Mozambique’s Instability

Mozambique faces an array of security challenges, from resilient illicit economies to an enduring kidnapping problem and a long running insurgency in the north. Many security challenges play out at community level, yet local perceptions and experiences of insecurity are underrepresented, with existing evidence based on top-down research privileging established donor and governmental narratives.

Drawing on research across Angoche, Maputo, Montepuez, Nacala and Pemba, this report gives voice to bottom-up perspectives. It does so by analysing local-level inputs and lived experience to provide a more holistic appraisal of the security landscape and support more effective interventions.

The findings depict a multilayered system of local-level insecurity, with lived experience of key harms varying across time and place. Across the board, interviewees expressed an overriding concern with petty crime, street-level insecurity, state abuses and corruption – with terrorism perceived, in all sites but Pemba, as a more peripheral issue. This speaks to a potential discordance with the dominant approach of national and international stakeholders – one perceived by many interviewees as reductionist in its primary focus on countering Islamic fundamentalism.

Many of the threats playing out at a local level were viewed as interconnected. Notably, endemic corruption was positioned as the foremost factor shaping local political economies, enabling many of the challenges cited, from kidnap-for-ransom to burglary and transnational trafficking flows. Beyond system-wide harms, interviewees described individual risks where entrenched perverse incentives were held to divert officials from public protection mandates. Echoing past studies, the professed result was deep-seated antipathy to state authority and a breakdown of trust between communities and the state.

Dissatisfaction with formal state provision was held to have triggered a search for alternative security structures. Interviewees described experiences of both private security provision and community self-policing initiatives – with a range of legitimacy and access issues raised. At times, these issues were described as themselves fuelling instability and violence, with further negative impacts on citizen security across the board.

At a deeper level, many forms of insecurity experienced at the local level were perceived as expressions of deep-seated, long-term trends. While appearing at times to represent discrete phenomena, these were understood to be shaped by the same enabling environment, as part of a continuum of violence and contestation based on grievances stretching back to Mozambique’s colonial experience. Here, elite capture of the state was presented as a key cross-cutting context of insecurity – with booming international investment, counterposed with entrenched political and economic marginalisation, held to aggravate latent tensions that could be exploited by criminal and insurgent actors.

Over the decades, national and international interventions were perceived as having ignored – and at times aggravated – these conditions. International stakeholders were frequently accused of overlooking corrupt behaviour and supporting a flawed security infrastructure, with harmful effects at the local level. Here, a clear sense of community-level exclusion from the design of security interventions was expressed, with a lack of understanding of local dynamics purportedly limiting the effectiveness of external programming and resulting in a range of harmful consequences.

Based on this research, the report presents the following recommendations for those engaged in national and international programming, mindful of the juxtaposition between existing approaches and the experiences of communities on the ground:

• External programming must be designed, adapted and validated based on meaningful local-level consultation that internalises the interests and lived experience of community actors across Mozambique.

• Efforts to professionalise law enforcement agencies must have at their core a focus on improving relations with the populations they are mandated to serve.

• Focused support should be provided to Mozambican civil society groups and NGOs performing potentially crucial roles in monitoring crime prevention and strengthening resilience.

• Technical assistance programming across Mozambique must take politics seriously, with interventions tailored to the political economies of key sectors and institutions at the local and national levels.

• International programming must better account for the role of some state actors in the production and maintenance of insecurity in Mozambique, and as potential spoilers in reform efforts.

• A focus on the drivers of crime and violence should be prioritised, with programming moving beyond symptoms to address underlying structural dynamics.

• Opportunities to address the role of local government and support decentralised structures capable of allowing meaningful popular participation should be explored.

• Regulation and oversight of private security companies should be strengthened to mitigate risks of human rights abuses.

Introduction

In February 2025, more than 9,000 people in Cabo Delgado, the northernmost province of Mozambique, were displaced due to ongoing violence and insecurity, including attacks by non-state actors in the Macomia and Meluco districts. The situation is one of many acute challenges, including widespread criminality, a longstanding kidnapping problem, and sustained protests following the disputed 2024 general elections, that are indicative of a precarious security situation across the country. With key threats linked to unresolved public grievances and marginalisation, there are growing calls for the government to engage with local communities to establish trust and rapport and to address popular discontent with existing governance and security arrangements. A solid, bottom-up appraisal of the varying local perceptions and experiences of security in Mozambique is required amid these ongoing challenges.

Although the benefits of “bottom-up” analysis are widely recognised, there is often little in the way of empirical data available, leading to knowledge gaps and an evidence base overly reliant on elite assumptions. Drawing on fieldwork across five locations in Mozambique, this report aims to build an understanding of how security and insecurity, including organised crime and terrorism, are perceived and experienced at the local level.

This primarily involves unpacking forms of insecurity as determined by local stakeholders and examining their mechanics, resonance and impact on communities, and how, in some cases, they may be seen to diverge from government and donor priorities. Such insights in turn offer valuable entry points for interrogating the underlying conditions and broader mechanisms that feed into and facilitate violence. Understanding these conditions is crucial: while discrete forms of insecurity have their own (thematically and spatially defined) drivers and dynamics, they often coexist within a shared social ecology, the complexity of which needs to be factored into programme design and policymaking.

Understanding how these conditions diminish trust in formal governance, diffuse and normalise violence, and propel social fragmentation is essential for contextualising local experiences of insecurity, identifying plausible pathways of change and developing subsequent intervention strategies. Accordingly, this report examines the following research questions:

• How are crime, terrorism and other forms of insecurity perceived at the local level in Mozambique?

• What are the mechanisms conditioning local insecurity?

• What are the implications for interventions by national stakeholders and international donors?

In examining these questions, the report offers recommendations to governmental stakeholders and donors on how they should adapt their approaches to better address insecurity at the subnational level.

This Whitehall Report is part of a broader three-year project, “Organised Crime, Terror and Insecurity in Africa” (OCTA), exploring local perceptions of these issues across three case study countries: Kenya, Mozambique and Nigeria. These states all face significant problems linked to crime, terrorism and related security issues; share common challenges in responding to these threats (including limited resources, capacity gaps and corruption); and have been significant recipients of long-term donor assistance and advice on these issues. Across these countries, the project’s contribution is to complement conventional, top-down understandings of insecurity (which generally privilege government and donor perspectives) with “bottom-up” efforts that harness local knowledge, providing a holistic appraisal of the security landscape and contributing to more effective interventions.

This focus is in line with the UN Policy Toolkit on The Hague Good Practices on the Nexus Between Transnational Organized Crime and Terrorism, which acknowledges the need to develop “context-specific responses based on evidence-based research” that draws on local engagement and “inputs from all sectors of society”. Yet concrete initiatives in this area have been few and key knowledge gaps persist, with limited bottom-up engagement to build the evidence base. The project contributes to addressing this gap via in-depth research on insecurity at the local level, considering the findings alongside existing government and donor approaches. It aims to provide guidance to inform policy that is more firmly grounded in analysis of local-level security dynamics.

Methodology, Challenges and Limitations

The report is based on research conducted between February 2022 and February 2024, which included an in-depth review of open source literature, extensive country-based fieldwork and the use of political economy analysis tools to examine key aspects of the security landscape in Mozambique.

The authors first undertook a rigorous review of open source literature on the security situation in Mozambique to establish a firm contextual basis for both fieldwork and policy analysis. The review covered relevant academic literature, NGO and government reports, policy briefs and media sources. The authors prioritised research on crime, terrorism and other key security threats. A key focus was placed on analysing civil society, government and donors’ understandings of and approaches to these threats. This was supplemented with a review of quantitative data relating to Mozambique, including Worldwide Governance Indicators – Political Stability and the Absence of Violence/Terrorism; Transparency International’s Corruption Perceptions Index; Varieties of Democracy (V-Dem) project data; and Afrobarometer surveys on institutions, leadership and governance, among other sources.

Researchers from RUSI then worked with a Mozambican research institution, Centro para Democracia e Direitos Humanos (Centre for Democracy and Human Rights – CDD), to conduct extensive country-based fieldwork, gathering data on local perceptions of insecurity in five urban sites across Mozambique: Angoche, Maputo, Montepuez, Nacala and Pemba. These sites were identified as both offering the best potential to explore the impact of insecurity, and being safe areas in which to conduct research. For example, Pemba and Montepuez, as the provincial capital and second-largest city in the northeastern province of Cabo Delgado, have been impacted by a six-year insurgency. The port cities of Nacala and Angoche in the northern province of Nampula, as well as the capital Maputo in the south, are known hubs for illicit trade and organised criminality, as well as experiencing other pressing security issues.

As part of the country-based research, 98 individuals were interviewed between September and October 2023 (15 in Angoche, 24 in Maputo, 20 in Montepuez, 19 in Nacala and 20 in Pemba). An additional focus group of nine local officials was also held in Angoche. An additional focus group of nine participants was also held in Angoche, where less time was available for interviews as a result of security issues affecting the area.

Interviewees were selected by CDD, using its existing networks and knowledge of the research study sites, from across different communities at the site level and/or based on its knowledge of local patterns of insecurity. In light of the sensitivity of the topic and the potential risks involved, it was agreed with all interviewees that their identities would remain confidential. Interviewees included representatives of NGOs, civil society activists, representatives of community groups, traditional and religious leaders, local officials, law enforcement stakeholders, lawyers, journalists, businesspeople, teaching professionals (schoolteachers and university lecturers) and other key stakeholders.

RUSI researchers travelled to Maputo in November 2023 to conduct validation workshops with CDD analysts, as well as a further 10 interviews focusing on wider trends in insecurity across the country, with experts on politics and security, and representatives specifically from the business and banking sectors and donor communities.

▲ Table 1: Interviews and Focus Groups Conducted

Researchers asked a series of open-ended, broadly framed questions to build an accurate picture of local perceptions. Participants were asked to describe the main forms of insecurity within their locality, the type of actors they identified as responsible for insecurity, those they considered to be the main providers of security, and the characteristics of these individuals and groups. During these discussions, interviewers explored the importance that participants attributed to threats such as “crime” and “terrorism”, but particular care was taken to ensure that initial questions did not lead with these terms.

In conducting research across five sites, the project sought to prioritise depth over breadth. Inevitably, the responses will not cover every form and manifestation of local insecurity across a country as large and diverse as Mozambique. However, in combining interview-based research in sites experiencing significant security challenges with in-depth political economy analysis, the project identifies trends and patterns that may also be relevant across other parts of the country.

Structure

The report has two chapters and a conclusion. Chapter I disaggregates perceptions and experiences of insecurity as viewed by participants, examining their perceived mechanics, resonance and impact. Chapter II explores a range of underlying factors that help to situate these perceptions of insecurity, assessing how they fit within a wider system of violence and instability, and considering how far existing national and international interventions are viewed as addressing these issues. The conclusion highlights the implications for government and donor policy and programming, and provides recommendations to improve the collective response to localised experiences of insecurity.

I. Local Perceptions and Experiences of Insecurity

In all issues related to conflict or insecurity, communities always lose.

- Interview with civil-society activist, Pemba

Mozambique faces a range of acute security threats, from high rates of violent crime to an enduring kidnap-for-ransom problem and a long-running insurgency in the country’s northernmost coastal province. Across specific contexts subnationally, perceptions and understandings of these forms of insecurity are contested, at times feeding on a mix of context-specific historical tensions and experiences of inequality.

Against this backdrop, it is crucial to interrogate local appraisals of Mozambique’s threat landscape. Accordingly, this chapter examines local-level experiences and perceptions of insecurity in Angoche, Maputo, Montepuez, Nacala and Pemba, drawing on interview and focus group data, and supplemented by analysis of other published information. It analyses the mechanics and perceived impact of key threats, considering their juxtaposition with established donor and governmental narratives.

Street-Level Crime and Personal Insecurity

Across all five sites, among the most common issues raised by interviewees were sustained high levels of crime targeting property and the person. Many recounted direct experiences of victimisation in the form of “petty crime” or street-level insecurity, with muggings, burglary, robbery, pickpocketing and phone-snatching portrayed as routine occurrences. Such patterns of criminal activity are well established and mirrored in past survey-based data. For example, in an Afrobarometer survey of a nationally representative sample of 1,200 adults in October–November 2022, 59% of respondents reported feeling unsafe while walking in their neighbourhood at least once in the previous year, with 39% reporting feeling unsafe “several times”, “many times” or “always”. These survey results represented an increase on those a decade earlier, with the proportion of respondents reporting feeling unsafe in their communities “several times” increasing by 12 percentage points since 2012.

