) Lack of ‘enabling environment’ The project relied on a series of on-site stakeholders as beneficiaries to mainstream several SLM practices (e.g. fire management, de-bushing on communal landsthrough community trusts). Trusts, in particular, were created following the drafting of the CBNRM policy. As a policy, there are no specifically defined quality standards for management. The project has suffered from the lack of professionalization of Trusts to ensure activity development. Projects need to include resources for ensuring that a conducive policy governs stakeholders; in the case of Trusts, an act and regulations should be amongst the highest priorities of project support or alternatively, PMUs should resolve themselves in not collaborating with stakeholders in an unclear institutional framework. At a higher level, it is necessary to include an ‘enabling environment’ or policy component in projects ensuring that legal instruments and project activities are in line for smooth project implementation. On the one hand, it is somewhat surprising that a policy component was missing in the PRODOC as the objective of the project was to mainstream SLM. On the other hand, there are socio-economic/political pressure groups that wish a status quo on TGLP and other SLM-related policies. Nonetheless, future interventions should open up the debate for policy reforms.

(viii) Delivery acceleration versus project extension One of the criteria for success is the budget delivery rate by project’s end. In the SLM project, it is nearly 100% with no requested project extension. A lot of emphasis by donors is put on this particular criterion and it is often viewed as success if a project delivered within the originally planned timeframe. This, however, does not account for the quality of results – in particular, the level of appropriation of results and the empowerment by stakeholders. These are complex processes that need time. The SLM project did complete (nearly) all activities as per PRODOC but lacked time to ensure adequate ownership and empowerment: this is the case for many activities that were initiated very late during implementation (2017/8) and for which support just stopped with the project’s end (not quite with additional UNDP support after project’s endthough): HLMA was both tested and demonstrated on selected (commercial) farms at the same time, meaning participating farmers were both assessing new management practices and expected to adopt those most promising; for other activities, trainings were delivered but no activity implementation carried out; this raises final beneficiaries expectations only to experience project closure before actual activity implementation or visualise activity effects. With low delivery at project start-up, it is necessary to systematically request a project extension even if the PMU expects to accelerate budget delivery. Otherwise, this can be done at the expense of output quality. This is particularly the case when behavioural change is sought as project outcomes, such as in the SLM project with improved rangeland management.

(ix) Project governance The SLM project governance mechanism through TRG and PSC was very participatory, ensuring that all stakeholders were aware of progress and could contribute both at planning and technical levels (TRG) and in the decision-making process (PSC). The analysis of minutes showed that technical and senior department staff were often participating in both structures. This may be an issue for ensuring adequate responsibility-taking by participating departments. TRG and PSC ToRs need to be more precise on their composition; PSC structure should be streamlined with only senior staff present (e.g. one per ministry) and for TRG, only technical staff (one per department). This would ensure (i) more balanced representativeness of Government stakeholders in project governance structures (the same would be expected from Non-Governmental Organisations participation) and (ii) a more proactive role of heads of departments in briefing their hierarchy on project progress and issues prior to PSC.

(x) Adaptive management The PMU has shown strong adaptive management with procurement process streamlining, additional activities (measles/ticks/Tsodilo & Lake Ngami events) when required and always in line with the project objective, flexibility in amending activities for efficiency purpose (e.g. association with volunteer commercial ranchers on control/pilot farms). The governance structures were however shy in curtailing quantitative objectives (e.g. acreage under SLM) although it was clear these would not be met. Decision making by the project governance structures should be based on advice from as many relevant stakeholders but primarily take into consideration the opinion of PMU, project stakeholders and the MTR. Under the SLM project, reducing the scope at MTR level would have been very beneficial to the objective with resource consolidation to ensure closer follow-up and more intense advice to final beneficiaries with a view to empowerment and ownership.

(xi) Final beneficiaries commitment The low commitment of final beneficiaries and difficulty in achieving behaviour change in SLM and in particular for rangeland management comes from (i) weekend farmers whose primary income is no longer livestock-related (mostly urban-generated) and often unable to fully participate in and oversee project activities and (ii) farmers entirely relying on livestock for a living and resisting change for fear of losing income. There is little evidence that the project adopted a differentiated approach for collaboration with the intent of establishing control farms (in commercial ranches and on communal rangeland). For (i), participation is mostly seen as an add-on on the conditions of little extra effort required; they have a tendency to move over/abandon if there is limited project advice; emphasis must be put on accompanying closely these farmers that often rely on poorly trained workers. The approach must be sensibly different for (ii): it should be based on the actual beneficiary needs that may be revisited during the inception workshop, and require a more holistic approach with different issues that are critical for new rangeland practices adoption (e.g. grazing rotation requiring additional boreholes, 24/7 herding requiring professional herders…). In this particular case, commitment may also be secured at design stage through farmers’ / community cofinancing (most often in-kind such as labour e.g. 5-10% of activity budget). 

