1. Pay No Bribe: Analysis of Lessons Learned September 2017 Coffey International

12. 12 implementation of the programme since 2014 and will continue to do so until December 2018, when the ACC will assume full management. 11.2 The DFID Business Case states that the programme should bring about institutional behavioural change by the end of its life , and that an e valuation will ascertain whether the database continues to be needed, or whether the dat abase can be graduated to provide more broadly focused performance reports on districts in line with decentralisation . 11.3 The principle direct threat to achieving the overall goals of the PNB and maintaining an impact over time is the lack of engagement and lacklustre performance of MDAs , both at local level and at senior levels . This will require concerted effort over the next few months. 11.4 The activities carried out by the CSOs will not be transferred to the ACC as there will be no funding for this work after the end of the programme. However, contracted CSO s currently facilitate the clear majority of reports to the PNB system through the mobile application. This should be integrated in ACC workplans in the 1 5 months to the planned closure of the programme. 11.5 Over the next period, DFID and ACC would like to see an expansion of the programme to all districts of Sierra Leone . Full programme implementation will take place in the additional district of Kono, where t here is an ACC regional office and an office for NMJD, one of the existing CSOs. For the remaining 9 districts a ‘hub and spokes’ model is envisaged with existing CSOs and ACC offices carrying out PNB activities in short visits from their bases. This shoul d be combined with nationwide outreach through radio and other media. 12. Sustainability: Lessons Learned 12.1 Sustainability should become a fundamental consideration throughout remaining the implementation period . Given the relatively short period of time remaining to programme closure, it will be important to focus on the most effective ways of maximising the impact of activities and to ensure a planned handover to the ACC. 12.2 A sustainability plan should be develope d with the ACC as soon as possible , as it is important to start the process of handover sooner rather than later. Careful identification of tasks for handing over, will be required, along with a timeline and names of persons responsible. Any capacity needs should be identified and factored in . 12.3 The dependence on the mobile app for making reports is not sustainable. Outreach should instead focus on use of the free 515 number , especially in areas where ther e is reasonable network coverage. CSOs should work wi th their animators to help support this shift, and outreach materials and communications should disseminate the 515 message loudly and clearly. This should be supported by longer hours at the Call Centre (especially in the evening) and the availability of Call Centre Operators with Mende and Temne proficiency. 12.4 In the remaining 15 months, a focus on broad and effective outreach will be required if all Sierra Leoneans are to learn about the PNB . In the Kenema meeting , PNB Partners all agreed that communications should become more creative and more targeted to different groups, that a range of different radio programmes should be widely employed.

15. 15 inform ed the ACC that SIM - boxe s had been banned in Sierra Leone , but the PNB SIM - box is sti ll operational. 15.3 There were questions about whether the Call Centre equipment had been used elsewhere prior to its installation. Coffey was un able to prove that the equipment was new, as most of it is no longer in production and details can not be retrieved . The installers, SBT S, provided invoice s and added an additional warranty . 15.4 Under MSI , a developer was contracted to build the website and database using a platform designed to gather testimonies submitted via SMS or by a call centre agent. By early 2016 , this developer had used most of its allocated hours but t he platform was not functional. In April 2016, it was decided to collect quantitative rather than narrative data and to replace the platform . The developer’ s contract was exten ded to build the new platform . 15.5 A second developer was then introduced to build the mobile application , as they offered a more cost - effective solution. Problems aro se when the new mobile application proved difficult to integrate with the original database , and the work became resource intensive . The resulting platform was fragile, error prone and difficult to update. 15.6 Coffey wanted to replac e the platform with a more sustainable solution which would allow for updates. They collected feedback from users and stakeholders to inform the new platform , but poor communication between the them and the ACC meant that there was limited understanding of current issues and no shared agreement on way s forward. 15.7 In 2017 , the ACC initiated the development of an Enterprise Resource Plan ning System, to host all their various mana gement systems and processes, including the PNB . As a result, the development of a new PNB platform had to be designed into the new ERP framework , and decisions were made to conduct a more comprehensive integratio n. 16. Information Technology Management : Lessons Learned 16.1 A detailed assessment of require ments for the call centre would have helped to avoid the issue concerning the SIM - box . Additionally, t he contract entered into between SBTS and Coffey was not reviewed by any staff with procurement/contracting experience , th us providing document ary evidence on the condition of the equipment was not included in the contract. This was an omission. 16.2 Coffey should have taken robust technical advice on development of the platf orm and subsequent technical decisions . It would have been better to have written off the costs of the initial failed development after underperformance by the first developer instead of giving the responsibility for the new platform to the same developer . With the redesign of the data collection logic , an entirely new procurement process , to include website, data base and app should have been undertaken . This would have ensured that any new developer would have had to provide a holistic approach to the new pla t form . 16.3 Coffey should liaise more closely with the ACC to ensure that the re are no miscommunications and that the support Coffey provides is in line with the needs and ambitions of the ACC.

