Conflict Map
11 December 2023

Competing armed groups pose new threat to Rohingya in Bangladesh

Growing competition between armed actors, a surge in violence and crime, declining donor support and restrictive government policies are pushing Rohingya refugees in Bangladesh towards a crisis point.

An aerial view of Kutupalong refugee camp near Cox’s Bazar, Bangladesh. (Photo: Azim Khan Ronnie/Alamy).
An aerial view of Kutupalong refugee camp near Cox’s Bazar, Bangladesh. (Photo: Azim Khan Ronnie/Alamy).

More than six years after the Myanmar military perpetrated a ‘textbook case of ethnic cleansing’ against the Rohingya population of northern Rakhine State, killing at least 6,700 and pushing more than 700,000 across the border into Bangladesh, the Rohingya refugees are at their most vulnerable since the crisis began. Driving this insecurity is an escalating conflict between Rohingya militant and criminal groups.

The violence has been exacerbated by the role of Bangladeshi security forces, who have contributed, in part deliberately, to the rise of armed actors in the camps. Moreover, policies put in place by Dhaka prevent Rohingya refugees from pursuing education and earning sustainable livelihoods. Combined with a decline in donor support, those policies risk fuelling a vicious cycle of insecurity.

As the situation in Bangladesh becomes increasingly untenable, the Rohingya refugees are left with no good options. The camps are increasingly miserable and desperate. Dhaka is unwilling to consider integrating the Rohingya into Bangladesh and most Rohingya have no desire to stay indefinitely anyway. Bhasan Char, a US$300 million development built by Bangladesh on a silt island in the Bay of Bengal to host Rohingya, has fallen out of favour with Rohingya because of its isolation and poor services, and now looks like an afterthought on the part of a Bangladeshi body politic pressing for repatriation. Third- country resettlement remains out of reach for all but a lucky few, and with rare exceptions the Rohingya are unwelcome throughout the region.

Across the border, Myanmar as a whole and Rakhine in particular are vastly changed from when the Rohingya fled in 2017. The military responsible for their expulsion is now back in power in Naypyidaw, locked in a violent struggle with opposition forces that continues to escalate across the country. In Rakhine, conflict has resumed between the military and the Arakan Army (AA), a powerful armed group that has gained significant control over the state but whom the Rohingya do not trust. With few indications that a durable peace will emerge, the underlying conditions that forced the Rohingya to flee Rakhine largely persist, frustrating prospects for safe returns to their home villages.

Bangladesh’s approach to the crisis — still focused on repatriation, as it has been since 2017, and predicated on a misplaced belief that Myanmar can be compelled to accept returns — has complicated prospects for future repatriation and risks significant blowback within Bangladesh and in bilateral relations with Myanmar. Bangladesh’s support — both tacit and overt — for Rohingya armed actors has facilitated the rise of Rohingya militant groups at the expense of a more moderate, non-violent and legitimate Rohingya leadership. Moreover, policy restrictions on Rohingyas’ ability to pursue education and earn livelihoods have contributed to their growing desperation and a marked decline in donor support, which could potentially push the Rohingya into the arms of criminal and militant groups, not out of affinity but out of necessity. These policies also leave the Rohingya poorer and worse equipped to return to Myanmar in the future: a discourse dominated by Rohingya militancy will make repatriation unattractive to the current and future powers-that-be in Rakhine and complicate reconciliation with the Rakhine community, a critical element in any sustainable return process.

Bangladesh’s forthcoming elections present a rare opportunity for a reset, though hope for a shift towards greater policy openness should be tempered by historical precedent.

Cycles of exodus

On 25 August 2017, the Myanmar military began a series of ‘clearance operations’ against the Rohingya population in northern Rakhine State. The attacks, if not unprecedented, were unparalleled in scale and marked the culmination of decades of persecution, discrimination and dehumanisation in official policy and rhetoric. The military and the National League for Democracy government led by Aung San Suu Kyi argued at the time that the actions of security forces were a legitimate response to coordinated attacks by the Arakan Rohingya Salvation Army (ARSA), a poorly organised and ill-equipped Rohingya militant group. The violent response, however, was grossly disproportionate to the threat posed by ARSA, and Myanmar military documents show that the campaign was planned in advance and executed with the intention of forcing the Rohingya population into Bangladesh.

The 2017 refugee crisis was not the first time Bangladesh had opened its borders to the Rohingya. In the late 1970s and again in the early 1990s, some 200,000 and 250,000 Rohingya, respectively, were forced to flee to Bangladesh as a result of Myanmar military operations. In each instance Myanmar, then as now under a military dictatorship, quickly reached bilateral repatriation agreements with Bangladesh.

The Rohingya were compelled by Bangladesh to return to unchanged conditions in Rakhine, thus perpetuating ‘a cycle from exodus to repatriation, to intensified persecution, and to exodus again’.

In 1978, summarising the Bangladeshi government’s approach to the Rohingya, a senior official said ‘we are not going to make the refugees so comfortable that they won’t go back to Burma’. Deprived of food aid, some 12,000 Rohingya died in refugee camps in less than a year, a mortality rate more than eight times higher than in the rest of Bangladesh. The threat of starvation, combined with physical violence and intimidation by Bangladeshi security actors, ultimately compelled most Rohingya to return to Myanmar. A similar scenario played out in the 1990s after Bangladeshi officials withheld aid from the refugees, causing famine-like rates of acute malnutrition. Moreover, in just a few weeks in 1992, Bangladeshi police and soldiers killed at least 20 and injured dozens of Rohingya who were protesting against repatriation. As in the late 1970s, most Rohingya were coerced into returning to Myanmar within a few years.

