Grace Cahill/Oxfam CC BY-NC-ND
U4 Issue | Development cooperationPeople's engagement

Downward accountability in humanitarian aid

The example of UNHCR Uganda

By Sophie Komujuni and Saul Mullard
U4 Anti-Corruption Resource Centre, Chr. Michelsen Institute (U4 Issue 2020:18)

Download as PDF

If aid organisations are accountable to their beneficiaries, there is potentially less risk of corruption. Interventions may also be more efficient and effective with accountability measures in place. Here researchers use a helpline and call centre set up for refugees by UNHCR Uganda to explore lessons around this idea. They demonstrate the importance of involving beneficiaries from the start in designing accountability mechanisms, as well as the need to provide adequate training and resources.

Main points

  • UNHCR Uganda’s Feedback, Referral and Response Mechanism (FRRM) is an example of a downward accountability mechanism. The aim of this research was to see if FRRM might be adaptable as a model to reduce fraud and corruption in other settings.
  • FRRM provides a helpline and call centre to field feedback, provide information, and point refugees/Persons of Concern in the direction of support and services. It also allows reporting of fraud and corruption, and ‘whistle-blowing’ on cases of violence, abuse, and sexual exploitation.
  • An important part of the mechanism is confidentiality. People reporting sensitive cases of fraud, corruption, or abuse can do so anonymously.
  • The literature shows that downward accountability does not always show clear benefits. There is no ‘one-size-fits-all’ design and every mechanism has its challenges.
  • The case study used data from the FRRM’s interagency dashboard. The researchers also interviewed UNHCR and partner organisations, and refugees in Bidibidi refugee settlement.
  • Incidents of mismanagement, corruption, and fraud within its refugee protection system cost UNHCR millions of dollars each year. Through FRRM, UNHCR Uganda is hoping to improve accountability, empower communities, and increase protection.
  • Key challenges to the system’s effectiveness included lack of resources, lack of organisational commitment, unintended consequences, and an absence of guidance and/or expertise.
  • Recommendations include using participatory approaches in the design of downward accountability mechanisms, improving communication methods, and involving beneficiaries in the planning, design, and roll-out. Better training is also important, as well as making downward accountability a requirement of donor funding.

Downward accountability can reduce corruption

The term 'downward accountability' refers to aid organisations being accountable to their beneficiaries. Effective downward accountability measures, directed towards beneficiaries of development or humanitarian assistance, have the potential to reduce corruption risks and improve the efficiency and effectiveness of outcomes. Downward accountability, after all, is not only about being held responsible by others; it also involves taking responsibility for oneself and staying true to the needs of clients, as described by Sawandi and Thomson and Ebrahim. Yet many organisations, including non-governmental organisations (NGOs) and international NGOs, have struggled to fully implement successful downward accountability mechanisms. Empirical studies of selected charities in Europe reveal that selecting a reliable accountability measure, ensuring confidentiality, and obtaining resources to develop such a measure is a difficult undertaking. In effect, downward accountability (accountability to beneficiaries), may be given less attention compared to upward accountability (accountability to donors). This is because of the limited power of the downward stakeholders.

Aid organisations might give downward accountability (accountability to beneficiaries) less attention than upward accountability (accountability to donors)

This U4 Brief asks what positive and negative lessons can be learned about downward accountability, feedback, and resolution mechanisms in general. It also asks what lessons can be learned from the humanitarian context for wider downward accountability in other development sectors. In order to address those questions, the paper takes the example of the start-up period of the UN Refugee Agency (UNHCR) Uganda’s Feedback, Referral and Response Mechanism (FRRM). The FRRM is an interagency common service which seeks to enhance and improve two-way communications between Persons of Concern (PoCs) or refugees and assistance organisations. It is aimed at creating a better informed and more accountable protection environment (see the FRRM data portal).

In October 2018, UNHCR piloted the FRRM in different refugee settlements in Uganda, a beneficiary engagement mechanism that was officially launched in January 2019. Since gaining independence in 1962, Uganda has provided asylum to people fleeing armed conflict and persecution in neighbouring countries. In the recent years, it has been in the limelight for its refugee open-door policy and for hosting the largest number of refugees in Africa. As of March 2019, Uganda was the third-largest refugee-hosting country in the world, with a total of 1,209,476 refugees coming from Democratic Republic of the Congo (DRC), South Sudan, Somalia, Burundi, Eritrea, Sudan, and Rwanda, among other countries.2c2e8ab7e7b5 In July of the same year, the number had risen to 1,313, 802, with an average of 507 asylum seekers arriving each day during that month.98a78d55966b By October 2019, the number had risen to 1,331,565. This extremely large refugee population translates into an extensive humanitarian presence. Indeed, UNHCR had more than 1005e64b66c1791 implementing partners in its programmes by October 2019.