(xii) Usefulness of baseline studies Experience has shown that most baseline studies requiring international procurement usually results in a long process often not finalised by MTR. Yet, these exercises have value only if integrated at project start-up (under the SLM project, the rangeland assessment was available in 2017, 3 years after project start-up). UNDP needs to revisit the mechanism for initiating baseline studies, alternatively: (i) baseline studies should be integrated into PPG, with TRAC funds if necessary, meaning they would be ready at project signature, TORs drafted by the PPG team during formulation, (ii) baselines studies procurement should be initiated immediately after Government project signature through direct UNDP procurement (without PSC approval / before PMU contracting), and be ready ideally 1 year after project signature, when the PMU is in place. This would require TORs integrated into the PRODOC beforehand.

(xiii) Project area and logistical costs The SLM project covered huge sways of land resulting in extensive PMU travel. For piloting, it is advisable to seek close-by project sites for the sake of costs and ease of organising training sessions. Combining outputs in limited areas may turn the project more into an integrated intervention under which results can benefit one another. For demonstration, scattering is preferable to increase beneficiary outreach.

29

4.1.2 Actions to follow-up or reinforce initial benefits from the project

(i) Need for non-governmental beneficiaries follow-up The project remaining timeframe was insufficient to ensure adequate appropriation of several project results; there is a need to continue accompanying the mongongo processing project, Lake Ngami Trust charcoal production, piggery women group, NAMA abattoir by relevant sectoral departments to support their activity development ; they also lack entrepreneurial skills and should be redirected to LEA (or similar) for closer follow-up (e.g. assignation of a LEA Business Advisor), relevant trainings in entrepreneurial skills (planning, business plan, management, accountancy…) and exploring organisation’s capitalisation (loans through CEDA or grants through Gender A and Youth Independent Office), with a view on customised support.

(ii) Charcoal production The Ministry of Energy should be involved in supporting Ngamiland charcoal production as part of its Green Energy strategy and explore the potential VC; there are indications of local interest for charcoal through the high-end tourism sector, for low volumes, and from Namibia’s charcoal industry for high volumes intended for export. 

(iii) HLMA further analysis and piloting/demonstrating With new project proposals/concept notes currently being drafted, the Government has clearly adopted HLMA as one of the tools to ensure more sustainable rangeland management and livestock rearing modernisation. However, it should be reminded that HLMA has not been thoroughly tested – in particular when it comes to costs-; amongst the priorities in HLMA would be a comprehensive CBA of HLMA techniques and their potential impact on rangeland regeneration within the Ngamiland context. This might require the collaboration of academia.

(iv) De-bushing and animal feed There are prospects for cattle feed from de-bushing activities. However, the project did not purchase the required equipment (shredder/hammer mill). A VC analysis should be commissioned to assess feasibility and economic viability (in particular on potential / actual demand in the district and beyond.)

(v) TAC / LEA support to Lake Ngami Trust TAC was not fully following-up the Trust and now it requires technical advice to continue charcoal production that has stalled altogether. The Trust needs to be advised on how best to streamline the production process and involve Lake Ngami community members in production. The trust lacks entrepreneurial spirit and skills with the imminent collapse of charcoal production at the end of the project support. LEA should be involved in the professionalization of management and support the trust in seeking capital investment.

(vi) CBT and quarantine There is a current spike of cattle off-take with CBT up and running and the current drought. With this trend in mind, there is a risk that grass will be rapidly exhausted before the end of the dry season: the project support to the quarantine enabled the opening of 6 paddocks out of 16; however the required investments vastly surpass the project support and one should reconsider how to effectively run the quarantine on a continuous basis (100% Governmentrun, farmers’ small/large contribution to maintenance…).

(vii) SLM mainstreaming in district institutions With the project’s end, it is not ensured that the Okavango Delta Wetland Management Committee will pursue dialogues on SLM mainstreaming; it is entirely dependent on the members to put forward SLM on the forum’s agenda. It may be relevant to seek TNC36 support to follow-up on SLM mainstreaming. 

30

4.1.3 Proposals for future directions underlining main objectives

(i) Trusts strengthening: The lack of legislation on trusts’ operations has been an impediment for project delivery: there is insufficient accountability of trust management to community members; this was evident with most trusts the SLM project interacted with: there was insufficient commitment, poor management, incomplete revenue reporting to DWNP, a lack of professionalism and overall a lack of vision for trust development. To ensure improved management of trusts, all stakeholders should lobby for the endorsement of the CBNRM act, complete with regulations on trusts. With rules and regulations, it may be easier for projects to secure commitment and professional involvement in project activities

(ii) SLM policy changes and mainstreaming into interventions: UNDP should lobby Government together with civil society on policy reform to integrate SLM into the main policies on land use, agriculture and wildlife, and related economic development. SLM should also be systematically integrated at the design stage into interventions focussing on rangeland and agriculture development. 

