*** 3/12/26 - Info re Insight Crime (think tank/media organization that seeks to inform the debate about organized crime + citizen security in the Americas) + Criminal Violence, the State, and Society from 2025 Annual Review of Political Science
Buzz Sawyer
(2) Criminal Violence, the State, and Society
Omar García-Ponce1
Review Article Open Access
The study of criminal violence has received increasing attention in political science over the past 15 years, as organized criminal groups have grown and diversified worldwide, unleashing unprecedented waves of violence. This article presents a critical assessment of the current state of political science scholarship on criminal violence. It discusses the sources and dynamics of organized criminal violence, emphasizing the reconceptualization of state–criminal group relationships in the literature, shifts in illegal markets, and the political incentives fueling criminal wars. It also examines how states and societies respond to criminal violence. State responses include punitive approaches, institutional reform, and community-based interventions, while societal responses can be examined through the lenses of exit (e.g., migration, disengagement), voice (e.g., political participation, collective resistance), and loyalty (i.e., compliance with state authorities or criminal groups). The article also addresses conceptual and methodological challenges, policy implications, and ethical considerations inherent in this field of study and identifies promising pathways for future research.
KeywordsOrganized criminal violence is a global security threat. From drug-dealing street gangs to transnational mafia-type organizations, organized criminal groups (OCGs) have grown and diversified around the world, reaching historically high levels of lethal violence in recent years (Davies et al. 2024). The United Nations Office on Drugs and Crime estimates that intentional homicides—primarily caused by OCGs in low- and middle-income countries—accounted for approximately five times as many deaths as state-based armed conflicts and 20 times as many deaths as terrorism between 2019 and 2021 (UNODC 2023). This form of organized violence differs from conventional types of conflict typically studied in political science, such as civil war, state repression, and political violence in general. It is carried out by nonstate actors involved in illicit activities or illegal markets and is mainly driven by economic profit rather than political or ideological goals. However, as Durán-Martínez (2023) points out, the diverse manifestations of criminal violence are not completely apolitical, and the line between criminal and political violence is often blurred.
Large-scale criminal violence not only exposes the inability of the state to retain the monopoly of violence and maximize social welfare but also reveals a complex network of interactions between OCGs and state and societal actors (Arias 2017, Durán-Martínez 2015, Yashar 2018). OCGs tend to be deeply embedded in society, infiltrate political institutions, and shape local power dynamics, redefining the social and political order (Lessing 2017, Trejo & Ley 2020). Despite the clear political dimensions of criminal violence and the security and governance challenges it poses to citizens and governments—especially in countries of the global South—its systematic study within political science is relatively recent. Organized violence lacking ideological or broader political goals was not considered a fundamentally political phenomenon, leading to its relative neglect among political scientists (Barnes 2017). It was not until the early 2010s that research on criminal violence or organized crime began to receive significant attention in the discipline.
Over the past decade and a half, political science research has advanced our understanding of the causes and consequences of criminal violence, often bringing insights from the civil wars literature (Kalyvas 2015) and contributing new concepts, theories, and methodological approaches. This growing academic interest is reflected in publication trends. As shown in Figure 1 , the number of research articles explicitly examining criminal violence or organized crime in the top political science and comparative politics journals has increased dramatically, from only three articles published in 2010 to 27 in 2023. During the same period, the share of articles on criminal violence in these journals grew from less than 1% to nearly 6%. Similar trends are observed in leading journals from other subfields and specialized outlets. This expanding body of research is methodologically diverse and incorporates interdisciplinary perspectives drawing from criminology, psychology, sociology, and economics. The scope of these studies includes a diverse set of substantive topics, broadly organized into three main areas of focus: (a) the origins and dynamics of criminal violence, (b) the strategies states employ to combat it, and (c) the societal impacts and responses it generates.
Research on Latin America dominates the field, a trend that is expected given the region's status as the most violent in the world, as measured by homicide rates (UNODC 2023). Homicidal violence caused by OCGs has afflicted Latin America and the Caribbean for decades (Rivera 2016). Some of the most extreme cases include Brazil, Colombia, Jamaica, Mexico, and Venezuela, but in recent years, several other countries in the region have experienced a sharp escalation in criminal violence. Ecuador, for example, has seen its homicide rate increase sevenfold over the past five years, reaching 40 homicides per 100,000 people in 2023 (Freeman 2024). At the same time, the region is also home to some of the most dramatic short-term security improvements. El Salvador's experience under President Nayib Bukele illustrates the risks of sacrificing democratic rights in pursuit of immediate safety.