3. 3 Introduction The Pay No Bribe (PNB) Programme pilot phase was launched in September 2016 with UKAID funding , and is led by the Anti - Corruption Commission o f Sierra Leone (ACC) . The overall objective the programme is to enable the ACC to design, manage and implement a system for real - time reporting of corruption by the users of services. The system targets the points of service delivery that are routinely prone to bribery and wher e demands for bribes impact heavily on the poorest and most vulnerable, including the growing numbers of urban poor. These are:  Health: User fees for basic services under the Free Healthcare Initiative  Education: User fees in primary education  WASH: Conne ction fees  NPA: Connection fees  Police: Carrying out of police duties Programme activities began initially in 4 pilot districts: Bo, Bombali , Kenema, and Western Area ( Rural and Urban ) . A Baseline Survey covered these districts plus two control district s: Bonthe and Koinadugu . Full implementation of the PNB programme will commence from October 2017 with the aim of broadening and deepening the impact to date, and building a strong momentum for change . Implementation will include Kono as an additional district. Coffey International supports the ACC with the implementation of the PNB . One of Coffey’s commitments to DFID is to produce a Lessons Learned document on the experiences of PNB to date , and this paper is the result of that process. Sources This paper uses information drawn from several sources, including the PNB Sharing and Learning Meeting and report, the End - of - pilot survey ; the DFID Annual Review s 2016 and 2017 ; ACC report; PNB launch and prelaunch reports; monitoring and verification visit reports, The Political Economy Analysis; Coffey internal documents; and reports of conversations with CSO heads; Structure The paper is structured around the different strategic ar eas of design and operation of the PNB programme: Programme Design, Theory of Change/PEA; Knowledge Management; Outreach and Communications; Coffey Capacity and Management ; Sustainability ; the Anti - Corruption Commission’s Roles ; and In formation Technology Management.

14. 14 13.9 At the Kenema meeting, t he ACC presented a summary of MDA responses on PNB repo rts. The zero response of the Ministry of Education to date was noted and discussed. 13.10 Discussions in Kenema also included agreements on communications and outreach strategies, IEC materials and working relationships with CSOs . 14. The Anti - Corruption Commi ssion ’s Role : Lessons Learned 14.1 The breakdown in communications between Coffey and the ACC has been detrime ntal to the progress of the PNB. Two vacancies within the Coffey office have now been filled, 18 and the stre ngthened Coffey team is working closely with the ACC to re - establish relationships and to ensure that the PNB programme is back on track. 14.2 The Strategic Plan and Work Plan are key documents. Coffey should work closely with the ACC to ensure that these are completed as soon as possible. Similarly, a training/ capacity - building plan should be agreed and implemented, focusing particularly on the sustainability aspects of the programme. The ACC should fo llow up proactively with Cof fey if progress slows. 14.3 MDA responses are a key strategic area for the ACC. As noted above, a plan should be developed to strengthen the PNB approach to MDA performance . Central to this will be i) the ACC using its powers under NACS to bring pressure on MD As to address corruption and ii) the ACC and CSOs working together at district and regional levels to develop and strengthen opportunities for change and build momentum for responses from headquarters. 14.4 Strong outreach strategies are required if PNB is to extend across all districts in Sierra Leone . Discussions in Kenema highlighted the fact that the listenership and impact of the current format of radio programmes could be strengthened , along with the quantities and relevance of IEC materials. In accordan ce with its Outreach S trategy, t he ACC agreed to identify the needs of different target groups, to reach out to different groups and districts with appropriate and creative local language programmes carrying punchy, interesting messages. The ACC and CSOs a lso agreed to work together to create lively PNB messages for IEC materials to appeal to a wider range of groups. 14.5 Coordinated M&E across the ACC and Coffey teams will be critical over the next 15 months of the programme , to ensure that robust data is collected and analysed and used to inform implementation ; citizens receive regular feedback on PNB; the media have access to good stories; and pressure for MDAs to improve delivery grows. 15. Information Technology Management : Background 15.1 The PNB platform con sists of a call centre with a free hotline (515), a website, and a mobile application. Members of the public either make reports themselves or are facilitated in doing so by CSO community animators, contracted by Coffey. The p latform faced delays in its la unch, eventually being operationali sed in September 2016. Since the n , over 30,000 reports have been made to the PNB platform, 3,500 of which used the hotline, 900 the website, and 25,000 the app. Logframe targets in relation to numbers of reports were met. 15.2 The call centre receives calls through a SIM - box, which contains SIM - cards from all telecommunication service providers in Sierra Leone. In September 2016, NATCOM 18 The posts of CSO coordinator and Deputy Team Leader