In Uganda, UNHCR’s reputation had been tainted with frequent reports of fraud and corruption involving its workforce and the organisation more broadly. These reports hinted at practices such as cover-ups, collusion with government institutions such as the police and the Office of the Prime Minister to conceal wrongdoing, among others. These are incidents that might impede downward accountability.1752d2a9d9e8 Trivunovicd1ddc6b9c63f observes that donors have become vigilant on issues of corruption in the projects they support. In Ebrahim’s view, this is in part a response to reports of scandals involving NGOs, which have eroded public confidence in them. FRRM provided refugees with options for communicating with UNHCR and its partners,144c73d5d3b1 including a toll-free helpline.

Impressively, FRRM registered a total of 52,749 calls and handled a total of 19,756 individual queries/complaints in the first 12 months from when piloting began. Of these calls, 475 were sensitive reports of fraud and corruption.aaee61f32fbf The evidence demonstrates that the mechanism has been embraced by the refugee community and that people feel comfortable in using the platform to access information, make referrals, and lodge sensitive complaints. In September 2019 alone, some 5,015 calls originating from 25 individual refugee settlements were received through the helpline. With further upscaling of FRRM, along with community sensitisation and improved mobile network coverage across Uganda, the monthly number of calls was expected to rise further.

This U4 Brief draws on empirical data from the FRRM’s interagency dashboard as of July 2019, UNHCR and its partners, and refugees in Bidibidi refugee settlement to answer the questions raised earlier. In the sections that follow, we review literature highlighting the challenges in establishing downward accountability measures. We discuss the methodology used, as well as the rationale for introducing the FRRM. We also briefly touch on some the operational details of the mechanism, including challenges faced during the piloting period. Finally, we analyse the findings and draw a conclusion.

Drawing on past studies

From UNICEF’S U-Report, Action Aid’s Accountability Learning and Planning System, ESAR’ referral mechanism, and case management procedures, legibility, and quantification, to accountability technologies, accountability has become a limitless phenomenon. This trend partly results from donors’ increased vigilance for signs of corruption and other inconsistencies in the projects they fund. As a result, some institutions have embraced a ‘“systems-approach”; i.e. working across multiple sectors (not just within one sector such as education, health, or child protection) and with multiple stakeholders (not just with social workers, but also community volunteers, para professionals, or auxiliary workers at community level’. FRRM seemingly embodies all the strategies highlighted, embracing as it does multiple methods and stakeholders – as demonstrated later in this Brief. UNHCR has previously responded to criticisms by designing techniques such as accountability technologies, like result-based management (RBM), biometrics, and cash based interventions (CBIs). Notwithstanding, establishing an effective mechanism can be a daunting task.

Yuesti, Novitasari and Rustiarini’s study24647a21a762 elaborates on the challenges facing accountability. They argue that weaknesses in accountability can be seen from ‘financial reports that are mostly submitted to the donors, lacking of information about organisations’ activities to the communities and lacking of supervision of the disbursement that disregards the culture of community’.

Accountability is difficult because of the scandals and abuse of power often reported concerning organisations. Thus, there is an increased leaning towards downward accountability or answerability to beneficiaries. Yet, according to Bawole and Langnel, there is no ‘one-size-fits-all’ framework for this and each mechanism comes with challenges. In some cases, systems have either not been sufficiently developed or formalised, or they have serious implementation challenges. This creates a disjuncture between the policy and practice, as we observed during the FRRM start-up period. Related to this, Sawandi and Thomson posit that downward accountability processes tend to be less formalised or institutionalised. Consequently, it is not always clear how organisations realise downward accountability.

Downward accountability sometimes falls shorts of a key element: empowerment of communities. In their research, Bawole and Langnel revealed superficial engagement by communities in project planning; their roles were limited to endorsing pre-prepared plans, decisions, and mundane aspects of the project. Downward accountability may also be difficult to achieve. For instance, it may be challenging to engage beneficiaries, let alone to identify stakeholders willing to participate. Yet, if downward accountability fails to incorporate beneficiaries in the different stages of project planning and management, then interventions will fail to reflect their needs and aspirations. A participatory approach is important.

If downward accountability fails to incorporate beneficiaries in project planning and management, then interventions will fail to reflect beneficiaries’ needs

In summary, and based on Connolly and Hyndman’s view, the ‘effectiveness of downward accountability is not always evident and while it may bring benefits, such as more intensive focus on key goals and greater legitimation, it is difficult to achieve’. Key consideration in the design of accountability systems should involve context specificity, with intended beneficiary vulnerabilities taken into consideration.

Quantitive data and qualitative analysis

The study aimed at understanding and assessing the impact of the start-up of the UNHCR-led interagency Feedback, Referral and Resolution Mechanism. The case study descriptively explored the rationale, achievements, and operational challenges of the project in order to suggest it as a potential model adaptable to fraud/corruption reduction undertakings. The study drew on both quantitative data obtained via the FRRM’s public dashboard (up until July 2019) and qualitative analysis using face-to-face interviews with UNHCR staff and five of its partners.