Although pronounced in high-density urban areas, criminal activity of this kind was reported across all five sites. Hotspots were described as relatively transient – shifting as a function of a range of factors, including law enforcement activity. The distribution of target populations was also reported as relevant in many cases. Here, those of greatest means were at times perceived to be more frequently targeted. Reinforcing this, a number of (albeit dated) studies on the socio-economic determinants of individual and household level victimisation in Mozambique have found the probability of victimisation in property crimes and various kinds of physical assault to increase commensurate with income.

Based on countrywide representative household surveys, the significance of this effect was found to vary by crime type, while across the board, poorer households – although less likely to be targeted – were found to suffer relatively greater losses when such incidents occur. Some interviewees commented on the deterrent effect of additional protective security measures. As one respondent in Nacala observed, where “wealthy individuals can afford private security, … middle-class families become the primary victims”. Meanwhile, other research points to the vulnerability to violent crime of inhabitants of low-income neighbourhoods, often linked to a lack of state presence in such peripheral locations.

While low levels of reporting, a lack of standardisation in collection systems and inconsistent record-keeping affect the reliability of official crime data, petty criminality was depicted as pervasive across interview testimony. In many cases, severe disruption to residents’ day-to-day lives was reported. Those interviewed in Maputo, for example, spoke of high levels of vulnerability spanning a wide array of crimes against property and the person. In the words of one civil society activist, “In the city, you can easily become a victim of carjacking, robbery or assault. Incidents like having … a mobile phone stolen are common in downtown Maputo”.

Other interviewees described the cumulative effect of these patterns of criminality. In the words of one respondent, “It’s difficult to refer to ‘security’ in a city where individuals are abducted nearly every week; where you can’t walk safely due to the constant threat of robbery or even murder; where your mobile can be snatched at any given moment; and where your car can be vandalised for parts or outright stolen”. Street-level theft of mobile phones, car parts and motorcycles was also reported as pervasive by respondents in Angoche, Montepuez, Pemba and Nacala, with further concern expressed over high rates of home invasions, and burglaries targeting residences and workplaces in these locations.

The police response to these crimes was consistently described as inadequate. Despite a recent increase in police presence and visibility in Maputo bairros (neighbourhoods), including in those bairros which are experiencing high levels of crime, respondents frequently described themselves as underserved by the police and other security agencies, with some bemoaning a lack of police patrols, stations and other security infrastructure in key locations. Others complained that police are slow to respond to calls and, when they do, are slow or unable to act. They blamed this on resource constraints (particularly a lack of vehicles or fuel), as well as institutionalised corruption, as explored further below.

The violent nature of some street-level criminality stands out in interview testimony. In Montepuez, respondents spoke of machete-wielding men targeting both wealthy and poorer individuals (such as kiosk owners), while accounts of the use of cords across roads to bring down motorcyclists and assaults conducted using hammers speak similarly to the potential level of physical threat to victims. As noted by one respondent in Maputo: “Nowadays, people are killed in broad daylight – just for holding a simple mobile phone, one might lose one’s life”. Yet interviewees rarely oriented their accounts explicitly around the individual physical (or emotional) harm inflicted. Instead, accounts generally focused on the transactional nature of these incidents and the material losses involved.

In many cases, street-level crime was described as unsophisticated, disorganised and opportunistic. As noted by one interviewee in Maputo: “Increasing poverty in major cities, particularly in Maputo, has resulted in more opportunistic crimes … such as cell phone theft, which can escalate to deadly outcomes”.

Others cited high demand for car parts as driving opportunistic instances of carjacking and theft amidst an unregulated market for vehicles. In the words of one respondent in Nacala, rather than “structured or organised groups with leadership”, those responsible are often “poor youths … spontaneously coming together to commit crime”. Others spoke of unsophisticated crime, but with a degree of organisation; in Montepuez, interviewees referenced “organised groups of young people” committing robbery and theft with impunity.

Differences in the perceived prevalence of such incidents emerged across the different sites. Notably, Angoche was the only site where any (albeit few) respondents reported feeling safe on a day-to-day basis. Some interviewees attributed this to relatively tight community links and the city’s relative geographical isolation, which limits inbound and outbound population flows. Despite this, other Angoche-based respondents reported that crime was widespread, revealing a marked divergence in perceptions in this location.

In other sites there was consensus on the pervasive nature of street-level crime. Other distinguishing features were apparent, however, that marked out particular locations. For example, a distinctive characteristic in sites in Cabo Delgado was the way in which respondents wove accounts of petty crime into broader discussions of insecurity, linking these to experiences of terrorism, transnational crime, corruption and state predation. These links are examined further in Chapter II.

In Maputo, depictions of street-level criminality often focused on the dynamics of kidnap for ransom as perceived by interviewees. Here, far from the politically motivated abductions witnessed during Mozambique’s 1977–92 civil war, contemporary kidnapping incidents were identified as criminally motivated, affecting ground-level security in different ways. It is assessed that these incidents predominantly affect businesspeople of Asian descent and their families, as well as other nationals with a presence in the country.

Interviewees reflected regularly on the nature of the law enforcement response. While under-reporting makes it difficult to establish accurate kidnapping rates, in March 2024, Minister of Interior Pascoal Ronda announced that the Police of the Republic of Mozambique (PRM) had recorded 185 cases of kidnapping since 2011, with more than 288 people detained on suspicion of involvement. In general, levels of kidnapping are regarded as high by southern African (and wider continental) standards.

The changing nature of this phenomenon was articulated in interview testimony and has been laid out in the published literature. Previously linked to criminal networks within relevant Asian communities, kidnap for ransom is perceived as lucrative and low risk and has increasingly raised concern. Growing sophistication of the criminal networks has also been observed. As noted by one interviewee: “Over the past 10 years, we have witnessed the strengthening of groups involved in kidnappings”. These groups were described by others as “notable crime syndicates”, holders of “economic power” and “behind not only kidnappings but also other threats from fraud to robberies”.

These views are in line with other sources. As noted by corporate intelligence consultancy S-RM, kidnappings should be viewed as more than simply “a consequence of the petty criminality associated with Mozambique’s struggling economy and increasing poverty levels”. Instead, the phenomenon is held to “reflect … broader concerns with the country’s political leadership and weak security apparatus, which has produced an environment conducive to the proliferation of organised crime”. Frequently marked by violence and tight coordination, many kidnapping cases are reported to be well-planned operations, with targets’ activities reconnoîtred in advance and, in some cases, prospective victims extorted to avoid being victimised.

Interviewees spoke of collusion among state officials as key to these operations, describing some police officers as “paid operatives” of kidnapping syndicates – complicit in facilitating criminally motivated kidnapping events and protecting perpetrators. At times, respondents held that compromised police officers sabotaged cases to avoid implicating senior officials. Such views come in the wake of past cases involving officials across the state apparatus (albeit with prosecutions being a rare occurrence). Past research by the Global Initiative Against Transnational Organized Crime (GI-TOC), among other institutions, similarly points to wide-ranging state involvement, from the supply of weapons to direct participation in abductions. Key governmental actors have recognised the extent of collusion in a range of instances: as warned by former Attorney-General Augusto Paulino, kidnappings “will not cease while there is no purification of our own ranks, while we do not eliminate those who have been infiltrated into the judicial system”.

While the kidnapping threat has terrorised Mozambique’s business community, its impact on other constituencies was also highlighted. Beyond local businesspeople, other interviewees expressed concern over criminally motivated kidnappings, including of teachers, journalists and civil society activists, among other stakeholders. As observed by one respondent: “Kidnapping may seem to affect a limited community, but its effects are felt throughout society”. This sentiment is mirrored in long-running public protests, reflecting sustained societal outrage at the government’s failure to respond. At times, the situation has made international headlines, as with the abduction and murder of a 13-year-old boy in 2013, triggering marches by thousands of protestors in Maputo.

Public anger appears to centre on impunity as a common thread across successive kidnappings and associated retributive events. Interviewees lamented the contribution made to a wider climate of fear and insecurity, to the normalisation of a lack of respect for the rule of law, and to the hollowing out of criminal justice institutions. In the words of a Maputo-based respondent: “When you have police officers participating in kidnappings for personal gain, it epitomises the dark side of corruption”. Resentment appears to be heightened where high-level actors are found to be complicit, as seen in the sentencing in 2013 of a member of the presidential guard, alongside five others. That year, former first lady Graç a Machel described all of Mozambican society as “held hostage to the kidnappers”, a sentiment articulated by interviewees over 10 years later.

State Predation and Corruption

Reports of complicity among officials were provided not only in relation to kidnapping. In all sites, interviewees perceived abuse of power and corruption within state security institutions as a core feature of the security landscape. Almost all described corruption as endemic, with the influence of corrupt relationships articulated as the defining feature of the country’s governance and the foremost factor shaping local political economies. Such attitudes are entrenched in a long history: accusations of corruption and other serious violations have plagued Mozambique’s domestic security institutions for decades, from the country’s police to its wildlife services and military.

Interviewees alleged a wide range of predatory behaviours among state actors. Reference was made most frequently to extortion and bribery on the part of the police, against which antipathy was often levelled. This institutional focus is understandable: as the closest government authority to local communities, and given its mandate to maintain public safety, the police are often the first point of contact for citizens experiencing security threats and harms.

A range of respondents expressed the view that some police officers patrol or investigate crime only in exchange for bribes. Others provided accounts of extortion: in a typical account, one social activist described instances where officials “stop citizens during the night to intimidate them”, noting, “I have personally experienced situations [such as one] where 10 intoxicated individuals with only … [their] identification cards stopped me [for no legitimate reason, to extort money]”. Other accounts were provided of officers extorting money at road blocks and police stations, reportedly demanding pay-offs from citizens accused of minor illegal activity (such as gambling), or “shaking down” individuals directly for money or possessions.

The accounts provided are consistent with perceptions revealed in recent survey data. An Afrobarometer survey of 1,200 adults in October–November 2022 saw 51% of those requesting assistance from the police in the previous year reporting having to pay a bribe, with 47% reporting doing so when encountering police in other situations (at checkpoints, during ID checks and in the course of investigations). Over half (55%) of those surveyed described “most” or “all” police officers as corrupt – a higher proportion than for any of the other 10 institutions covered by the survey. In terms of behaviours, police were perceived to “often” or “always” stop drivers without a sound reason by 45% of respondents, to engage in criminal activity by 39%, and to use excessive force against suspected criminals and protestors by 37% and 34%, respectively.

While not disaggregated by institution, this comes against the backdrop of Mozambique’s stagnating position on Transparency International’s Corruption Perceptions Index. In 2023, the country was ranked 145 out of 180 countries surveyed, with a score of 25/100, a progressive decline from its ranking of 123 (of 176 countries surveyed) in 2012, with a score of 31/100.

Some interviewees reflected on the incentive structures at play. Low pay and morale were frequently assessed to be enhancing police officers’ susceptibility to bribery – in the words of one respondent, “what drives them [police officers] is the profit they’ll make at the end of the day, not the safety of people and their property”. Here, interviewees pointed to discrepancies between low salaries and the cars and property reportedly owned by some police and customs officers, with well-off “businessmen” held routinely to “offer luxury vehicles to police chiefs in exchange for protection”. It was often held that these “businessmen” had come to the local area from abroad. As summarised by one respondent: “Foreigners … dominate commerce in Nacala. With the money they have, they can secure police protection. It’s frustrating – we are in our own land but don’t have protection. It feels like we are the enemies of the police.”

Interviewees pointed to further perceived links between police and other criminally motivated individuals. At the far end of the spectrum, in Nacala accounts were provided of links to serious crime, with respondents in Maputo also speaking of an active perceived role by police units in kidnapping and burglary. The impact on communities were described as encompassing an erosion of the rule of law, while a damaging effect in terms of deterrence was also reported. As expressed by one interviewee: “Where police are bought to work in favour of the criminal, this neutralises the preventive and combative functions of the state. Corruption is like a poison; once it stings, the organs die. Thus, the state’s organs die, allowing organised crime to prosper.”