Though Dhaka ultimately provided shelter to Rohingya fleeing persecution in 2017, Bangladesh’s leadership continues to view its previous repatriation efforts as successful and has responded to the current crisis in ways that mirror prior approaches. For example, in 2017, Bangladesh quickly agreed to bilateral repatriation instruments — 1 2 3 — with Myanmar, modelling the agreements on that of 1992. Evoking the language of prior repatriation efforts, Bangladesh has consistently called for the ‘early repatriation’ of the Rohingya, as it did in the 1970s, and sought to make the Rohingya in Cox’s Bazar uncomfortable by imposing restrictive policies to signal that their stay would be temporary. Bangladesh has also tried to leverage international pressure on Myanmar to accept returns.

Myanmar’s military and civilian governments have sought to expel the country’s Rohingya population during rounds of violence in the 1970s, 1990s and 2010s.

The latest exodus began on 25 August 2017 after poorly armed fighters from the Arakan Rohingya Salvation Army (ARSA) launched attacks on police posts in Rakhine State. Security forces responded with brutal ‘clearance operations’, pushing 742,000 Rohingya across the border into Bangladesh.

Data source: UNHCR

As of August 2023, more than 963,000 Rohingya are registered as living in refugee camps in Bangladesh, with the majority spread across five major settlements near the border with Myanmar.

Data source: UNHCR. Figures do not include unregistered Rohingya.
Since 2020, more than 30,000 Rohingya refugees have been relocated to Bhasan Char, a US$300 million camp built by Bangladesh on a silt island in the Bay of Bengal.

Though an estimated 630,000 Rohingya remain in Myanmar, persecution there and trying conditions in Bangladesh have pushed tens of thousands of Rohingya to other countries in the region, where they are often unwelcome and unable to register as refugees.

Data source: UNHCR. Figures do not include unregistered Rohingya.
Loading map ...

Deteriorating conditions and challenges for the Rohingya

The refugee camps in Cox’s Bazar would be a difficult place to live even in the absence of restrictive policies. Over 900,000 Rohingya, more than half of them children, are crammed into 33 camps, with more than two-thirds of the Rohingya population residing in the Kutupalong mega camp. Though the settlements are a maze of single-story bamboo and tarpaulin shelters, the population density is nearly five times that of Singapore. The overcrowding, along with unsanitary conditions and inadequate access to clean water, fuels the outbreak of diseases, including scabies, dengue fever and water-borne illnesses including cholera. The elderly and the disabled face significant challenges in navigating the camps, and also encounter barriers when trying to access basic services, as do women-headed households. Sexual and gender-based violence, including intimate-partner violence, is widespread, as are child marriage and human trafficking. In the dry season the camps are unbearably hot, especially for women and girls confined to sweltering shelters. Though the months-long monsoon brings cooler temperatures, it is bookended by cyclone seasons and accompanied by deadly flooding and landslides. Climate change is set to make things worse. With current conditions exacerbating the trauma they suffered in Myanmar, the refugees suffer from acute and chronic mental-health stressors.

Compounding the quotidian challenges of camp life, Bangladeshi government policy restricts Rohingyas’ access to education, freedom of movement and ability to work to support their families. Basic services are inadequate and often of poor quality. A non-signatory to the UN Refugee Convention, Bangladesh does not refer to Rohingya refugees as such, but as ‘Forcibly Displaced Myanmar Nationals’ (FDMNs), a euphemism particularly grating to a population long denied the right to self-identify as Rohingya in Myanmar. Rohingya voices are sidelined both politically and in the humanitarian response. And though they are heavily reliant on humanitarian aid, a funding crunch, set to worsen in 2024, is already taxing Rohingyas’ declining physical and mental well-being.

Rising insecurity and violence in the camps

The deteriorating security environment is compounding the challenges facing the Rohingya in the camps. There have already been at least 76 Rohingya-on-Rohingya murders in the camps in the year to 31 October, more than double the number of killings that occurred in 2022. Should this rate continue until the end of the year, the total number of murders in 2023 will surpass the figure of 91 recorded in the four-year period between 2018 and 2021. It would also make the intra-Rohingya homicide rate double that among Bangladeshis in Cox’s Bazar, and nearly five times higher than for Bangladesh as a whole.

In addition to becoming more frequent, violence in the camps is increasingly carried out during the daytime, driving fear and the possibility that innocent bystanders, including children and humanitarian workers, will become victims. Kidnappings and abductions for ransom, which are probably underreported, have increased from a few dozen last year to more than 100 per month in 2023 and are usually accompanied by beatings and torture. Though Cox’s Bazar District has long been a hub for drugs trafficked from Myanmar, the trade — and the corruption that facilitates it — has flourished since 2017, in part because of an explosion in production in Myanmar after the 2021 coup and the growth of Bangladesh and India as markets for consumption. Although it is home to less than 2% of Bangladesh’s population, Cox’s Bazar has accounted for more than half the methamphetamine seizures reported nationwide over the past three years.