While the presence of resilient OCGs may seem unsurprising in low-capacity states, various forms of organized criminal violence persist even in industrialized societies with democratic institutions and robust legal systems. Prominent examples include Italy and the United States (Gambetta 1993, Skaperdas 2001). Paradoxically, OCGs can emerge, adapt, and thrive under democratic rule. Drug cartels in Mexico and Colombia, much like the American and Italian mafias, flourished within the boundaries of democracy. Recent research has deepened our understanding of this phenomenon, aiming to develop effective strategies to combat the pervasive influence of criminal organizations and mitigate their harmful effects on society.
In this article, I discuss the current state of political science research on criminal violence. The scholarly work across the three key areas outlined above offers novel theoretical and empirical contributions that challenge conventional wisdom. First, understanding the sources and dynamics of organized criminal violence requires reconceptualizing the relationships between states and criminal organizations, which often coexist not as substitutes but as complements (Dipoppa 2025, Lessing 2024). Furthermore, extensive research demonstrates that economic and political incentives jointly influence criminal organizations’ strategic behavior and use of violence. Second, state responses to criminal violence vary considerably across contexts. Although punitive approaches remain predominant, they have produced largely ineffective or mixed results. Emerging scholarship highlights the potential of institutional reforms and community-based interventions, although rigorous evidence on their effectiveness remains limited. Third, the diverse ways society is impacted by and responds to criminal violence can be examined through the classical framework of exit, voice, and loyalty (Hirschman 1970). For instance, exit manifests through migration, political disengagement, or withdrawal from public spaces; voice includes political participation, protest, and collective resistance; and loyalty can take the form of compliance with either state authorities or criminal groups. I close the article with a discussion on the limitations, policy implications, and ethical considerations of this research, while identifying promising directions for future work.
Organized criminal violence represents a state's failure to monopolize the legitimate use of violence in the Weberian sense (Weber 1946). From this perspective, it is often assumed that the organizations behind such violence operate independently of the state, forming parallel coercive structures. Recent scholarship challenges this view and sheds light on the symbiosis between the state and the criminal underworld across illegal markets, from drug and human trafficking to extortion and illicit crop cultivation, offering a more nuanced understanding of the sources and dynamics of criminal violence. In this section, I discuss three key aspects: state–criminal group relationships, the organization of illegal markets, and the political logic of criminal violence.
Since the seminal work by Tilly (1985), several scholars have examined the parallels between state-making and organized crime (Koivu 2016). A prominent view in the literature is that states with weak institutions or dysfunctional legal systems generate a demand for private protection, making them more likely to have criminal organizations that engage in violence (Frye & Zhuravskaya 2000, Hay & Shleifer 1998, Varese 2001). The basic logic is that when states fail to provide basic services such as security, contract enforcement, and conflict resolution mechanisms, criminal organizations step in to fill the power vacuum with structures of coercion that produce violence.
The proliferation of criminal organizations and their violent activities is largely facilitated by weak state institutions, widespread corruption, and ineffective law enforcement (Snyder & Duran-Martinez 2009). These conditions create fertile ground for the establishment of criminal governance systems, where gangs, drug cartels, or protection rackets exert control over specific territories and populations. While particularly prevalent in urban areas of the developing world—especially in Latin America and the Caribbean, as illustrated by Arias (2017)—such criminal enclaves can also be found in certain neighborhoods of developed nations with strong institutions, such as Italy (Varese 2006), the United Kingdom (Campana & Varese 2018), and the United States (Howell 2015).
Although high levels of criminal violence and impunity indicate that the state has failed to maintain its monopoly on violence, it is important to emphasize that state presence often persists in meaningful ways and coexists with varied forms of criminal rule. Fiscal capacities and electoral democracy, for example, may continue to function, albeit under the shadow of corruption and criminal violence (Arias 2009). The prevalence of criminal rule in many of these societies gives place to duopolies of violence (Lessing 2024, Skaperdas & Syropoulos 1996). In most cases, the state and OCGs are not substitutes but rather strategic complements, giving place to symbiotic relationships between criminal and state governance (Arias 2017, Blattman et al. 2024, Lessing 2021, Magaloni et al. 2020a).
The phenomenon of criminal governance can be situated within the broader literature on nonstate armed governance, which includes studies of paramilitary groups, rebel organizations, and other nonstate actors that use violence to establish systems of rule in areas under their control (Acemoglu et al. 2013, Arjona 2016, Mampilly 2012, Stewart 2018). There are parallels between the governance systems created by rebels and OCGs. However, key distinctions exist. Rebel governance is typically understood as a state-building strategy to govern the territory controlled by insurgents (Florea & Malejacq 2024, Kalyvas 2006, Weinstein 2006). In contrast, following Lessing (2021), criminal governance tends to be more subtle. OCGs may exert influence without explicit territorial control and adapt to various economic and political conditions in both rural and urban settings. As Lessing (2021, p. 862) argues, “[Criminal organizations] vary widely in their organizational structures, but as with all organizations, these can be arrayed along a Weberian spectrum from personalistic-charismatic to rational-bureaucratic.”