Beyond system-wide harms attributed to police corruption, individual risks were also described. A number of interviewees articulated the view that reporting a crime came with a risk that information would be passed to the criminals responsible, who would then engage in targeted intimidation. Others complained that citizens who refused to pay off police risked being targeted by both criminals and the police themselves. As expressed by one respondent in Maputo, “If something happens to me and I turn to the police, I’m uncertain whether I’m in safe hands”.

Corruption risks across the wider criminal justice system were also reported. Interviewees lamented an entrenched culture of impunity, readily exploited by criminal actors. Corruption risks were identified from the early stages of the criminal justice chain, with public officials perceived to engage in a range of behaviours for bribes, including releasing suspects, obstructing investigations and destroying evidence or otherwise ensuring its inadmissibility. Similar concerns were expressed further along the criminal justice chain, with compromised officers across prosecution and judicial services perceived to jeopardise cases in many different ways, from falsification of evidence to intimidation of witnesses and delaying of due process. These views echo previous survey data, with 31% of those surveyed by Afrobarometer in October–November 2022 describing “most” or “all” judges and magistrates as corrupt, and a further 42% judging “some” of these professionals to be corrupt.

In some cases, interviewees spoke of the dual nature of citizen exploitation enabled by a compromised criminal justice system, describing corrupt officials as leveraging both sides in a criminal case. For one respondent, “if you commit a crime and seek resolution, you have to bribe [while] … simultaneously, the victim is charged to expedite their case”. The perversion of the system that this entails was outlined, with interviewees describing patterns of prosecution determined by the ability to make corrupt payments, rather than a disinterested application of the law. In such cases, one interviewee perceived, “the person who is in the right might not see their rights upheld”, with officers focusing “their attention on scrutinising honest individuals while facilitating criminals in return for bribes”.

While bribery and corruption were the primary forms of misconduct alleged, others spoke of misuse of force, sometimes with deadly outcomes. At times, these behaviours were depicted as intertwined, with requests for bribes linked to the threat and use of force. In other cases, interviewees pointed to police violence at public demonstrations. As expressed by one respondent in Pemba, “the PRM violate citizens’ human rights through the use of rubber bullets, mock executions, tear gas, and unwarranted arrests when they take to the streets to protest or express discontent with the regime”.

Such incidents have been condemned internationally. In the case of a lethal police response to protests following local elections in October 2023, Amnesty International equated the “use of excessive force by the PRM, which resulted in deaths, serious injuries and the arbitrary detention of protesters and bystanders” to “clear violations of the country’s Constitution and international human rights obligations”. Most recently, protests over the disputed 2024 general election triggered unrest, a violent security force response (condemned by Human Rights Watch), the violent intimidation of opposition party members, the detention of journalists and a reported 250+ fatalities – with the government resorting to banning protests, condemning these protests as “acts of subversion” and “terrorism”.

In interview sites in Cabo Delgado, respondents reported human rights violations in the context of a militarised counterinsurgency response. Here, a range of predatory behaviours were alleged, with “officers of the FDS [Mozambican Defense and Security Forces], municipal police, local or community force” referenced simultaneously as “providers of security” and “agents of insecurity”. As described by one respondent in Montepuez, “the community fears the military during patrols – they are known for their impatience and intolerance”. At times, human rights abuses and a lack of trust in authorities were perceived as drivers of the conflict itself. This is documented across the published literature, with the pursuit of a “military-first” strategy to tackling violent extremism identified as alienating communities and aggravating underlying grievances in a range of cases.

Local discontent over both state and private sector activity here predates the insurgency. For example, allegations of human rights violations and excessive use of force against artisanal miners were levelled at security forces around the Montepuez Ruby Mining concession from 2012, with part-owner Gemfields paying compensation of $7.6 million to 273 artisanal miners and villagers in 2019 (albeit without admitting liability). In 2021, Amnesty International accused not only insurgents but also government forces and private security company Dyck Advisory Group of war crimes, noting that “the security forces of Mozambique have abused the very people they are meant to protect”.

Mindful of these issues, various frameworks have been enacted to guide security relationships between extractive companies, state and non-state security actors, and communities. The government has indicated its intention, for example, to draw on the Voluntary Principles (VPs) on Security and Human Rights as a multistakeholder membership-based initiative, with a number of companies using these principles in practice. As recognised by researchers Christopher Vandome and Alex Vines, however, adherence to the VPs is not sufficient and needs to be “matched by a wider national, government-led effort to improve the human rights environment”.

Echoing past studies, interview testimony indicates that this situation has translated into a deep-seated distrust of state authority. In the words of one respondent in Pemba, “there is a breakdown of trust between communities and those who should ensure their security – the state itself”. In day-to-day encounters, this was positioned as challenging the very notion of policing by consent. As expressed by a Nacala-based respondent: “The state has lost legitimacy in the eyes of communities … our police officers are not well trained … they address citizens without civility, professionalism, or ethics – how can communities respect and acknowledge their legitimacy?”

It should be noted that perceptions of corruption and state abuses across the five sites were not homogenous nor consistently held – nor could they be independently verified. While the critiques of domestic security provision mentioned above were articulated by a significant majority of interviewees, some provided more positive evaluations of state security activity. A number of individuals in Angoche and Montepuez credited the police with detaining thieves, active patrolling and maintaining constructive community relationships, while also highlighting extortion and bribery as entrenched problems. Positive accounts, however, represent a minority of views expressed, with the overall picture emerging from interviewee testimony being one of dissatisfaction and mistrust of a state security apparatus viewed as preying on the people it is mandated to serve.

Beyond State Responses: Private Security and Vigilantism

Amidst this perceived public sector ineffectiveness, interviewees touched on a variety of non-state responses. Most prominently, interviewees pointed to growing use of private security services, with rising demand for effective security solutions among Mozambican households and businesses presenting lucrative opportunities for private companies. In the country’s north, expanding natural resource exploitation has also seen extractive and other companies employ growing numbers of private security actors, targeting gaps in public provision amid an unstable threat landscape. Rising use of private security, however, raised various issues for interviewees, often depicted as a symptom of deep-seated inequalities and a lack of economic opportunity for average Mozambicans. Many denounced private security as the preserve of the wealthy, with resentment expressed and moral quandaries raised regarding those who are unable to afford these services and are left to live with insecurity.

A number of interviewees pointed to cracks across this fractured security architecture, citing perceived deficiencies in both public and private services. They noted that “public security agents may violate human rights, while private forces are more concerned with profit”. Intersections between public and private security were also raised as an issue in some cases. In the words of one respondent in Pemba: “Many companies operating in the city, due to the evident weaknesses of our public … security forces, end up hiring private security firms. However, a concerning factor is that these private firms are often owned by the higher-ups in the police or military, … contributing to the lack of improvement in public security conditions.”

Concern was expressed over alleged predatory behaviour on the part of private security actors. This reflects wider reported anxiety among civil society over the capacity of the Mozambican government to regulate private security companies. As noted by Adriano Nuvunga of CDD, “These companies have capabilities that often exceed those of the Mozambican state … We are concerned about the limited capacity … to monitor and inspect effectively the private security companies. Due to weak monitoring capacity, human rights violations by private security companies are common.”

This points to fundamental questions around democratic oversight. As noted by analyst Anna Marie Burdzy, “the provision of [private] security services, inherently, includes a possible need to use force … However, using force bears the risk of abuses and misconduct.” Burdzy also notes that “certain tasks which are [traditionally] restricted to law enforcement [are ones] … for which private security [services are] generally not adequately trained. As a result, PSCs [private security companies] may handle such situations in ways that escalate rather than de-escalate the situation.” Echoing such concerns, interviewees perceived the governance issues surrounding such a fundamental shift in security provision to have received insufficient attention. As argued by one respondent in Pemba: “Security is too sensitive an issue to be entrusted to entities outside the state” and the government “should not privatise or outsource security”.

Interviewees also pointed to other non-state responses. With a significant proportion of the population unable to access private security, these were depicted, at their core, as efforts by community members to take security into their own hands. Here, interviewees described cases of community involvement in lynching, vigilantism and mob violence against perpetrators of criminal activity. Such accounts were particularly prevalent in Nacala, although details on what activities lynching comprised were not offered. One respondent in Maputo gave some indication, with residents of one locality reportedly warning: “Do not rob [neighbourhood] T3 or they will burn you with a tyre”.

Such incidents have long been reported by other sources. Back in 2008, the US State Department warned that lynchings and vigilantism were tarnishing Mozambique’s human rights record, citing 26 instances of vigilante killings in Maputo, Matola and Beira. Specific manifestations of local-level vigilante justice have also been reported in the context of the insurgency. In December 2018, for example, an “improvised village militia” comprising residents of Nangade district was reported in the media to have launched a counterattack on insurgents to protect their community. As reported at the time: “Armed with knives, axes, hoes and similar weapons, the group captured two men believed to be insurgents … [but] instead of handing them over to the police, they decided to execute them on the spot. They were beheaded, in retaliation against the decapitations carried out by the Islamists”. In April 2023, local militias (força local) were granted legal authority by the government to help combat the insurgency, despite international concern over the move.

Across the country more broadly, while a deeply rooted community-level justice system exists, a resort to lynchings in particular was attributed by interviewees to a loss of ground-level confidence in official structures. This was articulated by some interviewees in emotional terms – as an expression of mounting public anger at perceived public service failings. As noted by a Nacala-based respondent: “When the community decides to take to the streets, everyone from children to the elderly comes out to ensure security, and they achieve results. When they find a thief, they don’t take them to the police because they believe the thief will be protected … [they] lynch the thief [instead].”

At other times, community responses were described as less impulsive and better organised, amounting to structured systems of local-level protection. They were described as such in the context of a wider tradition of legal pluralism as recognised in the country’s constitution, with many citizens seeking redress through customary and community-based institutions and mechanisms of justice and policing. As articulated by one interviewee in Nacala: “In some neighbourhoods, we have community policing groups: local volunteers who patrol during the night”. These efforts were described as effective in some cases, but as coming at a financial cost. The respondent continued: “In areas [patrolled by community policing groups] … there has been a reduction in home invasions and public attacks at night. However, each family must contribute 500 meticais per month to pay those who patrol the neighbourhoods.”

In some cases, local-level responses were described as escalating beyond compensatory security provision to encompass retaliatory violence against police officers and infrastructure. This was pronounced in interview testimony in Nacala. In the words of one resident: “The relationship between the police and the communities is not good – in neighbourhoods like Quissimajulo and Janga, there have been clashes between the police and the local population”.

Another interviewee described Nacala as a “dangerous zone for the police. … When the population feels that the police are not solving their problems, they start to act out”, continuing: “There are always riots and people destroy police stations … The police have gone to the extent of removing a police station from a neighbourhood, and the district government temporarily closed a health centre.” Police stations were again targeted and vandalised in unrest surrounding the contestation of election results in 2024 (alongside prisons and medical facilities). Against this backdrop, the state was described as “losing credibility”, with instances recounted of police officers killed in clashes with residents. Implications for senior officers were also described: “Nacala is known as the district where police commanders don’t last long”.

Terrorism and Conflict in Cabo Delgado

Despite their perceived prevalence, in recent years the threats described above have not commonly been prioritised in top-level national and international discourse on the security situation in Mozambique. Instead, over much of the past decade violent extremism in Cabo Delgado has routinely been framed as the dominant feature of the security landscape and the core driver of instability in the country, likely due to the wider potential implications of this threat. Referring to the actions of the extremist group known locally as Al-Shabaab (though unrelated to Al-Shabaab in Somalia), insurgent violence in the country’s poorest region has a long history: the rise of today’s insurgency can be traced back to Mozambique’s war of independence, with the beginning of Frelimo’s attacks against Portuguese positions in Cabo Delgado launched from across the country’s northern border with Tanzania. Yet, rightly or wrongly, the prominence of violent extremism in state and donor narratives may be increasingly out of sync with local-level perceptions of “lived insecurity”. Articulating this view, some interviewees argued that a singular counterterrorism focus risks coming at the cost of concerted action to address more immediate community concerns in a range of locations.