A field visit to Bidibidi refugee settlement148a5c3b12b9 in Yumbe district was conducted. Here the researchers held a total of ten interviews with community members and their respective representatives regarding uptake. In total, 16 interviews were conducted (from June to August 2019): one with UNHCR’s FRRM coordinator based in Kampala; five with UNHCR partner organisations; and ten with individuals from Bidibidi refugee settlement. UNHCR staff were the key informants. The partners that participated were: Tutapona, an organisation specialising in trauma and psychosocial counselling and healing, as well as Medical Teams International (MTI), the Danish Refugee Council (DRC), Humanitarian Initiative Just Relief Aid (HIJRA), and InterAid Uganda.

The researchers designed interview guides for the key informants and partners. These were informed by prior readings about the FRRM and a theoretical literature review on downward accountability. Interviews lasted between 30 and 60 minutes due to the busy schedules of the respondents (specifically the partners), with notes taken as the interviews were conducted. For most of the respondents in the settlement, the interviews were shorter since many of the participants were not aware of the mechanism. Some of the participants requested confidentiality and this U4 Brief respects that right. In effect, the researchers used the adjective ‘anonymous’ to make reference to such respondents.

Additionally, we reviewed selected literature on downward accountability and the challenges of setting up an accountability framework. This study further supplemented the interviews with grey literature in form of newsletters, fliers, UNHCR monthly updates, Policy Briefs, and newspapers, among others. Some of these materials – like the fliers, concept note, and newsletter – were obtained from the FRRM coordinator. The study further obtained information on the practical implementation of the mechanism, especially from the interviews.

One of the intentions of this paper was to document examples of sensitive cases addressed through the mechanism. However, given the sensitivity of these cases and the confidentiality principles guiding FRRM, it was not possible for us to carry out such a review. In addition, as a risk-mitigating measure, the location of FRRM’s call centre remained undisclosed. As such, it was not possible to conduct interviews with any of FRRM’s helpline agents. These factors combined made it impossible to employ observation, one of the data collection methods used in the study.

We already had a predesigned format for the final document at the start of the study. In effect, themes such as rationale, challenges, a description of how FRRM works etc. informed both the collection and analysis of the interview data. Given the specific nature of this research, we found everything we needed easily from the interviews and grey literature. We also reviewed and analysed data from the interviews alongside the literature.

We also benefitted from comments and correction provided by UNHCR and external reviewers on the first and revised draft of the U4 Brief.

Building on existing accountability tools

Refugees and asylum seekers need critical information in real time. They actively seek updates and advice, but often get a range of misinformation on social media. Setting up a call centre for refugees will immediately address some of these issues. – Joel Boutroue, the UNHCR Representative in Uganda)

The UNCHR Representative in Uganda gave this message during the launch of FRRM in January 2019. Underneath both it and the spirit of mechanism, lies UNHCR’S desire to increase accountability within the refugee response (although this is not made explicit).

The last updated version of the concept note for the establishment of FRRM is dated July 2018, a few months after an audit by the Office of Internal Oversight Services (OIOS) was concluded. The audit released its report later in November 2018. This revealed incidents of mismanagement, corruption, and fraud within the refugee protection system costing UNHCR millions of dollars.e7940ab10257 This specific audit, which covered 18 months from July 2016, allegedly found a conspiracy between staff of UNHCR and the Office of the Prime Ministerce359fd9c1d1 to inflate and exaggerate refugee figures. It also exposed other incidences of mismanagement, such as excessive fuel use by UNHCR vehicles assigned to the Office of the Prime Minister and payments to ‘ghost’ workers, among others.507436b30edb The report’s recommendations included that UNHCR should review its management, corruption, and fraud mechanisms.2161e8f20868

As a tool, FRRM has not sought to ‘reinvent the wheel’, but rather to build upon and enrich existing accountability/ communications tools within the refugee response. This has been achieved through the integration of previously siloed complaints and reporting mechanisms, such as protection desks, suggestion boxes, and community-based structures such as reporting to zonal leaders. FRRM combined these mechanisms with the establishment of the toll-free helpline and its own inbuilt referral network.

Why then does UNHCR value this downward accountability method? Connolly and Hyndman argue that ‘stakeholders commonly perceived as salient (donors) cede power and credit saliency to beneficiaries to support their accountability needs’. On the one hand, credibility is vital for UNHCR to maintain its funders.534a261f56fe Donors have not hesitated to withdraw or cease funding UNHCR activities when scandals have been reported.8babe6a64e6c With this trend, it can be implied that the fourth and fifth categories of FRRM (explained in the next section) are deliberately crafted with a confidentiality component to encourage safe reporting of sensitive issues such as fraud, corruption, sexual exploitation, and abuse. In effect, this also enhances accountability, both upward and downward.