Magaloni et al. (2020a) develop a typology of criminal governance by examining in depth the case of Rio de Janeiro. The authors categorize five types of criminal regimes based on whether criminal groups confront or collude with state actors, abuse or cooperate with the community, and hold a monopoly or contest territory with rival groups. This typology underscores the complex nature of organizations of the criminal underworld and the diverse strategies employed by OCGs to maintain control and influence. Furthermore, it helps us understand the heterogeneous impacts of public security and crime prevention interventions, which may exacerbate violence in certain territories but improve security conditions in others, depending on the type of criminal regime.
Prison gangs introduce an additional layer of complexity. As shown by Lessing & Willis (2019), prison gangs in Brazil have developed sophisticated bureaucratic procedures to manage their drug business and internal discipline, orchestrating violence and exerting social control beyond prison walls. Similarly, Skarbek (2011) provides evidence that within the US context, the narcotics trade in Los Angeles flourished through a system of governance that emerged within the confines of the penal system, where prison gangs have incentives to provide mechanisms of governance—such as enforcing deals or adjudicating disputes—to mitigate market failures among drug-dealing street gangs. Comparing cases from Brazil, Bolivia, England, Scandinavia, and California, Skarbek (2016) finds a similar pattern: Prison gangs tend to create extralegal governance institutions when formal structures prove inadequate. Such governance systems cannot emerge without the direct or indirect cooperation of state authorities. Thus, paradoxically, state–criminal group relationships also unfold within the very institutions designed to punish criminals.
Illegal markets, such as those for drugs, arms, and human trafficking, provide the financial backbone for many criminal organizations. Specific markets, like the drug trade, experience systemic violence (Reuter 2009). Limited enforcement and high profit margins are frequently regarded as sources of violent confrontations in these markets. For instance, rival criminal organizations compete for control over drug production and distribution in poorly regulated markets without legal mechanisms to resolve disputes. However, illegality does not necessarily breed violence (Snyder & Duran-Martinez 2009). While some illegal markets are highly violent, others operate with relatively low levels of violence.
Prohibition is a crucial factor in understanding organized criminal violence. The transition of a market from legal to illegal status fundamentally alters its dynamics, potentially leading to increased violence (García-Jimeno 2016). This phenomenon is exemplified by the Prohibition era in the United States (García-Jimeno 2016, Owens 2014) and contemporary drug prohibition policies (Miron 2013). There is also evidence that the violent consequences of prohibition extend beyond these well-known cases. This includes illegal trades in timber, ivory, gemstones, and other commodities. For instance, Chimeli & Soares (2017) document a surge in criminal violence following the prohibition of mahogany trade in Brazil.
A number of theoretical and empirical studies have challenged the common misconception that reducing supply in illegal markets can curb violence. In fact, supply reduction approaches to combat illegal markets may exacerbate violence. Castillo & Kronick (2020) show that government interdiction efforts that reduce drug supply can increase drug cartels’ aggregate profits when demand is sufficiently inelastic. Higher profits, in turn, raise the stakes of territorial control and incentivize violence, as has been empirically demonstrated by Castillo et al. (2020). Furthermore, the criminalization of these markets creates opportunities for corruption, undermining state institutions and fostering environments where violence can thrive unchecked.
More broadly, related work on the economics of crime acknowledges that malfeasance can reduce the effectiveness of law enforcement approaches (Becker & Stigler 1974, Bowles & Garoupa 1997, Kugler et al. 2005, Polinsky & Shavell 2001). For example, Kugler et al. (2005) analyze the role of corruption not only in diluting deterrence but also as a strategic complement to crime and a catalyst to organized crime. The authors present an oligopoly model in which differentiated criminal organizations globally compete in criminal activities and engage in local corruption to avoid punishment. According to their model, when bribing costs are low and the rents from criminal activity are sufficiently high, increasing policing and sanctions can generate higher crime rates.
In theory, the profitability of illegal markets can either fuel or mitigate violence, depending on various factors, such as market structure and law enforcement dynamics. In some cases, increased profitability may intensify competition for control over resources, leading to heightened violence. Conversely, when a single organization or a small group dominates a highly profitable illegal market, they may opt for less violent methods to maintain their position and avoid attracting attention. Whether higher profitability generates more violence remains an empirical question (Shirk & Wallman 2015).