A brief overview of the violent extremist landscape in Cabo Delgado is required before respondents’ perceptions of the attendant security threats are examined. The start of the insurgency is generally attributed to a string of attacks on police stations in the port town of Mocímboa da Praia in October 2017. This came in the wake of reports of repeated clashes between youth gangs and religious authorities in Mocímboa da Praia in the preceding years – with the violence attributed by the Mozambican government mainly to foreign destabilisation efforts. By 2021, a group known locally as Al-Shabaab (“the youth”) had grown in capability to the point of having dealt major blows to Mozambique’s military. While foreign fighters appear among Al-Shabaab’s fighting force, Mozambicans make up the majority of this force, which remains Mozambican-led.

The group appears to have developed as a small sect espousing a brand of fundamentalist Islam and anti-authority ideology at odds with the approach of established local Muslim organisations. While it initially issued few public messages, the announcement of formal affiliation with the Islamic State in 2018 drew international attention. The Mozambican group’s leaders subsequently stated its objectives – to establish an Islamic state and ensure the application of Sharia law – while shifting the group’s focus from primarily targeting security forces to widespread attacks on civilians. From 2017 to 2021, the conflict resulted in 3,000 civilian deaths and the displacement of close to 1 million people, with towns and villages burned and whole communities forced to flee. And despite losing ground in the face of military interventions, the group has demonstrated an ongoing ability to inflict damage.

The government’s principal response was to launch a large-scale counterinsurgency operation. This is widely viewed to have been ineffective, failing to quell Al-Shabaab activity. The heavy-handed response has also had damaging effects, with security forces destroying mosques and killing civilians, harming relations with local communities. From late 2019, the government turned to private military companies for help, including the Wagner Group and South African Dyck Advisory Group, but the impact on the insurgency appears to have been limited. By 2020, Al-Shabaab controlled a significant number of villages and towns, was active beyond Cabo Delgado in neighbouring Niassa and Nampula provinces, and in March 2021 made international headlines by seizing the northern town of Palma.

Palma is in close proximity to major gas fields, and the attack on the town was widely viewed as a turning point in the conflict. The incident put pressure on the government to accept foreign assistance. In 2021, 1,000 soldiers from Rwanda and 2,000 troops comprising a newly formed Southern African Development Community (SADC) Mission in Mozambique were deployed, with numbers subsequently fluctuating. These interventions are credited with having curtailed Al-Shabaab’s influence, isolating the insurgents to remote bases.

Yet the insurgency has not been defeated. Throughout 2022, Al-Shabaab undertook hit-and-run attacks, mainly against undefended villages and isolated military outposts. In 2023, there was a reported drop-off in violence, alongside unconfirmed reports of a shift in tactics to reduce civilian casualties and present the group as an alternative political authority to the Mozambican state. It has been observed, however, that this is more likely to be attributable to counterinsurgency operations and declining manpower, while doubts have emerged over the resilience of the group’s ties with the Islamic State. Despite this, Al-Shabaab remains a threat, capable – as seen in the Macomia district – of launching deadly attacks on military positions and laying siege to villages and towns.

Against this backdrop, interviewee perceptions of the threat posed varied. Outside Pemba, violent extremism featured comparatively less often in discussions of insecurity than the other threats covered in this chapter. When raised, the threat was usually covered in cursory detail in terms of those involved or their agendas. On two occasions, respondents identified “Al-Shabab” and “Islamist terrorists” as threat actors, but generally they referred simply to “terrorists”. Revealing a particular understanding of the nature of the threat, rooted as this is in a Muslim-majority province, respondents regularly articulated the view that there was “no connection” between “religion” or “Islam” and terrorism in terms of the impact it has on Mozambique. A religious leader in Angoche described the insurgents simply as “people who hide behind religion” in pursuit of other goals.

In Maputo, “terrorism” was mentioned just once in interviews. This is perhaps unsurprising given the physical distance between the capital and the insurgents’ area of operations. In Nacala, Angoche and Montepuez, some participants expressed fear in relation to the threat of terrorism, yet more often, respondents did not point to terrorism as posing a significant direct threat to themselves, their communities or their daily lives. Nor did they view this as more than a hypothetical risk in the future.

In Pemba, fear of terrorism appeared more proximate. This is illustrated by such comments as: “I get the impression that terrorists are in every district” and “I am afraid because I can’t tell if someone is a terrorist”. Interviewees in Pemba also reported a significant deterioration in security with the rise of the insurgency. Yet they did not attribute this to the terrorist threat alone. Instead, it was linked to a mix of interconnected factors, from popular opposition to violence suffered at the hands of security forces deployed to counter the insurgency, to the movement of large numbers of displaced people, and routine instances of corruption undermining the counterterrorism response. On the latter point, allegations were made of security officers supplying terrorists with food and arms and releasing suspected insurgents for bribes. One respondent in Pemba observed: “I see the police and military as key players [in fuelling instability] – instead of ensuring security, they become perpetrators of insecurity”.

In Montepuez, some individuals similarly perceived corruption within security forces as bolstering the insurgency. Here, respondents also outlined a fear that facilities for displaced persons might be used to harbour terrorists. The latter perspective forms part of a broader theme across sites, whereby “outsiders” and “foreigners” are viewed as amplifying security issues. At times, terrorist threats were attributed wholly to extra-territorial actors: one religious leader in Angoche proffered the view: “Terrorism didn’t start in our country; it is an act that migrated to our land”.

A distinguishing feature in Nacala, meanwhile, was the focus on recruitment in interview testimony. Respondents described the narratives used to convince local youths to travel north – often centring on job opportunities across Cabo Delgado’s fishing industry. These narratives were commonly held to be based on empty promises: in the words of one respondent, “The police have returned many youths to Nacala who claimed they were going to fish in Cabo Delgado … some leave with job promises but encounter a different reality”.

In such cases, recruits were held to be motivated by unemployment; in others, wider factors were cited. As expressed by one respondent in Nacala, “recruitment is not solely due to the economic vulnerability of the young – recruiters use money and religion, claiming they are fighting the Frelimo government in the name of Allah”. Yet the legitimacy of these religious narratives was questioned: for another Nacala-based respondent, “What’s happening in Cabo Delgado has nothing to do with Islam”.

In Angoche, a connection was made to natural resources. With the Cabo Delgado insurgency perceived to be driven by discontent over exploitation of large-scale natural gas reserves, respondents in Angoche expressed fears that the same might occur in their region. This comes alongside a growing focus on the hydrocarbon potential of the Angoche basin, and moves in 2024 to grant petroleum exploration and production concession contracts to exploit this. As observed by one Angoche-based respondent: “People believe that terrorism in Cabo Delgado is due to gas exploration taking place there – when similar exploration was announced in Angoche, locals started fearing its arrival”.

Often these concerns were linked to the dominance of natural resource extraction economies by foreign entities. Interviewees lamented the lack of local employment opportunities created, with significant ground-level resentment cited where riches are perceived to accrue to an unaccountable elite. Such a discrepancy in the benefit derived from major extractive projects has been identified as a driver of conflict in Cabo Delgado across the published literature. This was reiterated by a traditional leader interviewed in Angoche, who said, “I believe the substantial investments in hydrocarbons create envy. … When we see our ecosystem being destroyed, which used to sustain us, by foreigners, we start questioning. And because we lack training, we are nothing in this world.”

Illicit Economies and Organised Criminality

Interviewees often framed a deterioration in security with the rise of the insurgency as part of a wider picture of fluctuating instability. Many did not attribute instability to the terrorist threat alone, citing a range of interconnected factors as influencing the broader security landscape. Beyond entrenched street-level insecurity, corruption and human rights abuses, a further set of threats cited by interviewees in this context were illicit economies and organised crime – described as key sources of ground-level insecurity in a range of sites.

A civil society activist interviewed in Pemba perceived a growth in opportunities for organised crime with the rise of the insurgency in Cabo Delgado. They did so noting: “In addition to terrorism that began in 2017, … other types of crimes … including … drug trafficking, corruption, and organised crime … have become more prevalent … due to the conducive environment created by terror activities”. This points to a perception of causality whereby the insurgency is viewed to have established the context for organised criminal activity to thrive. Others perceived the situation differently; in the words of another respondent in Pemba, “we still face older crimes that have always been prevalent in Cabo Delgado province, such as smuggling of timber and precious stones”.

In other interview sites, organised crime was similarly described without reference to the insurgency. The principal dynamic cited was transnational drug trafficking – a particular concern among respondents in Angoche, Nacala and Pemba. Trafficking in people, timber, precious stones, marine and other natural resources were mentioned, but less frequently than drugs. The vulnerability of the country to illicit flows in all these commodities is detailed across the published literature.

Interviewees most commonly cited heroin and cocaine as concerns, as well as suruma (the local term for cannabis). While Cabo Delgado and Nampula provinces were perceived as “corridors” for large-scale transnational drug trafficking, descriptions of localised drug abuse were also proffered. In some cases, localised drug use was viewed as a consequence of transnational drug flows, and a driver of other threats. In Pemba, young people were described as “plagued by drug use and unemployment” and engaged in systematic “theft … to feed their addictions”. The public health implications of such localised consumer markets are documented across the literature.

Beyond Pemba, interviewees expressed similar concerns in Nacala and Angoche. This aligns with shifts in illicit drug flows identified by external analysts. In 2022, for example, GITOC pointed to a southwards movement in the Mozambican drop-off points used for heroin moving from the Makran coast of Iran and Pakistan en route to consumer markets in South Africa and Europe. Whereas Pemba previously served as the southernmost offload point, a desire to avoid potential insurgency-related supply-line disruption has reportedly seen growing use of landing points in Nampula and Zambezia provinces. Today, drug flows via Nacala and Angoche are widely documented, while GITOC analysis points to an absence of evidence of established links between the insurgency and transnational drug trafficking. This is despite public statements by Mozambican and regional forces pointing to insurgent involvement in this and other criminal markets.

Similar uncertainty over the interface between terrorism and organised criminality was expressed by many respondents. Some pointed to the imperative of adequate financing to sustain terrorist operations, citing organised crime as a likely funding stream. Others referenced the level of weaponry and equipment held by insurgents, pointing to potential links with arms trafficking flows. In further accounts, the conceptual separation between terrorism and organised crime was probed. One interviewee described the two phenomena as one and the same: “I don’t see a clear boundary between these concepts as one implies the other. If violent extremism exists, it’s a form of crime.”

At times, respondents pointed to similarities in the underlying conditions fuelling both phenomena. At a local level, lack of opportunity was described as driving both crime and terrorism, while the perceived role of outsiders was also stressed. In Angoche, frequent reference was made to the city’s coastal location as a source of vulnerability. As one respondent noted: “our coast … can serve as an entry point for individuals linked to terrorist acts. Some say this is linked to drug trafficking, for which our area is a transit channel”. As expressed by another interviewee, “Angoche … is located on the coast, where most violent crimes in Southern Africa occur. … Angoche, with its many islands, … has also been seen as a provider of recruits for terrorists in Cabo Delgado, contributing to … insecurity affecting not only the district but also the province and the whole country.”

Interviewees often viewed corruption as an enabler of organised crime, with compromised officers within police, military and customs authorities perceived to be facilitating drug and other illicit flows. Drug traffickers working via the Port of Nacala were described as so “sophisticated” and “powerful”, that “if a customs officer finds drugs in their goods, they will not report it for fear of retaliation”. Beyond facilitation, direct involvement by customs and police officers was reported by interviewees in Nacala. As alleged by one respondent, “The tax haven [enabling transnational organised crime] starts at the Port of Nacala and extends to the Ministry of the Interior”. As perceived similarly in Pemba, organised crime “is always linked to a power holder, be they community-based, economic or political”.

At a purely domestic level, organised crime was also referenced in relation to some dimensions of street-level insecurity, where a spectrum ranging from ad hoc to organised was described. At the less sophisticated end of the scale was spontaneous criminality involving under-employed youth with few job prospects – held by some to be perpetrated by “groups of criminals that form randomly, without structure or organisation”. Toward the other end of the spectrum, muggings, burglary, robbery and car-jackings were at times described as organised and premeditated, conducted by structured groups of individuals. While these groups were generally not named, in Nacala some respondents spoke of an organised entity labelled “Group 15”. The research team was unable to find external analysis or reference to “Group 15”, but interviewees described it as “well organised”, “having leadership” and selecting its victims strategically.