At the same time, looking at the comprehensive nature of FRRM, this downward accountability method also, as the concept note portrays, addresses the need to improve UNHCR’s protection mandate for refugees. Jacobsen and Sandvik note that UNHCR has been through decades of criticism for lack of accountability and not fulfilling its protection mandate. These (accountability and protection) are two obvious goals for the FRRM. In essence, FRRM fulfils both the downward and upward accountability needs of UNHCR.

Creating a better informed and more accountable response

The Feedback Referral and Resolution Mechanism (FRRM) is a multifaceted interagency common service, which seeks to enhance and improve two-way communication between PoCs and assistance organisations. In so doing, it looks to create a better informed and more accountable refugee response. Through the integration of previously isolated complaints and reporting mechanisms, combined with the toll-free helpline, the FRRM aims:

  • to provide a safe, confidential system for reporting misconduct, including sexual exploitation and abuse (SEA), fraud, and corruption;
  • to facilitate increased access to reliable, up-to-date information for refugees;
  • to enable UNHCR and partners to effectively manage feedback, queries, and complaints from PoCs and to provide timely responses to concerns raised;
  • to increase accountability to PoCs, creating a better protection environment; and
  • to strengthen data collection on the needs of refugees.

The interagency helpline element of FRRM was launched in October 2018, with country-wide coverage achieved by March 2019. As of October 2019, the system’s internal referral network boasts an impressive 487 individual focal points working for 38 different organisations across 31 settlements in Uganda. With mapping of additional operational and implementing partners ongoing, the platform was planned to host in excess of 70 referral-receiving organisations and some 800 focal points countrywide by October 2019.

The helpline is open 8am to 5pm, Monday to Friday, and has 12 specially trained agents who speak a total of 16 languages. Upon calling, refugees are greeted with an audio recording requesting them to choose their language, before being directed to the appropriate agent. The agent then proceeds to transcribe the case into English, making it accessible to individual case workers in the field.

Calls received through the helpline are confidential and only escalated to relevant focal persons in the field when the caller gives consent. The referral is then received as an automatically generated email and password-protected hyperlink, which directs the focal point to the individual’s case details. Specific case details are only visible to the designated focal person, as they are password protected and only accessible through the FRRM system interface. In addition, callers wishing to report sensitive issues relating to allegations of misconduct, fraud, and corruption have the option of remaining anonymous. Should they wish to do so, such cases are only channelled to designated senior focal point staff in Kampala.

Helpline agents classify calls received through the helpline into five categories, depending on the nature and urgency of the issue raised:

  1. general feedback, where no feedback is required from UNHCR or its partners;
  2. information requests, which are answered by agents guided by an extensive database of standardised FAQs;
  3. requests for support or services, which necessitate referral to UNCHR or partner focal persons for follow-up;
  4. sensitive cases related to allegations of fraud and corruption; and
  5. sensitive and/or life-threatening cases related to reports of sexual and gender-based violence, sexual exploitation and abuse, violence, human trafficking, and other security issues.d2950d5f478e

Issues can be raised that relate to any of the 15 sectors within UNHCR’s mandate, such as livelihoods, education, registration etc., as well as other issues which do not fit into these sectors. How the agent categorises a case will determine how and who resolves the case, and the timeframe within which it needs to be addressed.

FRRM is first and foremost a communication tool, with the system also providing low-level case management functions. This case management component works in two ways: first, cases received may be given immediate feedback using an extensive database of standardised FAQs; and second, cases may be referred to partners or focal persons within UNHCR. Once a referral is sent, the respective focal person is responsible for initiating follow-up within five working days.22ea162ff72c Each referral-receiving partner and UNHCR office appoints multiple focal points, who are mapped to activities under various sectors of work. This ensures that referrals are automatically and immediately channelled to the best-placed person in the field to respond to the request or query.

All cases received through FRRM are channelled through the system’s online case management platform. This allows UNHCR and partners to track individual case status, referral response times, and the quality of feedback given by helpline agents. FRRM’s interactive dashboard then displays this data by settlement, sector, call category, breakdown of age, gender, and diversity of complainants, and more. The dashboard also highlights allegations of misconduct by location, sector, and organisation. This provides risk management teams with valuable information for tackling fraud and corruption in refugee response.

With up to 1,000 calls currently received per month, and with an increasing call volume every month, these data give a unique insight into the needs and concerns of the community. This provides UNHCR and partners with a powerful evidence base for programming and project design. In turn, this then assists in ensuring that resources are prioritised according to the needs of the community and that the impact of UNHCR’s work is maximised to best effect. Through tracking referrals in such a way and monitoring the quality of feedback given and response times, the FRRM refugee helpline plays an important role in creating a more accountable protection environment and refugee response as a whole.

The FRRM refugee helpline plays an important role in creating a more accountable protection environment and refugee response

UNCHR and its partners were found to have constantly sought to improve and upgrade the system. They had, for instance, decentralised the referral network to the settlement level, increasing the number of focal points in the country by more than 360 individuals. They also added the option for partners to mark referrals as either ‘closed and resolved’ or ‘closed and not resolved’. In doing so, FRRM became more responsive and could measure partners’ capacity to meet the needs of PoCs. In addition to this, UNHCR piloted the establishment of connectivity and charging centres in settlements, where people could have easy access to the internet, better network coverage, and also charge their phones. It was hoped this would lead to greater uptake of FRRM among those without mobile phones.