Since 2017, Rohingya security in the camps has been defined by three distinct phases: relative security until 2018; domination by ARSA from 2018 to 2021; and since then, contestation between ARSA and other Rohingya militant and criminal groups — primarily the Rohingya Solidarity Organisation (RSO), a previously dormant militant group, founded in the early 1980s, that has long-standing links with Bangladeshi security forces.

At least initially, ARSA enjoyed some popular support among the Rohingya in Bangladesh, who perceived the group as providing protection and a voice to refugees even though its ultimate intentions remained unclear. Over time, however, camp residents reported that ARSA became increasingly predatory, asserting power through intimidation and violence. ARSA was referred to in the camps as the ‘night government’ because of its dominant presence in the hours after humanitarian workers and Bangladeshi officials had departed. ARSA activities included camp patrols, forced recruitment, threatening female Rohingya students and NGO volunteers, and taxation of informal shopkeepers and NGO volunteers at rates as high as 40% of their earnings. ARSA operatives also sought to control Rohingya civil society, including by threatening and attacking alternative sources of authority such as majhis, the Rohingya camp leaders appointed by the Bangladeshi authorities.

Some sources alleged that ARSA enjoyed cosy relations with Bangladeshi security agencies — including the Armed Police Battalion (APBn), the Directorate General of Forces Intelligence (DGFI) and National Security Intelligence (NSI) — even as officials repeatedly and implausibly claimed that there was ‘no ARSA’ in Bangladesh. It is likely that officials provided ARSA with space to operate in the expectation that the group would directly or indirectly support various Bangladeshi policy objectives, including Rohingya relocation to Bhasan Char and repatriation to Myanmar.

By 2021, however, Rohingya tolerance for ARSA was waning, especially after the group was implicated in the September 2021 assassination of Mohib Ullah, a moderate Rohingya civil-society leader who commanded both international recognition and popular support within the camps. Bangladesh’s security and intelligence agencies were also becoming frustrated with ARSA, having realised the group was making efforts to block, rather than facilitate, repatriation. Then, in November 2022, a DGFI officer was killed, allegedly by ARSA, during an operation in No Man’s Land, an unofficial Rohingya camp between the Bangladesh and Myanmar border fences from which ARSA operated.

No Man's Land before and after it was cleared by RSO fighters. Copyright Planet Labs PBC.

Mohib Ullah’s assassination and the death of the DGFI officer eventually pushed Bangladesh to redirect its support towards the RSO, which has gradually increased its presence in and around the camps. It has probably received support from the Bangladeshi security agencies, particularly the DGFI, and its battle for control over the camps with ARSA has been the main cause of the marked increase in killings this year. In January 2023 the RSO launched an operation to clear ARSA from No Man’s Land, reportedly with support from the DGFI and Bangladesh’s Rapid Action Battalion (RAB). Bangladesh’s APBn and RAB have also stepped up arrests of ARSA leaders in the camps in recent months. Yet while ARSA is under increasing pressure from the RSO and the Bangladesh government, the group remains well embedded within the camps. Ongoing contestation between ARSA, the RSO and other smaller groups is likely to drive continued insecurity for the Rohingya.

Armed groups operating in refugee camps

The Arakan Rohingya Salvation Army (ARSA) is a Rohingya militant group that formed in northern Rakhine State in 2016. It carried out attacks on security forces in October 2016 and August 2017, prompting crackdowns and the exodus of Rohingya. It has sought control over the Rohingya refugee camps in Bangladesh.

The Rohingya Solidarity Organisation (RSO) is a Rohingya militant group initially active between 1982 to 1998. The group began to reemerge in Bangladesh in 2021 with support from Bangladeshi security agencies.

The Nobi Hossain Group, which also calls itself the Arakan Rohingya Army, is an armed criminal group focused mostly on drug trafficking. Nobi Hossain is opposed to ARSA and nominally aligned with RSO.

The Munna Group is an armed criminal group formerly affiliated with ARSA and known for its links to drug trafficking.

Islami Mahas is an armed Rohingya group opposed to ARSA and aligned with RSO. In contrast to other Rohingya armed groups, Islami Mahas appears to espouse a more Islamist ideology.

Compounding the rising insecurity, Bangladeshi security and intelligence forces have demonstrated an improving, albeit limited, capacity to counter militancy and criminality throughout the refugee settlements. Deployed in 2020 to serve as the primary policing force within the camps, the APBn faces a number of challenges, including difficult terrain and a trust deficit that impedes intelligence gathering. Furthermore, the total deployment of 2,000 officers is insufficient to provide effective policing, and the force is not mandated to take case reports or to investigate crimes, creating gaps in policing, access to justice and community trust. The conduct of the force has also drawn criticism, with reports of indiscriminate arrests, corruption and the improper release of those allegedly involved in criminal or militant activity.

Armed groups control refugee camps
Hover a camp on the map or a label below to see where each group operates as of November 2023.
Islami Mahas
Nobi Hossain
No data
Pattern tile example controlled
Pattern tile example contested
Pattern tile example present
Loading map ...