Across all threats, a final note should be made of the lack of confidence expressed by some interviewees in their knowledge base. Some lamented a perceived lack of transparency on the part of the government in relation to illicit markets and insurgent and counter-insurgent operations. Others stressed their inability to answer certain questions, expressing awareness of knowledge gaps on the threats plaguing their everyday lives. As a businessperson interviewed in Angoche stated, “I can’t precisely explain the state of security provided by the government; as a population, we are in the dark”. These sentiments were often expressed alongside a deeply felt sense of resignation to the fact that many households, and indeed whole communities, ultimately feel left to provide for their own security.

II. Contexts and Wider Systems of Violence

The biggest mistake analysts make is viewing insecurity as an isolated issue. We must look deeper into the root causes … leading individuals to create problems.

- Interview with traditional leader, Pemba

The contents of Chapter I offer a snapshot of community perceptions in multiple locations across Mozambique. They should not be viewed as a comprehensive or fully representative reflection of other subnational or national perceptions, and are all subject to contextual sensitivities, personal preferences and imperfect information. Together, however, they provide an authentic interpretation of “lived insecurity” across specific moments and places.

Crucially, such insights also offer important opportunities to assess the underlying mechanisms sustaining the threats cited. Indeed, while individual manifestations of insecurity may appear to be discrete phenomena, many coexist as symptoms of the same wicked problem, the full scope of which must be considered for effective policymaking. Drawing on both interview data and published sources, this chapter examines the systems enabling the wider environment of insecurity outlined in Chapter I – an accurate understanding of which raises important considerations for national and international programming.

“A War Against the State”: Elite Capture and Insecurity

Perhaps the most prominent context of insecurity in Mozambique as viewed through a bottom-up lens is the interplay between insecurity and corruption and the role of elite actors. As noted, interview testimony across all sites reveals a deep-seated mistrust of state authority, with corruption across the state security apparatus viewed by interviewees as pervasive. In highlighting the perceived issues posed by elite dominance of corrupt networks, most respondents referred to politicians or political networks abstractly.

Despite this, the perceived involvement of political actors in corrupt networks emerged frequently, and was viewed as contributing to the impunity described in Chapter I. Interviewees expressed the belief that political and economic elites can easily circumvent relevant legislation, allowing them free rein in relation to land rights, the terms of transportation of licit (and illicit) goods, and other key sectors of the economy. As a civil society activist in Angoche expressed: “Nowadays, if you have money, you can do whatever you want … with millions, one can corrupt the judiciary and even the government”.

These views are hardly surprising against the backdrop of egregious corruption scandals involving members of the country’s political leadership. These include the so-called “hidden debt” scheme – the largest corruption scandal in Mozambican history, involving secret loans amounting to US$2 billion secured with undisclosed government guarantees. This episode saw corruption uncovered at multiple levels, implicating senior politicians – including former Finance Minister Manuel Chang – and plunging Mozambique into protracted economic turmoil. The cost and consequences of the “hidden debt” scandal to the country has been estimated at $11 billion – or $400 per citizen. With the trial of key players ongoing, the scandal has seen Mozambique’s ranking deteriorate on relevant global indices of governance, democracy and corruption for the 2010s and beyond.

In this environment, political elites were viewed by interviewees as responsible for corruption among lower-level state security actors. While the mechanics of these relationships were not described, respondents frequently depicted elites as holding mastery over the nature of security actors’ involvement in corruption and integration into criminal networks. These dynamics were interpreted by respondents as part of a hierarchy of corruption, with lower-level security actors paying rents and kick-backs up the chain. Senior officials were depicted as guarantors of protection from law enforcement action and were described as using their status to influence investigative processes and judicial outcomes. An official interviewed for this research provided an example: “I know of a case at the airport police station. They seized rhino horns. Who came to collect them? The chief police commander. He came to the station and took them [away]”.

Importantly for this report, threat actors were viewed as having seized on pervasive corruption as a key political narrative – to wield what academic Fernando Jorge Cardoso has called “a war against the state”. This narrative is held to enable both insurgent actors and organised criminal networks to recruit new members and legitimise their activities. The term was used by Cardoso in relation to Cabo Delgado, describing the transformation, with the support of growing numbers of foreign fighters, of an initial “revolt” into an “ideological, civilisational and terrorist war” against “the state and the way of life of the overwhelming majority of the population”.

Cardoso holds that the insurgency is targeted “not against the government or bad governance … [but] against the state” itself, although this does not “eliminate the government’s responsibility in relation to abuses of power, mal-practices … or the capture of mineral wealth, or its rent, by elites, … [or] the behaviour of government authorities or the inaction (by some considered complicity) against some type of urban banditry”.

While Cardoso distinguishes between the two, in reality it is challenging to separate state from government in a country in which Frelimo has governed uninterrupted, despite the transition to a multiparty democracy in 1992.

Indeed, public reports of abuses by government actors only exacerbate the “war against the state” he depicts – and the opportunities offered in terms of recruitment. In Cabo Delgado, for example, such a “war against the state” is reportedly being waged in an organised and explicit fashion by insurgent actors comprising both radicalised Mozambicans and foreign fighters. While drawing at times on the ideology of the Islamic State to organise and remain connected to the wider global movement – as well as to strengthen recruitment and fundraising efforts – attacks are often framed as addressing popular grievances against authority structures. Indeed, an anti-state, anti-Frelimo narrative features in much Al-Shabaab messaging, with the group presenting its legitimacy to rule as greater than the elected government.

Examples of this rhetoric abound. During a 2020 attack in Muidumbe, a spokesman stated: “We invaded to show the current government is unfair. … It humiliates the poor and gives advantages to the rich.” In this context, the insurgents claim that the creation of an “Islamic government” will offer a more socially just alternative, exploiting a perceived deep-seated lack of confidence in the state across communities in the country’s north. This approach is viewed by many as having resonance; in the words of human rights campaigner David Matsinhe, “the government has done for the radical preacher 80% of the job. He just comes to harvest.” As argued further by GITOC, “the policies (formal and informal) which have in the past secured the economic interests of the elite now play into the hands of the insurgents”.

Both corrupt governance and elite control of resources have been leveraged in insurgent rhetoric. Al-Shabaab leader Maulana Ali Cassimo has reportedly referenced past governmental injustices against artisanal miners in Niassa Reserve, presenting mining as one of the few remaining livelihoods for local people facing poor public sector service delivery. State security institutions are referenced using dehumanising terminology, with local people warned against “collaborating” with these agencies. Yet for all the efforts made to frame attacks in this way, insurgents have shown little interest in governing or maintaining power, when compared to negative strategic objectives such as forcing stoppages in the regional gas project and using terror to further undermine trust in state security.

Beyond support for the insurgency – and across the rest of the country – such a “war against the state” is experienced in different ways. In terms of support for organised crime, perceptions of entrenched state corruption and grievances around elite dominance of the formal economy serve as powerful factors incentivising local level involvement. Such dynamics have been observed in Nampula, Mozambique’s third-largest city, and in nearby Nacala Port, where they are held to manifest as part of a comprehensive “crime–conflict economy” featuring symbiotic relationships between organised criminal networks, some state institutions, the private sector and, at times and inadvertently, international donors.

While Nampula has long been a regional drug trafficking hub, this “crime–conflict economy” is assessed to have developed into an umbrella political–economic structure through which organised crime operates with impunity – and often explicit protection – from state security agencies. Local stakeholders are held to be easily incentivised to participate, exhibiting limited loyalty to a formal economic system deemed predatory and unfair. As noted by researchers Prem Mahadevan and Alastair Nelson, in a hub such as Nampula, “powerful stakeholders who derive tangible benefits from organized crime … may gain community respect by providing a loose and highly unpredictable form of employment to others who aspire to emulate them. Their wealth and accompanying social status allow them to act as in-person influencers, providing a twisted set of role models, particularly for younger community members, who see them enjoying immunity from both criminal justice and social criticism”.

Similar dynamics are held to play out in subsidiary hubs such as Pemba, Nacala and Montepuez, enabling varying levels of support for organised crime. Meanwhile, further factors may incentivise local-level stakeholders to sustain illicit economies, including perceptions of the interests of this crime–conflict economy as determining decision-making on security agency deployment. The same applies at an international level, with external donors at times accused of playing an unintentional enabling role in this illicit crime–conflict economy.

Controlling the Narrative?

The Mozambican government has proven unable to effectively dispel the narratives propagated by threat actors. An example can be seen in the official response to the insurgency, with the government seeking to wield an impression of control by censoring media reporting and suppressing information flows. A range of cases of detentions of journalists has also been documented. Prominent examples include the arrest of journalist Amade Abubacar on charges including “public instigation through the use of electronic media”, “slander against forces of public order” and “association to commit an offence”, prior to his eventual release following calls from human rights organisations. Other cases include the disappearance in 2019 of radio journalist Ibraimo Mbaruco, whose last text message – “surrounded by soldiers” – prompted speculation that he had been abducted by the military. Despite calls from international organisations to locate Mbaruco, investigations were suspended and the case closed in August 2024, a move denounced by CDD as “a continuation of the Attorney General’s Office’s denial of access to justice for journalists”.

Suppression of independent reporting has been criticised by analysts as a move to prevent the exposure of governmental weakness and state abuses in the region. Yet the resultant information vacuum has not always had this effect and at times has given rise to conspiracy theories. As reported by GITOC, these include narratives “alleging that the insurgency is a false-flag operation instigated by Frelimo politicians, connected to a bid for greater profits from LNG (liquefied natural gas) or the illicit economy”. Meanwhile, many citizens appear dissatisfied with a media landscape characterised by persistent attacks on journalists, with the country dropping to 105th of 180 countries on the World Press Freedom Index in 2024, and Reporters Without Borders pointing to a “worrisome decline in press freedom in Mozambique”.

Where the government has sought to control the narrative, messaging has occasionally been inconsistent. For example, descriptions of the insurgency put forth by government sources have varied, while the threat has sometimes been downplayed. Notably, the Mozambican government initially appeared reluctant to label Al-Shabaab as a terrorist organisation, portraying the insurgents as a collection of common criminals without legitimate grievances. At times, it has pointed to potential exploitation of these individuals by foreign groups to destabilise Mozambique. Yet not until 2019 did Mozambican President Filipe Nyusi use the term “terrorism” in relation to the violence in Cabo Delgado, while official institutional acknowledgement came in April 2020, with the National Defence and Security Council statement that the country was facing “external aggression perpetrated by terrorists”.

Gus Waschefort, Jessie Phyffer and Martha M Bradley posit that such delays may have been attempts “to avoid the application of protections offered by international human rights law (IHRL) and IHL [international humanitarian law]”. Meanwhile, heavy reliance at various stages on private military companies has seen the government accused of lacking transparency about operations in the region. Such perceived obfuscation has ultimately allowed threat actors to control the narrative – in a battle between insurgents and state that has become as much about public rhetoric as territorial advances.

This is concerning in light of reports of a shift in insurgent strategy towards less violent tactics and moves to win over local inhabitants. This has been variously reported to encompass efforts to limit civilian casualties, urge cooperation with insurgents and engage with community members in what some have labelled a “hearts and minds” campaign. As observed by the International Crisis Group (ICG), such a shift in strategy could be aimed at garnering popular support and securing supply lines amidst territorial losses to government and international forces. Such moves risk further entrenching the insurgency, complicating the response and establishing conditions for further retaliatory violence.

Political and Economic Marginalisation

Nationwide, underpinning local level concerns over state capture are a range of wider factors linked to political and economic marginalisation. A crucial context of insecurity in Mozambique as viewed through a bottom-up lens, these factors featured in interview testimony across all sites as expressions of discontent over entrenched poverty, exclusion and lack of employment opportunities. Intimately connected with the disaffection respondents expressed over corruption and inequality of access to key services, these issues must be considered as cross-cutting dynamics enabling these insecurities.