A key component of this mechanism is the element of confidentiality. Under the mechanism, confidentiality is classified. It is established in the concept note that data can only be given to designated users for the appropriate category, sector and settlement, implying limited access to this information. Callers may prefer to remain anonymous when reporting. This is their right. However, this applies only if there is no follow-up required and if the cases reported are considered sensitive. Cases requiring follow-up dictate a need to obtain a caller’s information to be used in giving him/her a response. For the cases that require investigation (cases often in category 4 and 5), it is expected that confidentiality is maintained – particularly in order to protect the whistle-blower. In effect, no external party, for instance, researchers, can access this kind of information. Later, this Brief will demonstrate that the confidentiality element is a difficult one to reasonably operationalise.

Handling sensitive cases

Sensitive fraud and corruption cases can be reported under category 4. Although category 5 also has a sensitive cases component, these cases are of gross violations or threats to life.bcf83a1dbd68 As at August 2019, some 466 sensitive complaints had been reported under category 4 since its rollout in October 2018.c6af8f70c1d9 This demonstrated a strong level of trust in the system among PoCs.

Callers wishing to report fraud and corruption are directed to one of 12 specially trained helpline agents, depending on their language preference. As mentioned above, the agent then proceeds to transcribe the call into English, before automatically referring the case to the designated preassigned focal person via the system’s client management interface. Cases concerning fraud and corruption committed by POCs are sent to the Senior Protection Officer (SPO), while those concerning Ugandan government staff and UNHCR and its partners are sent to UNHCR’s Risk Management Unit and Internal Investigation Department in Geneva. The referral is then received as an automatically generated email containing a password protected hyperlink, which directs the relevant focal person to the individual case details. All cases are password protected and only accessible to the two focal points outlined above. Upon receiving the cases, the Risk Management Officer is responsible for forwarding the complaint to the IGO (Inspector General’s Office) for investigation while the SPO will pursue investigations in line with set standard operating procedures (SOPs).

Cases related to fraud and corruption are of two types: reports on services and complaints about personnel. In terms of services, for example, if medicine which is supposed to be free (normally labelled ‘Not for sale’) is seen being sold, a complaint can be raised by a Person of Concern with an investigation then carried out.d5f42eeec60c POCs can also raise complaints about misconduct of the staff of an organisation within category 4. Fraud and corruption cases follow the same referral system as the other cases. This basically means that if a case of fraud or corruption by staff of a partner organisation is reported, it is forwarded to that organisation.

In such cases, UNHCR and its partners have to take appropriate measures. UNHCR has an internal system for investigating its staff before it makes a decision.1b9ca2148095 Equally, its partners investigate any allegations internally. Medical Teams, for instance, hires independent investigators or uses its human resources offices to investigate allegations. The independent investigators then conduct a hearing, all the while maintaining the highest level of confidentiality. Disciplinary action can be taken depending on the outcome of the investigation. For most of the partners who participated in this study, the highest form of disciplinary action was termination of services of the person found guilty. Again, this was dependent on organisation’s human resources policies. Other less serious actions, such as issuance of a warning letter etc., may be taken dependent on the gravity of the crime.

Investigation is extremely important, because not all cases reported are genuine.d0977cfc766e Investigations sometimes find baseless accusations and are thus ignored. For example, a complaint was made against a Tutapona staff member. Upon investigation, it was discovered that the accuser was in the habit of making false accusations. UNHCR and the Office of the Prime Minister get involved in investigations of high-level cases of fraud or corruption.16219d2ee2d3 An example was a case in which a member of staff from a UNHCR partner organisation, working in one of the settlements in the West Nile region of Uganda, allegedly stole food meant for refugees. UNHCR became involved in the investigation. Upon inquiry, it turned out that the person had simply mixed up stores where the food was kept. UNHCR dropped the case.cacf0ea013fd

FRRM treats confidentiality with great importance and adheres strictly to UNHCR’s confidentiality policies. As a result, our efforts to obtain resolved cases from the respondents were unsuccessful. An observation of the system was also not possible, therefore. In fact, even the location of the call centre was not disclosed, partly for security reasons.

Nevertheless, such a mechanism will significantly reduce corruption only if there is commitment to make it effective. This commitment is a vital requirement given that investigations are conducted internally. With such a mechanism, organisations are answerable to their beneficiaries. Hence its functionality will depend on willingness to accept criticism and act accordingly.