Dhaka’s containment policy

Bangladesh has also employed a policy of containing the Rohingya and associated security concerns within the camps, exemplified by the installation of barbed-wire fencing, watchtowers and checkpoints throughout the camps. According to a report by the Youth Congress Rohingya, a refugee-led advocacy organisation, the APBn uses coercive measures to enforce this containment policy, including arbitrary detention, abuse, extortion and the threat of forced relocation to Bhasan Char. With few exceptions, Bangladeshi officials block Rohingya from relocating outside the camps, even in instances of credible death threats. UNHCR and the international community have not been effective in providing or advocating for more proactive solutions. As a result, the Rohingya are only permitted to relocate to other camps, to Bhasan Char, or to a camp-adjacent transit centre where some Rohingya have been languishing in relative isolation for over a year.

Bangladesh’s containment policy generates further risks unique to refugees. During a massive fire in 2021, one of hundreds of suspected cases of arson in the camps, at least 15 Rohingya were killed and 45,000 displaced. On that occasion, barbed-wire fencing prevented residents from escaping and impeded the efforts of fire and rescue services. More recently, Bangladesh’s home minister said Rohingya would be prevented from leaving the camps as superstorm Cyclone Mocha approached Cox’s Bazar in May (it eventually passed to the south, devastating much of central Rakhine State). Cyclones pose a major threat to the camps, where shelters are poorly built and flooding and landslides are common.

Rohingya refugees search through the remains of their shelters following a massive fire at the Kutupalong Camp 11 in Cox’s Bazar, Bangladesh, 5 March 2023. The fire affected as many as 16,000 refugees. (Photos: Supplied).

Bangladesh’s control over Rohingya civil society and efforts to promote repatriation

The first couple of years after 2017 saw Rohingya civil-society groups flourish in Cox’s Bazar, where the environment was much more permissive than in Rakhine. Many groups focused on aspects of the humanitarian response or on establishing and maintaining community-led schools, but there was also space for more political organisations to take root, with the Arakan Rohingya Society for Peace and Human Rights (ARSPH) led by Mohib Ullah emerging as the most prominent. Though such groups could have served as partners to Bangladesh in the pursuit of repatriation, Bangladesh has instead restricted their activities and sought to co-opt them to advance repatriation on its own terms.

Rohingya civil-society groups never had formal permission to operate within the camps, but Dhaka did not seek to control them until the ARSPH organised massive demonstrations to mark ‘Genocide Day’ on 25 August 2019. Organised with the approval of Bangladeshi civil servants based in Cox’s Bazar and attended by 200,000 people, the demonstration provoked a strong backlash among the Bangladeshi public. In response, civil servants seen as sympathetic to the Rohingya were summarily transferred, while the NSI cracked down on Rohingya civil-society organisations in the camps, including by closing the ARSPH’s office. Since then, civil-society groups and public demonstrations within the camps have been tightly controlled by the Bangladeshi authorities. The outbreak of COVID-19 in early 2020 provided a further rationale for limiting not only the activities of civil-society organisations in the camps but also visits to the camps by diplomats, human-rights advocates and humanitarian organisations. Although COVID-19 restrictions have largely eased since then, most trips by foreigners to the camps are still carefully managed: Rohingya interlocutors are hand-picked by the Bangladeshi authorities or by UN and humanitarian agencies, which limits the influence of civil-society groups while elevating those that may not effectively represent the views of the refugees.

Though ARSA was well established by 2020, and its members already implicated in the killing, abduction and harassment of prominent Rohingya leaders in the camps as well as in Rakhine, the assassination of Mohib Ullah in September 2021 further raised the threat level against civil-society organisations active in the camps. While some Rohingya groups have been able to carry on with education and humanitarian activities, would-be leaders are reluctant to assume public roles, knowing that the Bangladeshi authorities, UNHCR and other international stakeholders cannot protect them should they become targets. This dynamic constrains the refugee community’s ability to organise around a multitude of issues.

In 2022, Dhaka sought to fill the vacuum of civil-society leadership by manipulating Rohingya groups in ways that would advance its own interests. Beginning with World Refugee Day in June 2022, the NSI encouraged Rohingya to organise a ‘go home’ campaign. The rally was the first significant demonstration Bangladesh had allowed in the camps since the 2019 Genocide Day protest. Subsequent demonstrations to mark Genocide Day in 2022 and 2023 purportedly involved efforts by the NSI to organise meetings among Rohingya civil-society leaders, providing them with protest banners, curated messages and safety guarantees. Moreover, multiple sources indicate that the NSI and DGFI have played a direct role in the establishment of several other Rohingya groups, including the Arakan Rohingya National Alliance, a coalition of Rohingya-diaspora activists with links to the RSO that enjoys relatively unrestricted access to the camps, as well as the recently established ‘Forcibly Displaced Myanmar National Representative Committee’, which, unlike other Rohingya civil-society groups, has been allowed to open an office and hold public events. Bangladeshi intelligence officials have also hand- picked Rohingya refugees to meet visiting delegations from Myanmar and to join a ‘go and see’ visit to Rakhine State.

Rohingya refugee voices — especially those of Rohingya youth — are almost completely absent from any discussions, let alone decision-making, about repatriation, justice and accountability, or humanitarian issues. Though there has been robust engagement around localisation in the humanitarian response in Cox’s Bazar, for example, the focus has been entirely on shifting power and resources to Bangladeshi organisations, rather than to the Rohingya who are best placed to understand and respond to their community’s needs. The threats towards, dismantling of and control over Rohingya civil society in the camps are emblematic of systematic efforts to disempower the Rohingya community as a whole, creating yet another gap that Rohingya militant groups have sought to fill.