9. 9 7. Communications and Outreach : Background 7.1 Communications and Outreach are an integral part of knowledge management (discussed above) and essential to the awareness raising of citizens, which is a central part of the DFID ToC model . 7.2 The DFID Business Case highlighted a gap in community action and media activity on corruption, and emphasised the role of the media in the publication of information on bribery to increase public discourse on corruption and demand for a robust response. 7.3 The PEA notes that radio is by far the most used source of information in Sierra Leone. TV, internet and even newspapers are less used, especially outside of Freetown and Bo. Afrobarometer data 12 indicates that 64 per cent of people said they would contact the media to complain about government performance or corruption if they had the chance, compared wit h 15 per cent would never do this. 7.4 At the meeting in Kenema , PNB Partners discussed a wide range of communications and outreach activities and methodologies to target different group s , including working with the media (particularly radio); us ing radio more effectively ; use of local languages; use and effectiveness of IEC materials; use of street theatre and cinema, the roles of animators; and ‘ informal ’ and/or social media outreach with groups such as Okada drivers. 7.5 PNB partners noted that there were differences in quality between the activities and reporting of different animators. Not all CSOs had been able to monitor or verify the work of their animators sufficiently . 7.6 PNB is not operating alone: several initiative s are focused on improving service delivery (especially in Health and Education). 8. Communications and Outreach: Lessons Learned 8.1 PNB PE materials and broadcasts are not reaching all audiences and need to be made more exciting and more targeted . This will in volve p artners develop ing more creative approaches to communications, perhaps using PNB songs and jingles to sh are information by nationwide radio (IRN) instead of panel discus s ions . This c ould include a range of citizens’ voices and responses (taking radio to citizens) , celebrating good news stories , sharing MDA responses and/or asking MDA officials to take part in debates with citizens and identifying people with outreach to ordinary citizens (such as religious leaders) or celebrities (such a s footballers) to speak out on PNB messages . Messages should be memorable and tailored to appeal to specific groups . 8.2 In terms of changing behaviours and culture around bribery and corruption, it will be important to target young people. This can be done through music programmes, using songs with PNB messages, and strengthening PNB outreach through social media regular users of social media. Identifying who these groups are and how to include PNB messages in different types of social media will be importa nt during the next 18 months. 8.3 IEC materials should be more creative and more widespread. A larger number of billboards and banners with strong PNB messages and target ing schools, health centres, police posts and other strategic locations will make strong statements. U sing humour and cartoons to tell stories is effective. 8.4 Informal outreach opportunities should be identified . The s e could include work on buses and taxis with passengers and at bus/taxi /Ocada stops, work in lorry parks; street cinema; 12 Round 6 2014 - 2015 www.afrobarometer.com

10. 10 street drama; comedian groups; community voluntary youth and women’s groups and community theatre. 8.5 The quality and consistency of animator activities and reporting should be improved. Animator selection criteria should be developed by Coffey with the CSOs, and each CSO should then review the animators it has and monitor their performance carefully . 8.6 Animator reporting should be made easier to save time and improve accuracy, consistency and comparability . Opportunities to use smartphone/tablet software such as Ko bo Toolbox should be explored. 8.7 Other organisations are working on themes like those covered by PNB , such as local Integrity Committees and Accountability Now Committees . Improved effectiveness would be achieved by collaborating with these organisations, sharing messages with them and perhaps holding joint community meetings where relevant. 8.8 DFID have highlighted opportunities for cooperating with SABI - Strengthening co mmunity - led accountability to improve service delivery in Sierra Leone – implemented by Christian Aid, with Restless Development managing the Communications and Outreach component. CGG is also working on SABI and could lead in supporting closer working wi th PNB partners. 9. Coffey Capacity and Management : Background 9.1 Coffey ’s proposal to DFID for the PNB was led by MSI – Management Systems International. The programme started in July 2014 but was suspended due to Ebola in December 2014. It was remobilised i n June 2015 and a new team leader and CSO co - ordinator were contracted. 