Mozambique is a predominantly rural country, with 61% of its population based in rural areas in 2023. Poverty levels are high: in 2020, 62.8% of the population were reported to live in poverty, based on a threshold of $1.90/day. While the country has abundant natural resources and access to arable land and water, energy resources and deep ports, persistent challenges from political instability, frequent natural disasters and acute vulnerability to climate change have suppressed progress on poverty reduction. Meanwhile, despite advances in a range of areas – from increasing access to education to declining child mortality rates – progress has been uneven, with economic growth assessed to be increasingly less inclusive and levels of inequality rising.

In this context, many respondents identified poverty and inequality as major drivers of insecurity. They identified causal links between unemployment, poverty (including frustration caused by poverty) and the high crime rates seen in a range of sites, albeit without reference to specific criminal activities. A civil society activist in Maputo pointed to opportunistic crime, highlighting as a driver the inability of many residents to purchase food and non-food items to meet basic individual or family needs. Beyond material poverty, inequality was cited as a driver of crime and other forms of violent disorder. A Maputo-based priest summed up a widely held view: “Poverty and social injustice, characterised by significant disparities between the haves and have-nots, create fertile ground for violence. While poverty cannot justify criminality or violence, unfortunately, the desperation caused by hunger may drive people to do anything to survive.”

Unemployed youth were seen as particularly vulnerable. Respondents often spoke of this demographic as being involved in criminality, regularly alluding to their resorting to crime to feed drug addictions. Some cited attitudinal as well as socioeconomic factors, referencing an obsession with money, a desire for rapid enrichment, and an intrinsic lack of motivation to work or study. Yet more often interviewees recognised the lack of basic income and employment opportunities as major drivers of youth involvement in petty crime, mirroring a number of published studies. In these accounts, young people were recognised to be literally hungry, but also figuratively hungry for opportunities, such as education. The barriers they face in actualising these opportunities were cited by interviewees. A youth mentor in Nacala described the need for young people to pay to access employment, marginalising those without the means to do so.

The contrast between this situation and growing investment across key locations was highlighted. The feelings of exclusion fomented where other, often foreign, stakeholders benefit from domestic resources and industries were stressed in all interview sites. Growth was viewed as a negative: in the words of a journalist interviewee, “Nacala is growing, there are new investments and this increases security risks”. Another Nacala-based interviewee viewed this as translating directly into rates of engagement in crime, noting that: “Big companies are not employing locals. Those few who are employed work as security guards, with bosses from [elsewhere] … who … become chiefs upon arrival. If at least 50% of those working in the companies were Nacala natives, we would have very low levels of criminality. Poverty pushes people into the world of crime.”

Large-scale displacement was viewed as exacerbating this situation. As of 2023, Mozambique hosted more than 850,000 internally displaced persons and 30,000 refugees and asylum seekers, according to the UN Refugee Agency. Internally, residents have been forced to flee both ongoing conflict (in the north) and natural disasters (across a range of locations), with the country among the most vulnerable to climate change globally. In the face of violent disruption and displacement, as well as variable experiences of spontaneous return, some scholars have pointed to an increased propensity to join criminal groups. Some interviewees also reported wariness of those migrating in search of employment opportunities. In the words of a teaching professional interviewee, “many people migrate to Montepuez seeking better living conditions, but resort to criminality when they fail to improve their lives”.

Poverty and a lack of realisable opportunities were also identified by interviewees as drivers of violent extremism. In Angoche and Nacala, fear around the ease of recruitment of youth from local communities appeared more substantive than fear of the terrorist threat itself. Particularly in Nacala, young people with limited opportunities were viewed as susceptible to recruitment narratives, including those oriented around job prospects in Cabo Delgado.

The perceived powerlessness of those offering counter-narratives was stressed. As one respondent noted: “Nacala is one of the sources of youth recruitment for Cabo Delgado. Poverty is one of the reasons young people join terrorist groups. Churches, mosques, the media and governmental entities work to sensitise young people to discourage them from joining terrorist groups. However, no speech is strong enough to overcome poverty”.

Beyond recruitment, widespread marginalisation forms a key backdrop for the emergence of the insurgency itself. Cabo Delgado is located in Mozambique’s poorest region, situated over 2,500 km by road from the capital in the south. Despite its economic potential, including sizeable natural gas reserves and deposits of precious stones, many residents remain unable to access meaningful development opportunities, as do others based elsewhere across the country. Instead, as noted by Julia Stanyard and others, they have “borne decades of government neglect, extreme poverty and widespread corruption [while watching] … elite politically connected groups seizing the benefits of the region’s natural resources and profiteering from criminal markets”.

Liazzat J K Bonate has described these grievances as the “emotional entry points” that allowed jihadism to take root in 2017. Yet insurgent exploitation of marginalisation among local communities has a longer history. Specifically, many of the roots of the current conflict can be traced back to Mozambique’s colonial period and 1964–74 War of Independence. Cabo Delgado was a pivotal area during the war, as local people protested oppression by Portuguese colonial authorities who were viewed as unfairly extracting, and denying local people access to, wealth and resources. The Frelimo movement advocated for independence as the route to social justice and a fairer share of the province’s wealth. As such, similarities exist between the narratives of today’s insurgents and those previously seeking to overthrow the colonial regime. As noted by Joseph Hanlon: “The grievance was the same in both wars. Independence had been the flag 50 years earlier; this time, the flag is Islam.”

Since independence, the memories of many individuals nationwide, from guerrilla fighters to political prisoners and other marginalised groups, reveal a sense of exclusion from post-colonial narratives around national identity and development. These issues played into the wider conflict waged between Frelimo and then-guerilla organisation Renamo from the late 1970s. Similar sentiments still resonate, with unrest following Frelimo’s declaration of victory in the country’s 2024 presidential elections causing many Mozambicans, as described by Marilio Wane, to “ask whether the values, symbols, and institutions created to give shape to ‘national unity’ are still legitimate today”. Indeed, while the ruling Frelimo party has asserted its “ownership” of the liberation war’s legacy, other groups across the country have historically felt marginalised in the official narrative. Notably, in the post-independence era, the government adopted a highly centralised model that sought to concentrate power and integrate diverse regions under one national identity. Yet this model struggled to incorporate many citizens, feeding the level of popular support for Renamo during the country’s civil war in many rural locations.

Indeed, many citizens have long felt excluded from the political and economic benefits that followed independence. In the north, meanwhile, the introduction of Wahhabi ideologies in the late colonial period, which diverged from local Islamic traditions, added another layer of complexity. Over time, radical Islamist ideologies began to resonate with segments of the population, as socioeconomic conditions in the region deteriorated. In recent years, insurgents have drawn on widespread discontent, positioning themselves in opposition to the state’s perceived failures, and exploiting inequalities drawn over socioeconomic, political, religious and ethnic lines, including between the Muslim-majority Makua and Mwani and the Christian-majority Makonde. In this context, the drivers of insurgency can be summarised as “a cocktail of interconnected internal factors: from socio-economic and socio-political factors to cultural and religious factors and factors related to the lack of public security that [have] allowed illicit activities to flourish”.

Nationwide, underlying discontent around marginalisation has also historically been intertwined with the broader relationship between the Mozambican state and traditional authority structures – a contentious issue, notably, during the country’s civil war. For much of the post-independence period, the state actively suppressed “customary” or indigenous leadership in favour of a more centralised form of governance. In recent years, such a contest between state control and local autonomy has persisted and, in some locations, feelings of exclusion may be exacerbated by the disconnect felt by community members from decision-making under a centralised model of government. Despite the implementation of a range of processes to expedite decentralisation, little power and authority is devolved to local administration structures in practice, limiting the potential for citizen participation and influence around issues affecting them at the local level.

Over time, large-scale exploration and resource exploitation have served to exacerbate some of these tensions. While Mozambique is a country blessed with natural gas and oil – as well as rubies, graphite and precious woods – the benefits of these industries have not been shared equitably with the population. Instead, extractive and other private industries are at times perceived to have exacerbated inequality, with local communities facing displacement, environmental degradation and a lack of economic opportunities. This fits into a longer-term trajectory, with recent decades having seen a deepening of the basic patterns of extractivism instigated by Mozambique’s integration into global markets.

Collaboration between foreign investors and local elites across the political economy of natural resource extraction in the country has exacerbated the sense of marginalisation felt by many citizens. Excluded from the benefits, some are reported to have turned to the insurgency as a means of expressing their discontent and seeking material rewards. Egna Sidumo, Marko Svicevic and Martha M Bradley point to the Palma attack as a pivotal expression of these dynamics in the north, given Palma’s proximity to major international exploration projects. Described as a “turning point” in the conflict, the attack forced the latest suspension of French energy company TotalEnergies’s operations at its $20-billion gas liquification plant in the region – the largest foreign investment in Africa. Sidumo, Svicevic and Bradley point to discontent over the granting of significant areas of land in Cabo Delgado on concession to prospectors, investors and mining companies, “reinforcing arguments according to which the main cause of the conflict … was the interest of global powers, allied with national political elites”.

In this context, it is clear that many of the security challenges witnessed, in Cabo Delgado and beyond, are deeply intertwined with the country’s colonial and post-colonial history and should be understood as the culmination of long-standing sociopolitical and economic trends. The insurgency in particular should be seen not only or even primarily as a product of radical Islamic ideology, but rather as a response to long-standing grievances about exclusion, injustice and deprivation. The insurgency is also a product of the Mozambican state’s perceived inability to address these grievances at best and active complicity in driving them at worst. At times, this may separate out the motivations of Mozambican and foreign insurgents. As noted by the ICG: “While foreigners have joined in the name of jihad, most of the Mozambican rank and file militants are motivated by their perceived socio-economic exclusion amid major mineral and hydrocarbon discoveries in the region”.

Elite Dominance of Illicit Economies

In this discussion of marginalisation, the legitimate economy is not the only domain of relevance. Instead, differential access to entrenched, and thriving, illicit economies was a further point of reference in interviews across all sites. Here, Many respondents pointed to elite control over both commercial natural resource exploitation and the proceeds of crime generated by the range of large-scale trafficking flows to which the country plays host.

Mozambique has long acted as an illicit economic corridor to the wider region. Due to its geographical location, extensive coastline and proximity to consumer markets in South Africa, the country is a transit hub for an array of illicit commodities. State-embedded actors are alleged to be involved in a range of illicit economies, as well as in providing protection to criminal enterprises. As noted by GI-TOC: “The ruling party tends to be reactive rather than proactive towards criminal entrepreneurship, allowing for high-level state corruption and virtual impunity for politicians accused of corrupt relationships with criminal traders”.

Popular discontent over a perception of elite actors profiteering from criminal markets has deepened as Mozambique’s role in a range of illicit supply chains has remained in flux. Drugs and illicit extraction of natural resources are reported to comprise the largest criminal markets, while the country also acts as a significant waypoint and destination market for human trafficking. In parallel, illicit trade in excise and counterfeit goods with neighbouring countries is assessed to be rife, damaging key aspects of the licit economy.

Since the 1990s the country has been a major hub along the so-called “southern route” for heroin originating in Afghanistan, moving via Pakistan and Iran to East Africa, mainly via Tanzania and Mozambique, and on to South Africa and Europe. In recent years, trade in methamphetamines along the same route has reportedly also grown, alongside increased production in Afghanistan. The criminal groups involved are reported to be active in all of Mozambique’s provinces, as part of what GITOC has described as a “tightly regulated network run by families connected to the political elite, with the state controlling most of the heroin market”.

Mozambique is also increasingly viewed as a node in the global cocaine trade. This comes as part of a growing set of connections between domestic criminal actors and organised crime groups in Latin America, especially Brazil. Indeed, the arrest in Maputo of a high-ranking associate of the Brazilian criminal organisation Primeiro Comando da Capital (PCC) suggests that Mozambique is becoming a more significant part of the “Lusophone route” for cocaine from Latin America to Africa, alongside longer-established nodes in Portuguese-speaking Angola, Guinea Bissau and Cape Verde. Mozambique’s pre-existing involvement in other illicit networks reportedly provides the infrastructure to facilitate the transit of cocaine, with allegations that the PCC collaborates closely with domestic political elites to protect their operations. This is reportedly enabled by increased maritime trade links with Brazil, initially sparked by a 2015 trade agreement between the two countries facilitating the import of containerised goods, including those in transit.