Restricted access to education and livelihoods

Bangladesh’s control of the Rohingya refugees extends beyond civil society to aspects of everyday life. Bangladesh dictates what materials Rohingya can use to construct their shelters, prohibits them from owning Bangladeshi SIM cards (though many nonetheless do) and forbids them to open local bank accounts. Moreover, it limits Rohingyas’ freedom of movement, access to formal education and opportunities to earn a living, all of which are intended to underscore the temporary nature of their stay in the country.

In 2017, Bangladesh almost immediately signalled a restrictive approach to the education of newly arrived Rohingya. Instead of allowing accredited education or even learning based on a standard curriculum, Bangladesh and humanitarian partners established a network of ‘temporary learning centres’ across the camps, designed only for those up to age 14. Rohingya children attended learning centres in just two-hour shifts and the quality of instruction was poor, viewed by many as more akin to daycare. Access to higher education was and remains even more limited.

With few other options, Rohingya refugees established an extensive network of community-led and home-based education centres, which grew to serve tens of thousands of students. Many of these informal schools were run by educated and respected members of the Rohingya community, who taught in the Burmese and Rohingya languages using unauthorised versions of the Myanmar curriculum in exchange for a nominal fee. Despite their popularity, Bangladesh forced many of these schools to close. Bangladeshi officials justified the closures on the pretext that the schools were operating without permission, were charging for tuition and could radicalise Rohingya youth. In reality Dhaka’s main concern was probably that access to quality education might prolong Rohingyas’ stay in the camps, given that access to education in Rakhine was even more difficult.

Though UNICEF now claims that 300,000 Rohingya are studying under the Myanmar curriculum within sanctioned learning centres, actual attendance rates are likely far lower and the quality of education has been questioned, with nearly 60% of students saying it is no better than what existed in 2017. And, like their predecessors, these educational programmes are unaccredited and provide few pathways towards higher education. Without significant improvements to current educational offerings, Rohingya fear the potential for successive ‘lost generations’.

Bangladesh also formally prohibits the Rohingya from working in the country, taking the position that their basic needs are met through humanitarian aid and that allowing them to work will encourage them to prolong their stay and undercut wages for Bangladeshis. Though Bangladesh does allow Rohingya to participate in cash-for-work schemes and to serve as paid ‘volunteers’ for humanitarian agencies, it limits the number of positions available and dictates rates of pay where even the most highly skilled Rohingya workers receive only US$0.77 per day.

Despite the restrictions, informal markets and shops have sprung up throughout the camps, providing refugees with access to goods and services from fresh produce to mobile-phone repairs. Moreover, despite Bangladesh’s official policy of containment, many Rohingya manage to find work in the informal economy outside the camps, largely in manual labour. These earnings provide a vital lifeline, supplementing the humanitarian-aid package that, by itself, has been insufficient to cover Rohingyas’ expenses (mainly food and medicine, plus occasional costs such as dowries or the purchase of new clothing for religious holidays).

A Rohingya man carries construction material in Camp 10, 9 September 2022. (Photo: Supplied).
A Rohingya man carries construction material in Camp 10, 9 September 2022. (Photo: Supplied).

Concerned that Rohingya would be disincentivised to return if they found employment in Cox’s Bazar, and believing that the flow of cash through the camps fuels criminality rather than diminishes it, Bangladesh has sought to limit the refugees’ income-generating opportunities. Though due in part to reduced funding, there are now fewer cash-for-work and volunteer opportunities. The authorities have arbitrarily demolished thousands of Rohingya shops in the camps, without warning, and regularly round up Rohingya who leave the camps to search for work without permission. In just one week in September 2023, for example, some 500 Rohingya were detained for travelling outside the camps. Though humanitarian agencies have pursued creative solutions and some livelihood projects are being quietly piloted in the camps, the combination of restrictive government policies and declining donor assistance makes it unlikely that the Rohingyas’ basic needs will be met in the coming years. The lack of opportunities to earn income inside or outside the camps could facilitate recruitment efforts by comparatively well-heeled criminal and militant groups that pay a monthly salary. Moreover, if Rohingya were able to work, they could purchase for themselves some of the items currently being provided by donors, which could also lead to humanitarian aid being better targeted towards those most in need.

Declining rates of international assistance

Until recently, international donors, particularly Western countries, had generously funded the annual Joint Response Plan (JRP), which sets humanitarian objectives, priorities and budgets for the Rohingya refugees. Between 2018 and 2022, donors provided an average of US$650m per year, about 70% of the JRP’s requested budget. In 2023, however, donors have provided just US$407m as of November, or 46% of the budget sought. Although the United States and United Kingdom announced new funding packages in September and October, a recent UN-organised high-level meeting on the Rohingya netted few new funding commitments and the 2023 JRP is still unlikely to be as well funded as in previous years. To make matters worse, further cuts are expected in 2024.