9.2 Some activities, including a Baseline Survey and a capacity assessment of the ACC were carried out before t he programme’s annual review in early 2016. Both documents were considered by DFID and ACC to be of poor quality , and a logframe that was produced in December 2015 did not reflect the objectives of the Business Case . As noted in under 2.3 above, underlying research was not carried out. 9.3 The new Coffey PNB Team Leader was replaced in November 2015 and the new person was apparently not provided with necessary background or guidance documents . The situation at that point was not helped by the fact that MSI appeared to be implementing a programme different to that outlined in the DFI D Business Case. 9.4 T he DFID Annual Review in February 2016 concluded that the programme had struggled to establish strong relationships with counterparts and had made little progress towards its objectives. It scored a ‘C’ and Coffey took over from MSI as lead agency . 9.5 Over the next few months delivery picked up although there were still delays around the setting up of the PNB call centre, the PNB reporting platform and the production of key documents. The programme was launched in September 2016 and substa ntial progress was made between then and January 2017, as recognised in the Annual Review, where the programme scored an ‘A’. 9.6 Relationships within the Coffey Team were not constructive however, and not all staff provided to the team perform ed strongly 13 . Relationships between Coffey team members and the ACC suffered, and contracting and/or payment of CSOs was not carried out in a timely or efficient manner. The CSO Coordinator left March 2017 and was not replaced until September 2017 . 13 DFID noted the poor performance of one member of staff in September 2016.

11. 11 9.7 Overall, th e amo unt of time taken for the inception phase was excessive, and slowed down project implementation. This w as due to a combination of several factors including weak project personnel, a poorly performing initial supplier ( MSI), and poor quality working relatio nships between the ACC and some members of the project team. 9.8 CSOs have complained about Coffey’s top - down management style, for example, not being included in budget discussions (especially in relation to budget cuts) and setting of work plans. 10. Coffey Capacity and Management: Lessons Learned 10.1 The slow start to the p rogramme was exacerbated by MSI/Coffey’s failure to grasp the essential vision of the programme, failure to quality assure activities and production of some outputs that were not fit for purpo se. Coffey should pay attention to programme quality and direction, and be prepared to provide clear steers on doing things differently and innovating when required. 10.2 P rogramme teams cannot deliver strongly when staff underperform, but Coffey’s senior management has been slow to pick up on poor performance of staff in the PNB team and slow to address it . S enior management should ensure that staff performance is reviewed regularly , that shortcomings in performance are recorded and that timely ( and occasi onally difficult ) performance management decisions are taken when require d. 10.3 Constant changes of key staff have been disruptive and delaying. Coffey needs to do what it can to ensure robust recruitment, continuity of staffing and smooth management of any handovers to minimise disruptions in financial management and technical delivery . 10.4 Coffey’s internal knowledge management and oversight of key programme documents have been unreliable and should be set in place . This includes maintaining a database of all necessary documents 14 which should be provided to incoming staff to enable speedy and effective programme induction, and made available to e xisting staff to support processes of ground - truthing and checking of direction . 10.5 Del ays in the establishment of acti viti e s encouraged a simplification of programme targets to ensure that they were met 15 . In effect, aspects of the ToC that address more strategic (and more challenging) behavioural changes in MDAs, in the judiciary and in the ACC itself have received less focus . 10.6 Part of the reason that i mplementation was disrupted from April to July 2017 was because CSOs did not have functioning contracts and funds had not been transferred to them. This i s a situation that Coffey must avoid in future . 10.7 Working relati onships are important: Coffey should ensure that there are clear and honest lines of communication with all PNB partners and within the programme team , and that interactions with partners and colleagues are informative and respectful. CSOs should be included in necessary discussions about budget and work - plans and other topics to do with their activities . 11. Sustainability : Background 11.1 The PNB programme was designed for full transition to the ACC by the end of 2018 . This will require all aspects of programme implementation to be carried out in a way that aligns with ACC systems , capacity and resources . Coffey supported the setting up and 14 Among others, the D FID Business Case, ToC; Annual Review d ocuments; PEA 15 See PEA comment under 3.6