(iii) Involvement of the private sector in charcoal production: The project has been constrained by the lack of proactivity of Lake Ngami Trust despite extensive support; furthermore, the trust is not reporting to revenue to DWNP as advised and has neither voluntarily sought assistance from LEA on capital investment nor formulated any business plan. As an alternative, the private sector interest should be probed to speed up charcoal VC development.

(iv) Export market development The project support on export markets was significantly constrained by beef quality (e.g. importers for Mozambique estimate that Ngamiland meat is not tender enough); this was an external factor not accounted for by the project. Interviews confirmed that Ngamiland cattle genetics is poor and needs improvement. While artificial insemination is available in Ngamiland, divulgation remains low and costs may be an issue for farmers on communal lands. In any case, improvement should be associated with quality feed, hence the need for rangeland management practices maximising grass availability and possibly higher quality feed through feed lotting. These should be tested at some point in the future through an integrated livestock development project, targeting livestock farmers on communal lands.

(v) CBT and rangeland management With successful CBT combined with higher prices, there is a risk for further pasture degradation with renewed interest in livestock rearing. The government should pick up on the project’s results, in particular on HLMA and accompany the modernisation of the sector with continued divulgation of HLMA.

(vi) Empowerment of HLMA by livestock farmers Combined with successive piloting and then demonstrating the viability and cost-effectiveness of HLMA techniques (paddocking, 24/7 herding, kraaling, grazing plans…), the divulgation of HLMA should be carried out through developing a mechanism of added support and training through behaviour change milestones (the more the attitudes are changing, the more technical assistance the farmer receives, based on cost efficiency, hence the need for CBA). Overall, the project did not change fundamentally the way livestock farmers in Ngamiland manage their grazing land: first, this was a pilot project and second SLM was a new concept that require time for behavioural change. Therefore, it is recommended that Government and donors alike invest time and resources in sequentially (i) testing HLMA resulting in graduating the most environmental-friendly and economic practices and (ii) demonstrating and divulging HLMA to district livestock farmers using voluntary farmers (using the concept of FFS for livestock). 

(vii) Human-wildlife conflict Interviews showed that this is concern number one by large, far ahead of land degradation. Any follow-up of project’s results should include a component on this issue (through lobbying for renewed hunting, buffalo-proof fencing and/or trenches digging, land reallocation, improved agriculture/wildlife demarcation…) to ensure more appropriation and ownership of results.

31

4.1.4 Best and worst practices in addressing issues relating to relevance, performance and success

--- Insufficient buy-in at project start-up: stakeholders were unimpressed for the first 1 – 1 ½ of project implementation: very little project activity was implemented with delayed PMU recruitment and placement of Coordinator in Gaborone for over 6 months to replace staff. This tended to undermine the confidence of final beneficiaries and Government stakeholders in the project, the more so as SLM was viewed as a new concept not fully understood by most stakeholders. It may be counterproductive to delay project implementation for any reason (e.g. lack of office space at the implementing partner, lack of staff at UNDP country office) after project signature. PMU unit should have been mobilised as soon as possible and have spent extensive efforts in explaining the project concept and discussed beneficiary expectations right at project start-up. 

+++ First Coordinator replacement: both Government and UNDP may be hesitant in removing underperforming staff. This may result in disastrous consequences on implementation including an accumulation of technical and legal issues by the end of the project. Good coordination between UNDP and MEWT resulted in a swift removal procedure followed by a new Coordinator that successfully accelerated project delivery.

--- Some of the project results included outputs primarily financed by BMC (outputs 2.3 and 2.4). This is a risky approach as the project relies heavily on cofinancing for achieving its objective. While this was successful or the SLM project, it may jeopardize an entire project if cofinancing does not materialise or is less effective than expected. Project design must avoid in all cases relying extensively on cofinancing for achieving outcomes. Cofinancing can contribute qualitatively or quantitatively in outcomes but not contribute nearly entirely to a project outcome. 

--- Project design did not include a policy component despite the overall project objective of SLM mainstreaming; this is somewhat baffling as (i) it has limited the effectiveness of the multi-stakeholder forum on dialogue district level but little or no mandate for policy changes at national level and (ii) it was well-known that CBO governance under the CBNRM policy was an issue while the project primary target was CBOs and trusts in particular. Projects should systematically include some sort of project component on ‘enabling environment’ or policy updating to (i) ensure that the implementation of project activities is in line with the legislative framework, (ii) the sustainability of project results is secured and cannot be legally challenged and (iii) ensure that policy changes resulting in project activities are endorsed by Government at the highest level. 

+++ UNDP country office M&E tools: although not project-specific, UNDP has devised a number of M&E tools to ensure that the top-management is always aware of project’s progress.This is most beneficial for PMUs that are encouraged to effectively monitor closely the project’s results (see examples under Annexe 10).

Latest Evaluations

Contact us

1 UN Plaza
DC1-20th Floor
New York, NY 10017
Tel. +1 646 781 4200
Fax. +1 646 781 4213
erc.support@undp.org