Commitment and willpower might not be enough to hold the government accountable. A paradigm shift could also be required

A participant interviewed noted that holding the government accountable could be challenging, since UNHCR could not investigate the Office of the Prime Minister.f1c0dfdf8cc2 Related to this, another respondent observed that in a situation like Uganda – where corruption is deeply entrenched across the entire government system – if investigations are done internally, commitment and willpower might not be enough. Rather, a complete overhaul of the system or radical leadership to enforce anti-corruption measures might be necessary.7e289d047d11 Without this kind of commitment, there is a risk of finding what Walsh calls ‘disjuncture between what is enacted and what is represented in relation to accountability’. Or what O’Dwyer and Unerman term ‘rhetoric commitment’ that does not transform into practice. Interestingly, the Office of the Prime Minister was not yet on the FRRM portal at the time of this study, despite corruption scandals having been reported in this government body.16b2b6b2c8a6

On the other hand, using the FRRM system of handling corruption-related cases internally does save the public face of the organisation, while weeding out bad elements. There is no doubt that downward accountability remains a vital consideration for the legitimacy of organisations and for efficient service delivery. Further, individual responsibility, as opposed to an entire organisation being implicated, should play a deterrent role.

Best practice, challenges and lessons

Challenges with the system and lessons learned

The greatest challenge faced by UNHCR and its partners during the FRRM start-up period was lack of resources. The volume of work for its partner organisations may have increased as a result of the system’s introduction. Yet this was not catered for by UNHCR providing additional human and financial resources. The country coordinator, Tutapona Uganda, for example, observed that UNHCR did not offer resources, so it could not hire staff to specifically work on the cases.2e70510e5983 HIJRA, DRC, and MTI all shared similar opinions regarding human and financial resources. While a total of 10,524 referrals were received during the first year or piloting, only 5,395 (49%) of cases had been resolved by the time of the research. In addition, a significant number of these cases were also not addressed within the stipulated time period of five days. This delay in closure rates and response times was in part linked to resource constraints. However, the delay was also intrinsically linked to the fact that FRRM was a newly introduced system which focal points had struggled to adapt to.

When it came to resources, the research found UNHCR Uganda to be chronically underfunded. This has stifled the success of the project and restricted the organisations’ ability to invest in the human resources needed for FRRM to reach its full potential. For instance, as of July 2019, according the monthly update, UNHCR had only 26% of the US$386.2 million it needed.9c349fa1fcbd With Uganda having the largest refugee population in Africa, and with settlements spread across a large geographical area, 350 individual complaints reported each day was definitely a huge task for the organisations. Likewise, this demanded significant funds. For UNHCR, training the hundreds of people on this mechanism was also a considerable task. Different studies, such as those by Bawole and Langnel and the Asia Development Bank, recognised lack of sufficient financial and human resources to be a setback in the enforcement of mechanisms aimed at curbing corruption and ensuring that downward accountability was conducted effectively.

As of August 2019, when the data were collected, the Government of Uganda, through the Office of the Prime Minister (OPM) was not yet mapped as a referral-receiving partner. Yet most reports on corruption and fraud reported seemed to implicate the government more than the partner organisations currently receiving referrals.7dc1b75c0899 Despite demonstrated indicators that the mechanism was developed to arrest some of these anomalies, OPM could only be investigated in other systems. Accordingly, FRRM could be criticised for targeting only the ‘smaller fish’.

While planning to control corruption in aid to tsunami victims in Malaysia, stakeholders highlighted an important issue: protection for whistle-blowers. Where there is a complaint or case reported via the mechanism, and especially if this is corruption related, effectiveness can only be achieved if there is a feeling of security.

‘All calls are treated with the strictest confidentiality and callers are given the option of remaining anonymous should they so wish.’ (UNHCR 2019a)

Accordingly, confidentiality is considered FRRM’s core value, through which refugees’ identities can be hidden should the caller so wish. But the right to anonymity makes investigations harder to conduct. Hence it is the responsibly of UNHCR and partners to instil confidence within the community to report such incidents ‘on the record’.

The UNHCR newsletter (January–February 2019) revealed incidents where people reported being threatened to prevent them using the helpline. This is unfortunately not a new phenomenon: previous whistle-blowers have reported being targeted by staff from UNHCR itself in Uganda and elsewhere.9d2e45202d34 This does point to tensions within the refugee response, as humanitarian workers and government staff may seek to actively stifle take-up of FRRM in order to silence the voice of refugees wishing to complain. This needs to be addressed. Otherwise the situation might reduce uptake of the system and decrease its effectiveness. For example, a respondent in Bidibidi expressed fears of using their personal phone, because the number was registered with the government. Such incidents clearly demonstrate that efforts to increase accountability need to be followed by policies which seek to create as conducive and safe a reporting environment as possible.

The aim of downward accountability should be empowerment of communities through participation. Participation for FRRM can be enhanced through constant information dissemination within the communities. Besides knowledge of the mechanism, it is important for refugees and host communities to have information about their entitlements from the different humanitarian organisations. O’Dwyer and Unerman use the rights-based approach to emphasise that beneficiaries’ knowledge of their rights in an NGO–beneficiary relationship is crucial for claiming those rights. In Bidibidi Zone, where we sampled ten participants, only two people knew about the toll-free helpline. These were leaders in the community, highlighting the need for increased advertisement of the mechanism.