Although they have probably provided some support bilaterally, regional actors are unlikely to fill this emerging gap. Thailand and the Philippines are the only ASEAN countries that have contributed to the JRP since 2017, providing a total of less than US$1 million between them. China, meanwhile, has provided only US$400,000, with the entire sum allocated in 2020. India has made no contribution at all. Though it has included brief references to the Rohingya in its formal statements, ASEAN remains largely disengaged. The ASEAN Coordinating Centre for Humanitarian Assistance on Disaster Management has done little to follow up on its ‘Preliminary Needs Assessment for Repatriation’, which frames the crisis as purely humanitarian in nature.

Contributions to the Joint Response Plan (By donor)
Hover donors below to see annual values
Total contribution (US$)
Reported contributions the Joint Response Plan as of 29 November 2023. Source: UNOCHA Financial Tracking Service.

Donor fatigue has culminated in successive cuts to Rohingya refugees’ food rations in 2023. Until February this year, refugees in Bangladesh had received the equivalent of US$12 per person per month in food support through the UN World Food Programme (WFP). The stipend was cut to US$10 in March and US$8 in June, with the WFP subsequently warning of a further cut to US$6. Even before these decreases, the WFP had estimated in 2022 that 83% of Rohingya-refugee households were ‘highly vulnerable’. Since the cuts, the percentage of households reporting acceptable food consumption has fallen from 56% to 22%, according to the WFP, while the percentage of households needing to borrow money for food doubled. If these cuts continue, the likely result will be an increase in severe acute malnutrition, thus impeding the physical and cognitive development of Rohingya children. The cuts could also exacerbate the deteriorating security environment by driving some refugees towards militant and criminal groups out of increasing desperation to feed their families.

Assistance for the Rohingya has fallen in part because of the other crises competing for international attention and a shift in political focus towards Myanmar as a whole following the coup in 2021. The sense among Western governments that no durable solution for the Rohingya will emerge until the military is removed from power, however, risks relegating the Rohingya issue to an afterthought rather than recognising it as an equally important challenge to be addressed in tandem with the broader conflict inside Myanmar.

Alternative options for the Rohingya

Facing growing insecurity, tightening restrictions, and waning international support and attention, Rohingya are weighing alternatives to remaining in the camps that are equally unattractive.

Since April 2020, Bangladesh has relocated 30,000 Rohingya to the island of Bhasan Char. The government advertised Bhasan Char as an improvement on the conditions in the mainland camps, with more durable, concrete facilities, better access to education and services, and greater freedom to work . But recent surveys released by the WFP and UNHCR found that 94% of Rohingya on Bhasan Char were deprived of education, 77% were unemployed (more than double the figure in Cox’s Bazar in 2019, according to a survey), and 90% were ‘highly’ vulnerable. Moreover, the island remains susceptible to extreme weather and climate change, and despite Bangladesh’s repeated commitment to ensuring that all relocations would be voluntary, human-rights groups have documented a pattern of coercion in moving refugees onto the island.

While the security situation on Bhasan Char is markedly better than in the camps, residents are isolated and often unable to acquire permission to return to the mainland even for short visits. Even after UNHCR signed a memorandum of understanding with the Bangladesh government to operate on the island, Bangladesh’s promises to Rohingya of better services have not been fulfilled and the island’s remoteness has increased the costs of aid and service delivery. The conditions on Bhasan Char have compelled thousands of Rohingya to leave the island through informal channels, with some detained and returned in the process. Although small numbers of Rohingya continue to be relocated to Bhasan Char, most do not see the island as a viable alternative to the camps. Moreover, Bangladesh has come to view further relocations as counterproductive to the launch of a pilot project to repatriate the Rohingya. With the pilot project now stalled, it is possible that Bangladeshi officials will again take a more positive view towards Bhasan Char. But the island remains unappealing to the Rohingya.

The Rohingya have also sought refuge elsewhere in the region. Malaysia, for example, has historically served as a destination for Rohingya and now hosts more than 100,000 registered refugees according to UNHCR. Over the past decade, tens of thousands of Rohingya have sought to reach Malaysia by land or sea, at significant expense and often with tragic consequences. In the mid-2010s, thousands of mostly Rohingya refugees were smuggled and trafficked towards Malaysia only to be held in dangerous jungle camps on the Thailand—Malaysia border or left to drift offshore, prevented from disembarking by the region’s maritime forces.

Although Rohingya boat departures decreased significantly following the 2015 Andaman Sea crisis, voyages by sea to Malaysia have again accelerated, with UNHCR estimating that some 3,500 Rohingya attempted the crossing in 2022, a fivefold increase on the 2021 figure. All risked being pushed back by regional maritime forces and almost 10% of them died at sea. In a change from the past, nearly half the Rohingya taking to boats in 2022 were women and children, probably because many young men had already made the crossing and wanted their families to join them or to start new families. This year, with November marking the onset of the ‘sailing season’ when conditions at sea improve, more than 1,000 Rohingya have already arrived in Aceh, Indonesia, and further departures over the next six months are likely. Meanwhile, Rohingya have also increasingly sought to reach Malaysia by land, going back to Myanmar and heading south towards Thailand despite the high risk of arrest and imprisonment for ‘illegal travel’ since the coup.