13. 13 Radio is the most effective broadcasting medium in Sierra Leone and reaches large numbers of citizens i n most areas. 16 12.5 Currently, the programme has no clear idea as to how reporting levels can be maintained or increased following the end of CSO engagement. Careful consideration should be given to the continued use of the database, whether it is necessary to retain it beyond the end of the programme and, if so, how ACC will manage it . 12.6 A critical aspect of s ustainable impact is the change of culture around MDA service delivery among officials at local level and senior management. This was discussed extensivel y at the PNB Partners’ meeting in Kenema: it will be important to develop a strategy to address MDA performance , highlight successes where they occur and build momentum for change. 12.7 Over the next 15 months, t esting results against the log frame and an upd ated ToC and PEA will help to learn lessons about impact and sustainability and keep the programme on track. I nnovation and flexibility should be encouraged where justified, and backed up with robust experience sharing and learning. 13. The Anti - Corruption C ommission ’s Role : Background 13.1 Th e PNB platform is a key component of the Presidential priorities . The initiative is led by the ACC C ommissioner w ho reports on a fortnightly basis to the President and Ministers in the President’s Delivery Forum. The Commissioner has deployed internal ACC resources, and has also engaged and deployed senior management of the Commission in key PNB areas. 13.2 The Commissi oner has delegated decision - making responsibilities on operational matters to the Deputy Commissioner and the Director of NACS (who is also the Programme Manager of the PNB). ACC staff r esponsible for different areas can respond to queries from partners an d Coffey on PNB implementation. 13.3 The ACC’s 2 nd quarterly report for 2017 highlighted several areas of dissatisfaction with the current functioning of the PNB and Coffey’s management of it. Additional areas where the PNB implementation could be strengthened were identified in the August 2017 PNB Partners’ Sharing and Learning meeting in Kenema. 13.4 The quarterly report raised questions around the s ustainability of the PNB after its completion. Lessons learned in relation to these are contained in Section 12 ab ove. 13.5 The ACC noted particularly that Coffey’s management of PNB had been disjointed and that by May 2017 the number of Coffey in - country PNB staff ha d reduced , despite a recommendation in the Annual Review that Coffey should strengthen the numbers and tec hnical capacity of its staff. There has been no agreement of a Work Plan with Coffey . 13.6 The ACC and Coffey worked together on recruitment for new consultant posts 17 but due to a breakdown in communications with Coffey, ACC received no information on progress with the appointments . Similarly, the appointment of a consultant to support the ACC in the develop ment of an ACC Strategic Plan was not confirmed. 13.7 The breakdown in communications between ACC and Coffey teams meant that the ACC had limited engagement wit h the Coffey team except through the Team Leader. 13.8 The ACC feels that the quality of training provided by Coffey has been poor and has not strengthened capacity within the ACC. 16 See also the Communications discussion under Section 8 17 An M&E Adviser post in the ACC and a Deputy Team Leader post in Coffey

6. 6  MDAs/ District Councils/politicians respond to pressure from public to tackle corruption in target services  Support to Anti - C orruption Commi ssion builds capacity and results in increased sanctions for corrupt practices  Support to judiciary results in follow through with sentencing and recovery  Public Service Pay Reform results in a living wage for officials. 3.4 These assumptions should have been tested on a regular basis and at Annual Reviews, but were not . 3.5 As noted above, the PEA that was originally developed for the PNB programme in early 2016 was considered inadequate . A subsequent PEA was not finalised until May 2016 and was not published on the website or used to check design or ToC assumptions . 3.6 Interestingly, t he PEA notes that the ‘there is no particular PEA - driven reason to doubt the ability of the programme to achieve purely outpu t - level technical results, such as installing a working PNB, and increasing knowledge of corruption issues amongst citizens’ 8 but it goes on to point out that this may not lead to more prosecutions for bribery and corruption. 4. Theory of Change and PEA Les sons : Lessons Learned 4.1 Reference to a ToC and logframe would have helped the direction of travel and bedding in of activities during the first year of the programme ; it would have helped to avoid delays and enabled MSI to learn lessons and develop a nuanced delivery closer to that outlined in the DFID Business Case. 4.2 Similarly, a timelier PEA would have supported more constructive challen ging of the model and questioning of the approach es which were not working (see below) . 4.3 Taking the time to understand and interpret DFID’s vision contained within the ToC would have enabled MSI/Coffey to have i n itiated a focused implementation of PNB earlier , achieved preliminary results and arguably avoided the ‘C’ mark awa rded in the 2016 A nnual Review. 4.4 Arguably, the programme has not evolved or adapted its approaches since its inception , in part due to dela ys, in part due to not having process es for questioning findings, and at times due to weak communications between partners . A n updated ToC and PEA should be developed to help PNB Partners challeng e some of the gaps and assumptions in PNB design . Some specific lessons emerged from the PNB Partners Kenema meeting as well as from the PEA itself :  Assumptions about whether simply raising peopl es’ knowledge levels will have an impact on citizen’s attitudes and actions must be questioned, along with whether traditional approaches to engaging citizens will be effective.  Under PNB, MDA responses to ACC reports has been largely lacklustre: alternative approaches are needed. Some MDA officials may not be committed to using PNB information to address corruption through administrative action and reform measures. Additionally, selected MDAs may have the greatest vested interests in preserving the status quo. 8 Support to Anti - Corruption in Sierra Leone : Political Economy Analysis. Coffey, May 2016