In terms of natural resources, a range of illegally sourced and traded products are of particular concern. Notably, Mozambique has substantive timber resources, which were exploited extensively under Portuguese rule. In 2013, the Environmental Investigation Agency (EIA) estimated that 76% of timber exports comprised illegally cut wood, with China the major consumer market. In 2024, EIA found that an average of 500,000 tons of timber had been exported to China annually since 2017, in violation of Mozambique’s log export ban.

Meanwhile, Mozambique’s mining industry has been plagued by illegal activity. The country has rich deposits of gemstones, trade in which was largely dominated by artisanal miners in the 1980s and 1990s, when the country was closed to large multinationals. Into the 2000s and 2010s, the sector was subject to greater regulation, with the government granting concessions to large companies and introducing compulsory mining licences in 2016 for artisanal miners. Yet reported control of concessions by political elites, combined with allegations of widespread corruption and a lack of alternative livelihoods, has pushed many miners to operate illegally as part of an extensive illicit market.

Wildlife crime is reportedly rife, with the country acting as a source and transit hub for numerous commodities. Mozambique has long played a significant role in transnational ivory trafficking, as indicated by a substantial decline in domestic elephant populations and large numbers of seizures elsewhere linked to the country. According to the Great Elephant Census, from 2009 to 2014 the country lost 53% of its elephants, while EIA has identified Pemba as a key hub for the processing of multi-ton consignments destined for China. It is also assessed that Mozambique acts as a significant source and transit point for commodities from rhino horn to lion and abalone, while suffering high levels of illegal, unreported and unregulated fishing enabled by corruption across supervisory bodies and political elites.

Alongside legitimate sectors of the economy, the alleged capture of these illicit industries by domestic elites was lamented in interview testimony. This was a particularly prominent theme in locations where the shadow economy exists as the pre-eminent sphere of economic activity. The country’s north, for example, has always served as a key economic corridor when it comes to illicit trade. With Tanzania serving as the closest source of supplies for many communities, centuries of informal cross-border trade across an estimated 40 informal entry points have allowed transnational smuggling to flourish.

Today, as perceived by a teaching professional interviewed in Pemba, “organised crime … appears to be on the rise”. In illustrating this perspective, interviewees cited major seizures, including in Mocimboa da Praia of more than 100 containers of unprocessed wood, as well as heroin and cocaine seizures in Macomia and Murrebué. In such cases, a perceived lack of action following the confiscation of illegal commodities was lamented across interview testimony. In the words of a civil society activist in Pemba, when the authorities seize drugs, “they don’t follow through on cases, leading me to believe that organised crime has political hands or powerful backers at the community, provincial, national or even international levels”.

Many interviewees framed perceived elite dominance of these illicit trade networks through the same lens as the capture of gas and gemstone profits within the legitimate economy, stressing the exclusion of local communities from the resultant benefits. Similar parallels have been drawn by international observers, who have highlighted the totality of elite dominance of the region’s politics and economy. Stanyard and others have shown financial rewards from Mozambique’s illicit economies to accrue to many of the same senior political stakeholders and business elites that have used corrupt practices to cultivate interests across the country’s mining, forestry and transport logistics sectors. The disaffection this has bred at community level is sorely felt; for Stanyard and others, “the scale of corruption and elite control is matched only by Cabo Delgado’s extreme levels of marginalization and poverty”.

National and International Responses

The majority of this chapter has examined cross-cutting issues underpinning the insecurities outlined in Chapter I. A vital question that remains is how far, and how effectively, have national and international stakeholders sought to respond to these cross-cutting problems? Interviewees offered perspectives on this question which, combined with analysis across published sources, provide insight into popular perceptions of the response to core drivers of the security issues identified.

Across all sites, a common view emerging in interview testimony was that national and international interventions have failed to engage with many of the underlying contexts and drivers of insecurity as viewed through a bottom-up lens. Most prominently, respondents denounced a perceived failure to address systemic corruption, improve governance or promote the rule of law, among other fundamental issues. International stakeholders were viewed as having done little to incentivise national-level actors to adjust their approach, with these views amounting to an overarching impression of pessimism regarding the potential for meaningful change.

As noted previously, for much of the past decade, the dominant concern of national and international stakeholders has appeared to be counterterrorism. Several interviewees expressed the view that this focus may be out of sync with other local-level concerns, coming at the cost of interventions targeting more immediate community security priorities in a range of locations. The national-level response was described as highly state-centric, based on an approach favouring military responses over a more holistic approach. In Cabo Delgado, for example, the government is held to have primarily relied on its armed forces and police, private military companies and foreign military forces to quell the insurgency, over concerted efforts to address the conflict’s deeper causes. Elsewhere in the country, a similar lack of engagement among national and international stakeholders is reported, with the underlying factors enabling insecurity.

Attempts have been made to redress this balance. In Cabo Delgado, for example, with donor support, the government has developed a $300-million reconstruction plan and a $2.5-billion Northern Resilience and Integrated Development Strategy (ERDIN) to include other northern provinces. The latter is designed in line with three pillars that speak to a more holistic approach: “support for peace building, security and social cohesion”, “reconstruction of the social contract between the state and the population” and “economic recovery and resilience”. Community-level participation is also planned in the development of a National Body for Peace and Reconciliation, charged with addressing the underlying drivers of violent extremism and preventing conflict across the country as a whole.

Yet these plans have been dismissed by some in the face of a perceived lack of political will. Some respondents remained sceptical of the government’s stated intention to provide broader development, viewing action in this space as a means of prioritising the government’s energy interests in areas where multinational companies operate.

This critique is echoed across the published literature, as are concerns over a perceived governmental failure to recognise the contribution of internal factors – including marginalisation and state violence – to insecurity. Beyond this, observers note that patterns of governance in the north remain little changed, with the ongoing perceived influence of corruption and political patronage viewed offering little hope that the benefits of development and reconstruction funds will be spread widely. As articulated by Stanyard et al. with reference to the situation in the north specifically, “the prevailing view in Mozambique is that the government will prioritize security for gas developments in Cabo Delgado, rather than sustainable development for the local population and improved governance.”

The governmental response to organised crime is similarly held to centre on law enforcement. Here, resources have been focused on the pursuit of individual actors, rather than efforts to prevent crime from happening or to tackle cross-cutting factors enabling illicit activity. Such an enforcement-focused response is itself held to suffer key gaps: although the country’s legal framework is generally judged to be adequate, capacity to investigate and prosecute remains limited. Meanwhile, the collective understanding of key threats and associated drivers remains incomplete across a number of forms of criminality.

While various administrations have stressed the need to tackle organised crime and corruption, the response is further complicated by the perceived extent of complicity of state-embedded actors. Indeed, with some state actors widely held to be complicit in hindering countermeasures, the main barrier to an effective response may be neither strategy nor law enforcement capacity, but a lack of political will to tackle the problem. This characterisation may be supported by the disproportionate targeting of law enforcement action against those involved in low-level transactions with little political protection. At the same time, interviewees lamented what was perceived as the regular dismissal of cases against organised criminals where political protection is in place.

As such, state-security institutions were viewed as having limited ability or interest in meaningfully addressing criminality. In this context, interviewees remained sceptical that programming aimed at professionalisation – such as a 1997–2007 $33.7-million reform programme led by the United Nations Development Programme that focused on capacity-building, training and equipment – had significant impact. Similar views were expressed in relation to efforts, with international support from the 2000s, to support community engagement and policing – although in some cases, non-state forms of policing and dispute resolution have evolved into alternatives to a mistrusted state system.

Donor-Led Approaches: Incongruent and Unbalanced

The role of donors in addressing insecurity was addressed specifically by some interviewees. Over the past decade, many described international funding as having disproportionately focused on supporting the Mozambican armed forces to counter the insurgency in the north. Military interventions by Rwanda and SADC were highlighted as the most visible element, with the US also providing counterterrorism training to the Mozambican armed forces and the EU launching a €15-million training mission in 2022, alongside providing €89 million-worth of non-lethal equipment and supplies. Nationwide, beyond the situation in Cabo Delgado, international support in other forms was also cited. Agencies such as UNODC have provided advice on policy and strategy, as well as technical assistance, to tackle organised crime and terrorism.

Such a donor focus, alongside the government’s prioritisation of law enforcement and military approaches, has prompted analysts to call for a stronger international focus on the underlying drivers of insecurity. Academic Albert K Domson-Lindsay, for example, has called for international security actors to adopt a broader approach to security in Mozambique that goes beyond a simple fight against Islamic fundamentalism. While a military response is required, he argues, this must come in conjunction with concerted, long-term efforts to address broader drivers of insecurity across the country – identified as poor governance, centralisation of power and marginalisation of minority groups.

Recognising these issues, various interventions have sought to strengthen governance and build integrity across the security sector. As part of its “Strategy to Prevent Conflict and Promote Stability, 10-Year Plan for Mozambique” (2022–32), the US has prioritised professionalising institutions across the country through the development of more transparent, accountable and effective systems, including in the security sector. Via the Maputo Roadmap framework, UNODC is engaged in work nationwide to strengthen the rule of law through improved justice systems and integrity in the field of crime prevention and criminal justice, among other outcomes. Other interventions have sought to build resilience through participatory approaches across the country, with the EU providing security assistance as part of an “integrated approach” encompassing humanitarian, development, regional and local peacebuilding efforts, as well as police capacity building and initiatives to foster community-level dialogue among civic leaders. Meanwhile, in 2021, the World Bank approved Mozambique’s eligibility for the Prevention and Resilience Allocation, which provides enhanced support for countries at risk of escalating into high-intensity conflict or large-scale violence, while granting $100 million from the International Development Association to support recovery activities focused on restoring livelihoods, building social cohesion and improving access to basic services.

From existing baselines, however, it is likely to take generations for these interventions to have the necessary effect. Crucially, despite this activity, Mozambique’s ranking has worsened on core governance matrices, with the country reported to be in urgent need of more transparent, participatory and decentralised governance. Further complicating international assistance in these areas is the need to account for longstanding concern over state actors misusing development assistance: over the last two decades, donors have regularly suspended budgetary support, the principal modality of aid to the country, because of concerns about corruption. In many cases, however, aid flows have resumed in the absence of meaningful reform.

These issues speak to further critiques of external engagement in these areas. A prominent local-level narrative questions whether donors are genuinely interested in improving governance as a major factor hindering sustainable development and improved security. This perspective draws on a longer-term view of international assistance, dating back to the aid operation put in place in the 1990s in the aftermath of the civil war, through large-scale efforts to privatise state-run industries, to pare back the public sector and to deregulate the economy according to the neoliberal mandate of the Bretton Woods institutions (the World Bank and the IMF). This orthodoxy saw the provision of aid linked closely to the country’s compliance with neoliberal economic principles, almost to the exclusion of other considerations.

Against this historical backdrop, donor efforts to support reform of the criminal justice system are recognised to have lacked cohesion. More fundamentally, the international community is itself accused of creating the conditions for insecurity to thrive. In the words of analyst Alexander Chance:

The priority attached to rapid economic liberalisation without adequate social security provision and regulatory oversight appears to have directly fuelled corruption. Trumpeting “good governance” whilst consistently failing to follow-up on concerns not only demonstrated donors’ weakness in this area, but also left members of civil society dangerously isolated … Similarly, an unwillingness to take seriously and act upon warnings over the financial clout and political patronage of organised crime left the criminal justice system incapable of halting its reach into the highest echelons of the state. Over time, Mozambique’s apparent success story in development terms was allowed to dominate donor narratives, whilst corruption and criminality exacerbated inequality and fuelled distrust of the state.

With international exploitation of the country’s energy riches booming, such concerns remain rife. A number of commentators have accused international actors of systematically overlooking corrupt behaviour to secure lucrative natural resource deals. Rwandan forces, for example, are accused of concentrating deployments on sites in the north known for their potential for gas exploitation and gemstone extraction, with speculation that France’s funding for the expedition is motivated by French company TotalEnergies’s interests in Cabo Delgado. As analyst Joseph Hanlon observes: “The government survived the very loud donor-cut off of significant aid in 2016 due to the $2 billion secret debt. … Donors and now the IMF have quietly crept back into the fold, without major concessions by government. Foreign governments want a presence because of the gas and the potential to win huge contracts. So they cannot afford to challenge Frelimo misconduct.”