Though Bangladesh had blocked Rohingya from being resettled to third countries since 2010, fearing it would stoke resentment domestically and serve as a pull factor for further Rohingya to enter Bangladesh, it ultimately allowed Mohib Ullah’s family to be resettled to Canada following his killing and has since permitted small numbers of other vulnerable Rohingya to be resettled to a handful of countries. The public announcement in December 2022 that the US would open a refugee-resettlement programme for Rohingyas in Bangladesh created heightened expectations, particularly among educated Rohingya and those with family members in Western countries, that third-country resettlement might become a possibility. But information about how many Rohingya will be resettled, who will be prioritised and how long the process will take has not been made available. The US has resettled less than 200 individuals since announcing the programme, in part because of delays in Bangladesh granting exit permits, while other countries have resettled less than 250 Rohingya refugees total.

Of the alternatives to life in the camps, only repatriation to Myanmar offers the potential to address the Rohingya refugee crisis at scale. But Rakhine State — and Myanmar more broadly — has undergone significant changes since the Rohingya fled in 2017. The military coup incited levels of violence unseen in Myanmar since the Second World War, with over 25,000 violent events recorded between February 2021 and November 2023. Many of the Rohingya villages razed by the military in 2017 have been reclaimed by vegetation.

The AA and the regime agreed to an informal ceasefire in November 2020, but the truce temporarily broke down between August and November 2022, a period in which the AA seized a string of regime outposts in Maungdaw Township, giving it control over segments of the Myanmar—Bangladesh border for the first time. Then, on 13 November 2023, fighting returned to Rakhine as the AA moved to synchronise attacks with an offensive by its allies that had begun in Shan State on 27 October. The AA’s renewed campaign has once again focused on seizing control of peripheral areas, including border posts occupied by the Border Guard Police. Myanmar’s border with Bangladesh is now characterised by increasing competition between the AA and the military regime, as well as rising Rohingya militant activity.

Pathways to repatriation? No safe options for the Rohingya

There are, broadly, four potential pathways for Rohingya repatriation, each of which carries significant risks. Firstly, there is the potential for repatriation through the formal bilateral process agreed between Bangladesh and the Myanmar military. Following two failed attempts in 2018 and 2019, and COVID- related delays, talks over a pilot repatriation project have picked up over the past year through a process facilitated by China, through which Beijing has managed to extract some modest concessions from the regime in Naypyidaw. Myanmar military officials have travelled to Bangladesh for ‘come and talk’ visits in the camps, most recently in late October, and allowed a delegation of Rohingya and Bangladeshi officials to travel to Rakhine for a ‘go and see’ visit in May, the first of its kind since 2017.

While publicly committed to the bilateral repatriation process and employing incentives and coercion to convince Rohingya to return, in private Bangladeshi officials remain deeply sceptical about the Myanmar military’s intentions, a sentiment shared widely among the Rohingya refugee population. From the perspective of many Rohingya, the military has shown little inclination to address the root causes of the crisis or provide guarantees of citizenship, safety and the right to return to their original land. There is also an enduring concern that, should they return, Rohingya will be confined to camps similar to those in and around Sittwe, the capital of Rakhine State, where at least 120,000 Rohingya have languished since 2012.

Moreover, the Rohingya are concerned about the conflict between the AA and the military, the latter’s use of divide-and-rule tactics to exploit Rakhine—Rohingya tensions, and the limited international presence in Rakhine, which has been further constrained by junta-imposed restrictions on humanitarian relief after Cyclone Mocha. Refugees also face social pressure from the Rohingya diaspora and within the camps not to participate in the military-led repatriation effort. Nonetheless, should a pilot repatriation project move forward, a small number of Rohingya may agree to repatriate through formal channels because of the deteriorating conditions in the camps and persistent pressure from the Bangladeshi authorities.

Secondly, Rohingya could also seek to return to Rakhine State informally and of their own accord. A small number of Rohingya have returned since 2017, often through arrangements with military- appointed administrators in their home villages, facilitated by the payment of significant bribes.

Informal mechanisms typically require the returnee to accept a so-called National Verification Card, an identity document that effectively labels the holder as an immigrant from Bangladesh. Despite the informal cooperation of some local authorities, numerous Rohingya have been arrested upon returning to Myanmar, including dozens in the past few months. While the number of refugees considering informal channels of return appeared to be growing in mid-2022, the resumption of fighting between the military and the AA has dampened the trend. Even if interest in informal returns rises again, the avenue does not offer a likely pathway to large-scale repatriation.

Thirdly, Rohingya could support efforts by Rohingya armed groups to fight their way back into Rakhine. Among a minority of the refugee population there is a view that the establishment of a Rohingya enclave within northern Rakhine, controlled by an allied armed group, is the only means by which the population can safely and sustainably return to Myanmar. This view is also espoused by some in Bangladesh’s security agencies, frustrated that ARSA and the RSO have focused most of their efforts on controlling the camps rather than fighting the Myanmar military. Should a Rohingya armed actor try to return to Myanmar through force, however, it would invite significant reprisals against the Rohingya population as a whole, fuel perceptions among the broader Myanmar populace that the Rohingya are separatists, and risk squandering some of the empathy gained after the 2021 coup.