8. 8 6. Knowledge Management: Lessons Learned 6.1 Knowledge/information management is a critical factor in sustainability. Partners at the Sharing and Learning meeting noted that not enough proactive knowledge management had been carried out within the PNB programme and that more effort should be made to generate and share relevant knowledge and information. 6.2 Knowledge management in PN B currently focuses on centralized data collection and management, whereas a model where information flows between partners, between citizens and partners, from citizens to database and from database to partners and citizens would be closer to DFID’s origi nal vision where information would become a tool for citizens to use to hold government to account as well as for government to provide information to citizens . 6.3 There is currently no direct mechanism for sharing good practices and ‘successes’ between PNB stakeholders . Regular experience sharing meetings between partners should be established , and direct links for prompt feedback are necessary (e . g. through social media, WhatsApp etc . ) to share stories with all partners for the website and media, and for animators to feed back to citizens . 6.4 Along with experience sharing meetings, PNB CSOs would like Coffey to organise joint M&E visits with ACC and Coffey , which would provide useful o pportunities for learning and sharing. 6.5 Points cited by PNB Partners includ e the following on the use s to which P NB knowledge and Information could be put :  Providing information in usable formats to all partners and citizens (different languages, pictures, cartoons etc.) .  Developing relevant f eed back to citizens on PNB progress and on MDA commitments in their districts .  Sharing MDA responses promptly with CSOs and citizens .  Enabling citizens to use information to monitor , track and comment upon MDA commitments at district as well as national levels .  Supporting citizens in engaging directly with MDAs at local level .  Using website data to put together analyses and reports on performance by MDA including comparison reports between sectors and districts 11 (for sending to MDAs and to post on the website).  E ncouraging journalists to review data , interview citizens a nd develop stories on aspects of PNB (comparing and contrasting information by sector, district etc.) and/or use reports posted on the website  Collating and posting stories/testimonies through the website or turning into news items in local languages.  Feeding stories and testimonies into a range of radio programmes designed to appeal to different audiences in local languages .  Encouraging d ifferent PNB partners and communities to develop their own i nformation and share with others. 6.6 It is essential to strengthen knowledge management, data collection and information flows in the remaining 16 months of the PNB programme. 11 The DFID PNB Business Case suggests using the i nformation database to collect and publish res ults including comparative data between services, and between districts

4. 4 1. Programme Design Background 1.1 At the time of developing the Business Case, there was strong support within DFID for tackling corruption. It was a Ministerial priority, was set out in DFID’s own Business Plan 2011 - 2015, the DFID Sierra Leone Operational Plan 2011 - 2015 , and was reflected in DFI D’s Partnership Principles for Support to G overnment 1 . In addition, the need for DFID to control corruption is highlighted in the 2011 ICAI report on DFID’s Approach to Anti - Corruption. 2 DFIDSL may have been under some pressure to design an Anti - Corruption programme. 1.2 The DFID Business Case ‘Support to Anti - Corruption in Sierra Leone 2014 - 2018’ identifies a strong culture of impunity in Sierra Leone but claim ed that citizens’ perceptions of government willingness to tackle corruptio n had increased in the three years prior to writing the Business Case. In addition, World Bank Institute Governance Indicators show ed that some progress had been made in control of corruption, albeit from a very low base. 1.3 There are questions around the programme design as set out in the Business Case, and whether it reflected the best available information on anti - corruption practice at the time and whether enough background research had been done . P oints are:  The PNB desig n did not represent a best practice anti - corruption programme. MSI managed the programme for the first two years 3 and did not understand or interrogate DFID’s thinking or vision. Research on the context to establish baselines, understand citizen definitio ns of bribery and identify the most effective style of engagement was due to be carried out during the first three months of PNB implementation but did not take place. The contextual analysis at the design stage was limited.  