The aim of downward accountability should be empowerment of communities through participation

Information needs to be packaged to reach the intended audience. This means accessibility is as important as comprehension of the message. Multiple methods of communication – including less formal ones like magazines, leaflets, and social media platforms provided/ updated on a regular basis (often the preferred methods of communication) – could speed the achievement of intended goals. Information can be translated into local languages that people understand. The Charity Commission interprets accountability as follows: ‘providing relevant and reliable information to stakeholders in a way that is free from bias, comparable, understandable and focused on stakeholders’ legitimate needs’.

This does not in any way suggest that UNHCR did no advertisement of the system during the period under review. It did in fact hold system demonstrations and FRRM workshops for partners and UNHCR focal points in each refugee settlement. It also held meetings with community leaders in parallel with these. In addition, refugee leaders we spoke to in Bidibidi settlement affirmed that consultations were held with communities there in February 2019. According to one of the leaders, UNHCR, together with OPM, held meetings with the communities and their leaders. Here, every person who could attend, was given the telephone number and instructed to call if they had a problem. This leader was one of the few people we interviewed who knew about the system; however, he did not know how the system worked. As a result, he could not confidently tell people in his area about it.

Further problems arose as a result of inadequate training of local project volunteers. For example, even though a poster adverting the mechanism was displayed at one of the complaints desks in zone one, the volunteers working there were not able to explain how the mechanism worked. Usage of the helpline in the West Nile (where Bidibidi refugee settlement is) remained low, with a 19% decrease in cases received between June and July 2019. However, information dissemination was later scaled up in the region in a bid to improve uptake of the helpline.

FRRM faced implementation challenges and unintended consequences arising out of the mechanism during its start-up. In addition to several difficulties – such as slow adoption of the new mechanism by stakeholders, and poor network coverage and telephone usage – we observed a disconnect at the operational level of FRRM. Many refugees did not have access to a phone and some called using borrowed phones. This complicated feedback from the partners.d9f77ab6794d It was also difficult to contact Persons of Concern given the poor network coverage. Partners were expected to give feedback within five working days. However, all the five partners interviewed said that this had not been possible. For example, we were informed by MTI on how for the entire month of May 2019, the organisation worked backwards in order to clear a significant backlog.ef66a3d07776 UNCHR updates and newsletters showed a similar trend. Yet Bawole and Langnel argue that these are the typical practical implementation challenges that organisations have to face. The FRRM concept note says regarding FRRM that:

FRRM ‘will employ and serve as add-on to the existing community channels and feedback mechanisms, while providing technical support and tools that will enable structured data collection and the provision of support and referral services.’ (UNHCR 2018)

While this was the intention, the start-up practice revealed a trend of bypassing existing channels. Some calls made reported cases that could have been handled by community leaders or the police. Two respondents from the partners interviewed (HIJRA and DRC) found this to be a disturbing trend. They saw that it would cause the system to become clogged with people who were not most in need of assistance.

In Bidibidi, and in all other refugee camps the world over, anti-corruption posters are ubiquitous. They inform refugees that services are free and that wrongdoing should be reported. Complaints desks are also popular, as observed in Bidibidi refugee settlement. The UN established the Inspector General’s Office and provides a mechanism through which reporting can be carried out, as well as emphasising its commitment to confidentiality. Refugees in Uganda are at liberty to report corrupt individuals to the police or to the Office of the Prime Minister. Many organisations also have an independent reporting mechanism.

A common component of most downward accountability measures is their use of technology.6f351048e64a Nonetheless, these mechanisms have similar challenges – such as the bodies responsible lacking commitment to give feedback or hunting down and targeting whistle-blowers rather than giving them protection.f66cbaea6757

A question that requires further study is whether FRRM will manage to address such challenges, to the extent that it ensures both upward and downward accountability?

Conclusion and next steps

Despite the desire for more holistic accountability, this and other research (referred to above) have identified major challenges in achieving meaningful downward accountability. This is even when a stakeholder with seemingly significant power (the primary funder) advocates on behalf of such accountability. These challenges include lack of resources, lack of organisational commitment, unintended consequences, and an absence of guidance and/or expertise.

This U4 Brief draws on data from the start-up of the FRMM. It cannot claim to be an evaluation of the mechanism’s effectiveness with regard to controlling corruption. FRRM is a relatively new mechanism with potential for further development. Indeed, the Brief recommends that future research on this UNHCR mechanism be carried out in a more extensive manner. This research could target more than one refugee settlement, as well as more staff members from UNHCR, the government, and partners.