Similar concerns emerged during interviews in a range of sites. In the view of a journalist in Maputo, “foreign entities continue to inadvertently prop up a flawed security infrastructure” without adequate oversight or accountability. Particular concerns were expressed over the manipulation of donor support in ways held to be harmful to the population. As summarised by the same respondent, “foreign entities … tend to supply weaponry and other equipment [that are used] not for the enhancement of security … but rather … to quash peaceful protests”. In this context, calls were made for international actors to “re-evaluate their role in security, considering not just the provision or sale of equipment to Mozambican authorities, but also the ethical terms of its usage to genuinely bolster security instead of furthering a specific, often harmful, agenda”.

A final perspective emerging from interviews concerned a strong sense of disaffection over programming undertaken without adequate local-level consultation. Such a sense of exclusion was expressed in all sites, with concerns raised around the effectiveness – and potentially harmful effects – of associated interventions. As noted by one respondent, “here in Nacala, we have national and international NGOs. Their interventions are not preceded by studies to understand local dynamics … they arrive with set goals, more concerned about numbers than necessarily solving problems for the youths”. As expressed by a civil society activist in Maputo, “one significant error made by foreign analysts when assessing Mozambique’s security context is their lack of a holistic understanding of the ground reality”. This was dismissed in Pemba as resulting in “recommendations that might not align with the actual situation” in key subnational contexts.

Conclusion

Outsiders don’t engage with the community facing these problems, they analyse based on government reports … many of [which] … don’t reflect the real situation.

- Interview with local official, Maputo

Many individuals at the local level in Mozambique live in environments characterised by pervasive insecurity. The objective of this report is to go beyond conventional elite discourse and donor perceptions to build the beginnings of a “bottom-up” understanding of insecurity as a starting point for analysing Mozambique’s threat landscape.

The findings across Angoche, Maputo, Montepuez, Nacala and Pemba – varied subnational contexts with diverse social, economic and political dynamics – offer a range of insights into community-level perceptions and experiences of key harms and dangers. While individual accounts of “lived insecurity” differed across key moments and places, consistencies emerged in the form of a dominant focus on the threats posed by petty crime, street-level insecurity, state abuses and corruption.

In some cases, accounts of petty crime – often targeted at property and the person – extended to a discussion of transnational trafficking and other forms of organised criminality. The threat from insurgent activity, meanwhile, was often perceived as a peripheral issue insofar as it affected interviewees’ day-to-day lives – in all locations except Pemba, where concern over this phenomenon was elevated.

This local-level positioning of violent extremism, often as secondary to more mundane, street-level concerns, speaks to a potential discordance with the priorities of national and international stakeholders. While a range of programmes have sought to improve citizen security and professionalise the police response to day-to-day public harms, the approach to insecurity taken by the government and its international partners over the past seven years was at times viewed by interviewees as reductionist – oriented around a simple fight against Islamic fundamentalism in Cabo Delgado.

Interconnections across threats were frequently drawn in interview testimony. In many accounts, public sector corruption was perceived as a core characteristic of the security landscape – bisecting and enabling many of the other threats cited. Viewed as the defining feature of the country’s governance, the establishment of perverse incentives was perceived to leave many citizens underserved by police and other security agencies. In this context, an overarching narrative of victimisation emerged, with many community-level stakeholders presenting themselves as abandoned or, in some cases, exploited by state security services. This appeared from interviews to have damaged their affinity to a shared social compact, challenging theoretical notions of policing by consent and undermining the perceived legitimacy of the state.

This situation was presented as having set in motion a search for alternative security structures. Interviewees outlined their views of both private security provision and community self-defence initiatives – with both strengths and a range of gaps, weaknesses and legitimacy issues described across these perceived alternative avenues for self and group protection. At times, the pursuit of these options was perceived itself to fuel instability. In the words of a local official in Angoche, “when people feel threatened, their first reaction is to defend themselves. … [Yet] automatically triggered defense mechanisms lead to misunderstandings and conflicts, escalating latent tensions into violence”.

In many cases, local insecurities were understood as expressions of deeply rooted, long-term trends. As argued in this report, they exist within the same enabling environment, as the product of historical path dependencies that have contributed to a widespread normalisation of violence. Most frequently, interviewees perceived elite capture of the state as a key cross-cutting context of insecurity as viewed through a bottom-up lens. Interviewees also pointed to entrenched political and economic marginalisation, often contrasting this with booming international investment in natural resource exploitation.

Elite extractivism and deepening inequality, in turn, were held to have exacerbated latent tensions that could be easily exploited by criminal and insurgent actors. Importantly, such dynamics have played out against a long history with a continuum of violence and contestation based on similar grievances stretching back to Mozambique’s colonial experience, the 1964–74 war of independence and the 1976–92 civil war. Over the decades, meanwhile, national and international interventions were viewed as having ignored – and at times exacerbated – these conditions, with calls made for more transparent, participatory and decentralised governance in the future.

These findings paint the picture of a complex and deeply rooted system of local-level insecurity, raising important considerations for policymakers and practitioners. Local-level perceptions of the gaps in existing national and international ameliorative programming offer further valuable insights for those looking to support more holistic security interventions. In particular, a marked sense of local-level exclusion from the formulation of external interventions was apparent, with this phenomenon held to limit the effectiveness of ground- level programming.

Drawing on the findings of this research, this report offers a series of recommendations for national and international programming, targeting key aspects of the juxtaposition between existing approaches and the experiences of communities on the ground.

Recommendations

• External programming must be designed, adapted and validated based on meaningful local-level consultation that internalises the interests and lived experience of community actors across Mozambique.

  Bottom-up engagement is crucial to the successful design and ground-level implementation of a range of security interventions. In planning such engagement, international stakeholders must account for a widespread lack of trust in past and present programming at the grassroots level. They should therefore work with national authorities where possible to map and learn from past engagement efforts, whether in relation to the insurgency in Cabo Delgado or wider insecurity experienced across the country. This must be done while ensuring adequate tailoring of consultation processes at the programme design phase while being mindful of the need to disaggregate specific approaches to account for the variety of conditions and experiences of lived insecurity at the subnational level.

• Efforts to professionalise law enforcement agencies must have at their core a focus on improving relations with the populations they are mandated to serve.

  In the face of diminished trust in state security and law enforcement agencies across Mozambique, the provision of external programming must be made contingent on national-level engagement in the professionalisation of key agencies into service-oriented institutions that protect all citizens. As a key part of this, interventions to address security force culture, management and accountability structures must be prioritised, with particular corruption risks and vulnerabilities identified and addressed. In parallel, local, national and international drivers of corruption within these institutions must be analysed and addressed as part of a broader approach focused on improving communication and trust between the population and the state.

• Focused support should be provided to Mozambican civil society groups and NGOs performing potentially crucial roles in monitoring crime prevention and strengthening resilience.

  Programming in collaboration with Mozambican civil society stakeholders should be prioritised, building the capacity of specific actors and allowing them to use the additional capability to solve problems on local terms, in response to ground-level priorities. This should be done while working to ensure a national-level environment that respects civil society rights and freedoms, facilitates access to relevant information and guarantees the ability of third-sector stakeholders in Mozambique to work independently across monitoring, reporting and conflict-resolution functions, including in relation to the insurgency in the north. Civil society should also be engaged wherever possible in crime and violence prevention debates and interventions, giving voice to local-level expertise and experience, and engaging, as appropriate, with community-level policing and security.

• Technical assistance programming across Mozambique must take politics seriously, with interventions tailored to the political economies of key sectors and institutions at the local and national levels.

  Interventions aimed at improving governance across the Mozambican security sector must be informed by in-depth political economy analysis, dissecting interest and power structures to identify agents with (and without) incentives to support sustainable reform on the ground. This analysis should extend beyond the central Mozambican policy establishment and consider specific contextual and institutional factors at the district and provincial levels – accounting, for example, for potential discontent at this level over the instrumentalisation of the security apparatus by elite actors, and over the conditions experienced in some cases by lower-ranking members of the police and military. Policy and programming should be formulated accordingly, drawing on real opportunities to influence a more publicly oriented form of security governance, and ensuring the feasibility of planned interventions in light of vested interests across the political and security spheres.

• International programming must better account for the role of some state actors in the production and maintenance of insecurity in Mozambique, and as potential spoilers in reform efforts.

  A central theme across the research for this report concerns the extent of perceived official complicity in criminality and violence and the fluidity between licit and illicit activity. There is an urgent need for reflection among donors on how programming is undertaken considering these conditions, and a need to be mindful of the long history of misuse of security and development funds in Mozambique. Alternative models of provision that account for the perceived role of some state actors and institutions not only as providers of security, but also as agents in the production of insecurity are required, based on in-depth political economy analysis across the Mozambican security architecture.

• A focus on the drivers of crime and violence should be prioritised, with programming moving beyond symptoms to address underlying structural dynamics.

  A prioritised research agenda should be supported to inform programming targeting the underlying drivers of insecurity in Mozambique – mindful of the ways in which these dynamics may vary at a district and provincial level. Based on this, investment to address economic inequality should be prioritised and delivered in a way that is locally-based and transparent, with a focus on training and capacity building, investment in skills and education, sustainable management of the natural resource sector as an avenue for local growth and opportunity, infrastructure development and enhanced access to finance. A specific youth focus should feature in such economic empowerment programmes, with specific vulnerabilities accounted for in delivery. This should come as part of a concerted focus on crime and violence prevention, engaging Mozambican civil society and private sector actors with the potential to contribute more actively.

• Opportunities should be explored to address the role of local government and support decentralised structures capable of allowing meaningful popular participation.

  A key focus of external stakeholders should be on identifying options to engage with elected local structures capable of enabling greater community participation in decision-making, mindful of the perceived need for more decentralised, accountable and participatory governance. Where appropriate, opportunities should be explored to deliver development assistance not only through central government, but also at a district and provincial level, based on in-depth risk assessment and analysis of the relevant political economies. Efforts to strengthen local-level participation could also see community-level civil society actors and leaders brought into governance roles across relevant initiatives.

• Regulation and oversight of private security companies should be strengthened to mitigate risks of human rights abuses.

  Monitoring of the activities of private security providers should be bolstered, with interventions prioritised to bolster inspection and regulatory capacity. This should come as part of a strengthened focus on the terms of engagement with the private sector operating in Mozambique more broadly. While the Mozambican government has indicated its intention to draw on frameworks such as the Voluntary Principles on Security and Human Rights to guide security relationships between extractive companies, state and non-state security actors, and communities – and although a number of companies operating in Mozambique use these principles – gaps remain across the extractive sector, with a wider government-led effort required to improve the human rights environment.

Cathy Haenlein is Director of Organised Crime and Policing Studies at RUSI. Her expertise lies in serious and organised crime, illicit trade, conflict and development. Cathy has a particular focus on transnational environmental crime and security, with regional expertise in East and southern Africa.

Joana de Deus Pereira is a Senior Research Fellow at RUSI Europe specialising in counterterrorism and preventing/countering violent extremism (P/CVE). She has particular expertise in the realms of geopolitical and country-risk analysis in sub-Saharan Africa. Joana has been involved in research and project management for a large number of EU-funded projects. Joana has also been engaged as an EU project evaluator for over three years and has deep experience in risk and security analysis with a focus on counterterrorism and P/CVE issues.

Michael Jones is a Senior Research Fellow in RUSI’s Terrorism and Conflict team, focusing on political violence, non-state armed group governance, and conflict ecosystems across Sub-Saharan Africa. He has led fieldwork investigative fieldwork across numerous regions, was accredited as a member of DFID/FCDO’s conflict advisor cadre in 2020, and has worked as a consultant analysing peacekeeping in Somalia and corruption in Sudan.

Liam O’Shea is a former RUSI Senior Research Fellow focused on police integrity, violence and corruption, security sector reform and politics and security in the former Soviet Union. His research examines how to address prominent problems in Western policing, including police brutality, racism and misogyny.