No Political Economy Analysis (PEA) was carried out as part of the design process or immediately after, and although two versions were subsequently produced by different authors, neither was integrated into programme design or plans . As a result, programme design tended to ignore the subtleties of political power and change and instead promoted easy deliverables such as numbers of reports.  The lack of research meant that preconditions for a programme were not explored sufficiently . The capacity of the media to investigate and report effectively , free from patronage, was not assessed (although some training was envisaged) and the capacity of the CSOs to carry out the tasks required was not review ed .  Limited attention was paid to MDAs in the design , and the strength of the approach in r elation to influencing MDAs and/ or to enlisting their support and cooperation. 4 No incentives were built in to encourage MDA cooperation .  The five sectors selected for the programme were those where demands for bribes impact ed heavily on the poorest and mo st vulnerable groups . The potential 1 All documents cited in the DFID Business Case and Intervention Summary 2014 - 2018, Strategic Case p.7 2 Independent Commission on Aid Impact (ICAI), Section 4.1: ‘ DFID needs to give significantly greater attention to the fight against corruption to manage this increasing risk. ’ 3 Including 6 months out of country due to Ebola. 4 The Business Case outlines a n alternative design option for the programme using a community scorecard approach. This was rejected as being too depend ent on service providers, local power brokers and politicians’ engagement to make it effective . A counter argument could be that engage ment with these groups could have made the programme more effective.

5. 5 effectiveness of these sectors (or others) to act as ‘drivers of change’ in government - wide reform process es wa s not analysed. 5  It was envisaged in the Business Case that an increase in the effectiveness of the prosecutorial capacity of the Commission, would allow ‘ clarity and a searchlight ’ to be shone up on the actions of the judiciary, but the logic and rationale for this was not spelt out in the approach. 2. Programme Design Lessons Learned 2.1 A robust contextual analysis prior to and/or as part of design is essential for a programme like PNB . An analysis of this type (a PEA or similar ) would have enabled a more nuanced approach to the realities of political power and patronage , service delivery reform , the judiciary, corruption behaviour , media capacity , cultural expectations around bribery and the strength or otherwise of civil society . 2.2 It is important in a programme of this type to factor MDAs into the design. Expectations of MDA behaviour change should be grounded in ince ntives and/ or punitive measures likely to challenge the status quo . 2.3 Coffey/MSI should have ensured that the research was carried out and used to i nform programming. The DFID Business Case states that ‘the research phase will determine how to maximise the effectiveness of the programme [and] on understanding the social, cultural and political context, and minimising harm’. The fact that this research was not carried out meant that opportunities were missed to fill information gaps and nuance PNB design , and for PNB to adapt and evolve as experience grew . 2.4 For the 16 months remaining, i t will be i mportant to learn from information being generated by PNB to modify activities and approaches where necessary. 6 3. Theory of Change and PEA Background 3.1 The Business Case contained a ToC and an incomplete logframe based on i) support to citizens and the media in reporting and exposing petty bribery, and ii) sup port to the ACC to increase both petty and grand corruption prosecutions and enforcement of public standards through the provision of technical assistance and financial aid. The two components of the programme were designed to complement each other and max imise the likelihood of enabling citizens to challenge corruption where it directly impact ed on them. 3.2 However, no revision or ground - truthing of the ToC took place and the logframe that was developed by MSI at the end of the first year of implementation b ore little resemblance to objectives set out in the Business Case or to the ToC. 7 The logframe was updated but there were several delays and changes before it was finally approved. 3.3 The ToC had six assumptions:  Citizens participate in debate on corruption and call for action  Underlying Principles and Mutual Accountability Framework incentivise high level political will 5 For example, t he PEA notes ‘there is a risk that there could be too many vested interests in preserving corruption in [PNB] sectors. An alternative approach to consider during the pilot is whether there are other sectors, with extant but less entrenched corruption , that could offer quicker wins.’ (p.15) 6 See also S ection 6 on Knowledge Management 7 MSI claimed not to have received the DFID Business Case or the ToC from Coffey until the end of 2015