Mindful of these factors, corruption control measures must draw positivity from FRRM. If they are to achieve effectiveness, they should adopt the successful aspects, such as internal investigation that protects an organisation’s public image, and find solutions to the challenges, as with whistle-blower protection, information sharing, and commitment of resources. As already indicated, commitment and willpower are vital considerations in FRRM. To a certain degree, this has been demonstrated by UNHCR’s partners, who actually investigated and took action when allegations were made.

Beneficiaries’ confidence to continue to participate in accountability measures will depend on them witnessing change arising out of their involvement’

Jordan and Tuijl treat NGO accountability as ‘an issue of plurality based on the need to apply common principles and universal rights in different contexts, as opposed to being an issue of common standards, tool-box techniques or mechanisms that can be applied universally’. In the end, confidence on the part of beneficiaries to continue to participate in accountability measures will depend on the effectiveness of their engagement; that is, on them witnessing change arising out of their involvement. It calls for ‘walking the walk and talking the talk’.


This U4 Issue has shown that in addition to the general challenges associated with downward accountability, there are some important lessons to be learned. These in turn lead to practical recommendations for donors and practitioners.

Key among these recommendations is that periodic studies of how beneficiaries experience the mechanism will be crucial to its success. For example, this research found that some beneficiaries were concerned about using personal telephones, while others did not fully understand the system. Therefore, designers of downward accountability mechanisms should:

  • Use participatory approaches when designing them. This will ensure that beneficiaries know and understand what the mechanism is for and have multiple methods for making complaints suitable to their individual circumstances.
  • Use a variety of communication methods to publicise any mechanism that is developed.
  • Include beneficiaries in the planning, budgeting, and roll-out of downward accountability mechanisms.

The role of donors and other development partners is important to ensure downward accountability. While in this case UNHCR was responding to existing corruption challenges in Uganda, general donor concerns over corruption have helped to create space for the development of FRRM. Therefore:

  • Donors should consider making downward accountability a requirement of funding.
  • Beneficiary feedback (beyond case studies and anecdotes) could be made a requirement when reporting to donors.
  • Donors should accept that reporting of corruption via downward accountability mechanisms is a positive indication of the functioning of the mechanism in contexts of systemic corruption. This is provided that donors handle the corruption cases that arise adequately.

Organisations that have established downward accountability mechanisms should ensure that frontline partner staff receive proper and regular training. They should:

  • Establish routine training of partner staff on the rationale and operational aspects of the mechanism. This will ensure they can pass this information on to beneficiaries.
  • Include beneficiaries in the development and roll-out of training and information campaigns related to the mechanism.
  1. UNHCR Global Focus 2019.
  2. UNHCR 2019b.
  3. UNHCR Uganda 2019.
  4. 2011.
  5. Hayden 2019a.
  6. Currently, 42 partners are referral-receiving organisations. There are plans to expand on this number to cover 80 operational partners (Interview with FRRM coordinator, 24 July 2019, UNHCR offices, Kampala).
  7. UNHCR 2019a.
  8. 2016.
  9. Bidibidi refugee settlement is located in Yumbe, in the West Nile region of Uganda. It hosts refugees from South Sudan. The settlement is one of the largest in the world, with more than 280,000 refugees. According to the UNHCR FRRM January–February 2019 newsletter, Bidibidi is among the refugee settlements with the least number of people using FRRM.
  10. Okiror 2018.
  11. Biryabarema 2018.
  12. Biryabarema 2018.
  13. The Office of the Prime Minister is the government body responsible for refugees in Uganda.
  14. See, for instance, Kisakye 2019.
  15. Ibid.
  16. UNHCR 2019a.
  17. Interview with FRRM coordinator, 24 July 2019, UNHCR offices, Kampala.
  18. UNHCR 2019a.
  19. UNHCR 2019a.
  20. Interview, MTI communications officer, 30 July 2019, MTI offices, Kampala.
  21. Interview, MTI communications officer, 30 July 2019, MTI offices, Kampala.
  22. Interview, anonymous, 10 September 2019, Kampala.
  23. Interview, anonymous, 10 September 2019, Kampala.
  24. Interview with communications officer, Medical Teams International, 30 July 2019, MTI offices, Kampala.
  25. Okiror 2018.
  26. Interview, anonymous, 10 September 2019, Kampala.
  27. Interview with HIJRA focal person, 6 September 2019, HIJRA offices in Kampala.
  28. Interview with Country Director, Tutapona, Uganda, 25 July 2019, Tutapona Offices, Kampala.
  29. UNHCR 2019b.
  30. Biryabarema 2018.
  31. Hayden 2019a.
  32. Interview, anonymous, 9 September 2019, Kampala.
  33. Interview, anonymous, 9 September 2019, Kampala.
  34. See, for instance, the story of the refugee whistle-blower, Gratien Zimy Ntezimisi, in Nakivale refugee camp; or of the Somali woman, Hamdi Abdullahi, in Dadaab, Kenya; or the Sudanese refugees.
  35. See, for instance, the use of biometric verification as used by the World Food Programme.

This work is licenced under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International licence (CC BY-NC-ND 4.0)