Moreover, despite recent rumours of links between the Myanmar military and ARSA, it is unlikely that either the regime or the AA, both of which are much more powerful than any other current or prospective Rohingya armed group, would allow a Rohingya rival to establish a foothold in Rakhine. The AA raided numerous Rohingya villages in September in what it said was an attempt to root out ARSA fighters, and clashed with ARSA in Buthidaung in July; it also fought with the RSO in Maungdaw in October last year. Perversely, the recent crackdown on ARSA in Bangladesh has forced the group back into Myanmar, where its increased presence is already stoking communal tension and fear among Rohingya civilians that they will be branded as ARSA and targeted with violence.

Fourthly, the AA could attempt to initiate a process of repatriating Rohingya to areas of Rakhine under its control. It would not be able to offer citizenship, but it could in theory enable some Rohingya to return to their villages of origin, and potentially provide them with greater security than the military could. Doing so would advance the AA’s objective of enhancing its international legitimacy, by wresting ownership of the Rohingya issue from the military and the National Unity Government of Myanmar, and perhaps open new channels for international assistance. And although the AA may not view the Rohingya as equal in status to the ethnic Rakhine, it has repeatedly claimed it intends to include Rohingya in its parallel governing structures and to welcome the refugee population back to Rakhine State. Moreover, the AA provided Rohingya with humanitarian aid before and after Cyclone Mocha, though Rohingya say they continue to be discriminated against in aid distribution. While an AA-led repatriation effort would face some resistance, the AA’s popularity gives it significant leeway to set policy and there has been little backlash following its progressive pronouncements towards the Rohingya.

Nonetheless, the AA has a mixed record in practice, with large gaps between its stated positions and its ability — and perhaps its willingness — to implement them. For example, Rohingya complain of abuses at the hands of the AA and its administrators, excessive taxation, and a lack of understanding of the ways in which Rohingya are caught between competing armed actors. Though the AA may be more attuned to such divide-and-rule tactics than in the past, potential efforts by the regime to manufacture ethnic divisions could leave Rohingya particularly vulnerable, especially since conflict has once again resumed between the military and the AA. Moreover, it remains to be seen whether the AA can conquer additional parts of Rakhine, and so its ability to deliver on its intentions, even if they are genuine, is uncertain. Given its trajectory of growing influence across northern Rakhine, however, the AA has sought to position itself as a critical player in any future repatriation process. If the AA can demonstrate an ability to bring the Rakhine and Rohingya together and to govern all residents of Rakhine equally, it stands a better chance of advancing its objectives and creating an environment more appealing to potential Rohingya returnees.

Outlook: a long and difficult road ahead

The next few years for Rohingya refugees in Bangladesh are likely to be grim. Even if conditions on Bhasan Char improve, which is unlikely, the accommodation there was built to serve only one-tenth of the Rohingya population currently in the camps. Though opportunities for resettlement will increase from the current trickle, they are unlikely to match annual Rohingya population growth in Cox’s Bazar. Journeys by boat and land to reach Malaysia will increase, but they remain expensive and dangerous. Without significant shifts in Rakhine, a return home risks persecution at the hands of the Myanmar military, the AA or both. And without a reversal of policy restrictions and support for Rohingya militant groups by Bangladesh, the best-case scenario for the refugee camps is a status quo that offers decreasing hope for the future amid rising insecurity and a dearth of opportunities for education, livelihoods or a flourishing Rohingya civil society.

While Bangladesh is not responsible for the Rohingya crisis, and Dhaka is confronted with an array of unenviable choices, the current situation is partly of its own making. Its response to the influx of Rohingya in 2017 was an understandable attempt to promote repatriation by putting pressure on Myanmar and on the refugees themselves, while also preventing a further inflow in the short term. Since then, however, it has stubbornly adhered to an approach that imposes severe restrictions on the Rohingya, fosters insecurity and deepens the humanitarian crisis even as international assistance is declining. In concert, these dynamics serve to enable and empower illegitimate armed and criminal actors who ultimately undermine prospects for repatriation in the longer term.

As well as complicating its goal of repatriating Rohingya to Rakhine, Bangladesh’s current approach entails significant risks to its own interests, and potentially to its relationship with Myanmar. For example, it is likely to become increasingly difficult for Bangladesh to contain the Rohingya militant and criminal groups it has fostered within the camps. And though there is little to suggest Rohingya are prone to extremism, despite numerous precipitating factors, extremist groups — both domestic and foreign — could seek to radicalise them. Moreover, while the risk of communal violence between the Rohingya and Bangladeshis in Cox’s Bazar has often been overstated, the potential for a breakdown in social cohesion will increase over time.

Furthermore, Bangladesh could be inviting reprisals from the Myanmar military and the AA, which Dhaka already believes is meddling in the sensitive Chittagong Hill Tracts, if it is seen as supporting Rohingya militant groups that carry out attacks or seek territorial control in Rakhine. There are also financial risks to maintaining the current policy approach: Bangladesh claims (without evidence) that it spends US$1.2 billion per year on the Rohingya. But if, alongside declining donor assistance, Bangladesh does not maintain its support, Rohingya will probably turn to criminality and militancy within the camps to meet their basic needs. Or they may increasingly look for work outside the camps, which could cause the collapse of Dhaka’s policy of containment.

Overall, the forced marriage of Bangladesh and the Rohingya is likely to be long and unhappy without significant changes in Dhaka’s approach to the crisis.

Website and graphics by Brody Smith.