7. 7  It is difficult to bring about reform in a situation where public servants are complicit and where internal control measures are usually undermined. Additional approaches m ay need to be developed.  Support (training ) for the ACC implies that a lack of ACC capacity is the major cause of the level of prosecutions , but ACC capacity is only one factor in the GoSL’s ability to secure convictions in corruption cases .  A strong medi a is an impor tant element in the remaining 15 months of the programme, but the media (and individual journalists) must be assessed in relation to their objectivity and their strength to resist pressure to censure certain stories, such as those close to gov ernment. 5. Knowledge Management : Background 5.1 The August 2017 Sharing and Learning Meeting with PNB partners in Kenema agreed that the PNB programme depends on robust management of the knowledge and information generated by activities of all PNB partners. 5.2 Since its inception, the PNB programme has had many opportunities to gather and use information to help citizens and this became very clear in the Sharing and Learning meeting held 23 - 24 August with PNB partners in Kenema. 5.3 The DFID Business Case identified information as a key component of the programme; “ The new data will also offer opportunities for DFID (and potentially collectively with other donors) to use this information in DFID’s influencing dialogue with GoSL over systemic public service reforms. T his would take the traditional ‘I Paid A Bribe’ approaches to a new level of influencing, and ‘close the loop’ between reporting and subsequent change ” 9 . 5.4 DFID also claimed that more available information would lead to a better informed and more vigilant p opulation: “a powerful benefit would be where information not only punishes individuals, but also results in systemic changes, illustrating the strength of citizen action, and tackling the culture of impunity”. 10 5.5 The PNB programme has the potential to gener ate information in the following ways:  Through data generated by citizens’ reports and presented on the website. This is a powerful source of information which can be analysed and used in many ways.  Throu gh meetings between PNB partners regionally or nati onally to share experiences , reflect and learn.  Through meetings with citizens in different contexts (comm unities, schools etc. ) where opinions are given and can be recorded.  Through public meetings with citizens and PNB partners (town hall meetings) and w ith MDA representatives  Through papers and analyses produced by PNB and/ or by others working on similar issues in Sierra Leone or elsewhere.  Thr o u g h monitoring reports and capturing of stories, testimonies and narratives in conversations with citizens.  Thr ough feedback from MDAs and their commitments on changes , and through citizen monitoring of commitments. 9 DFID PNB Business Case 10 Ibid

2. 2 Table of Contents Introduction ................................ ................................ ................................ ................................ ............ 3 Sources ................................ ................................ ................................ ................................ .................... 3 Structure ................................ ................................ ................................ ................................ ................. 3 1. Programme Design Background ................................ ................................ ................................ ..... 4 2. Programme Design Lessons Learned ................................ ................................ ............................. 5 3. Theory of Change and PEA Background ................................ ................................ ......................... 5 4. Theory of Change and PEA Lessons: Lessons Learned ................................ ................................ ... 6 5. Knowledge Management: Background ................................ ................................ .......................... 7 6. Knowledge Management: Lessons Learned ................................ ................................ ................... 8 7. Communications and Outreach: Background ................................ ................................ ................ 9 8. Communications and Outreach: Lessons Learned ................................ ................................ ......... 9 9. Coffey Capacity and Management: Background ................................ ................................ .......... 10 10. Coffey Capacity and Management: Lessons Learn ed ................................ .............................. 11 11. Sustainability: Background ................................ ................................ ................................ ....... 11 12. Sustainability: Lessons Learned ................................ ................................ ............................... 12 13. The Anti - Corruption Commission’s Role: Background ................................ ............................. 13 14. The Anti - Corruption Commission’s Role: Lessons Learned ................................ ..................... 14 15. Infor mation Technology Management: Background ................................ ............................... 14 16. Information Technology Management: Lessons Learned ................................ ....................... 15