Looking at the case of Venezuela, Kronick (2020) shows that the increase in profits from drug trafficking along major routes led to a surge in lethal violence as competing groups fought for control. Comparing the organization of illegal markets in Colombia and Mexico, Durán-Martínez (2015) finds that violence patterns are influenced by the degree of control drug trafficking organizations have over street-level dealers and local market monopolization. Independent dealers often result in sporadic and less organized violence. However, when drug trafficking organizations control dealers, significantly higher levels of violence are observed in competitive markets than under monopolistic conditions. These findings underscore how economic incentives and organizational structures within illegal markets can shape patterns of violence, particularly in the absence of effective state intervention.
A key aspect in the organization of illegal markets is the recruitment process, which often targets vulnerable populations by exploiting economic hardship and limited opportunities in legitimate sectors. Dube et al. (2016) show how the economic opportunities of rural workers in Mexico affect the dynamics of the drug trade. Exogenous shocks in the price of maize induce farmers to grow illicit crops, which then spurs violence as criminal organizations fight over drug production. Similarly, urban areas with high unemployment and poverty rates become prime locations for recruitment. Marginalization and income inequality can drive youths to perceive illegal markets as a path to socioeconomic mobility (Buonanno & Vargas 2019, Enamorado et al. 2016, Zepeda Gil 2024), foster gang socialization processes (Krakowski 2022, Rodgers 2017), increase the acceptability of criminal behaviors (García-Ponce et al. 2023a), and incentivize the formation of crime-specific skills that set individuals on criminal paths (Sviatschi 2022).
Recent research also highlights the remarkable adaptability of OCGs in sustaining and expanding their operations, both domestically and internationally, including their ability to penetrate legal markets. While the primary source of violence is, of course, the revenue generated from the illicit activities themselves—whether drug trafficking, extortion, or other criminal enterprises—criminal organizations often diversify their repertoire of criminal activities and income streams. Competition and state responses to organized crime can prompt OCGs to look for nontraditional sources of income. For example, some criminal organizations seek to control the production of legal agricultural commodities (Yoo 2022), mining exports (Herrera & Martinez-Alvarez 2022), or migrant exploitation (Dipoppa 2025).
Global economic processes have been identified as important factors shaping illegal markets and patterns of organized criminal violence. Andreas (2013) highlights that globalization and increased cross-border flows have facilitated the expansion of transnational criminal networks, complicating state efforts to combat criminal violence. Policy decisions in one country can have significant spillover effects in others. For instance, the expiration of the US assault weapons ban in 2004 resulted in greater gun supply in northern Mexico, leading to significant increases in criminal violence (Dube et al. 2013). Similarly, scarcity created by cocaine seizures in Colombia has led to increased violence in Mexico (Castillo et al. 2020). Furthermore, OCGs invest in legitimate businesses to launder money across national boundaries (Farfán-Méndez 2019, Levi & Soudijn 2020). Weak financial regulations and offshore banking systems facilitate the movement and concealment of illicit funds, providing criminal organizations with the financial resources to sustain and expand their operations. The role of money laundering is an important gap in the academic literature; further theoretical and empirical work is needed to understand how criminal organizations insert their profits into the financial system and the resulting impacts on violence levels.
Recent scholarship increasingly recognizes the critical role of politics and institutional arrangements in understanding patterns of criminal violence. Political conditions at all levels of government shape the dynamics of violence between rival criminal groups and their interactions with state security forces. Furthermore, even when OCGs lack explicit political agendas, ideology, or aspirations to overthrow the government, they exert substantial political influence. This influence manifests through various mechanisms, such as using violence against political actors and voters, funding candidates and electoral campaigns, and infiltrating political institutions.
Paradoxically, democratization processes have been linked to the escalation of criminal violence in several contemporary cases. Young democracies with entrenched patronage networks are particularly susceptible to social and political transformations that can disrupt existing local orders and destabilize mechanisms of social control. Some scholars argue that within the Latin American context, the public security and criminal justice reforms carried out during the political transitions shaped the ability of the new regimes to control the production of violence (Cruz 2011, Yashar 2018). However, if traditional power structures erode and state capacity is low, then social control mechanisms weaken and criminal violence increases. Notably, this phenomenon was observed and documented by Villarreal (2002) in rural Mexico during the period preceding the so-called drug war. This study provides valuable insights into how political transitions can inadvertently create conditions conducive to increased criminal behavior, highlighting the complex relationship between democratization, social structures, and criminal violence in developing democracies.
Trejo & Ley (2020) present a novel theoretical argument about the political logic of criminal violence in fragile democracies. They argue that growing electoral competition during the 1990s disrupted networks of protection and informal agreements between OCGs and local politicians in Mexico, pushing drug cartels to fight against each other and against state security forces for territorial control. More specifically, political alternation in state gubernatorial power undermined the informal networks of protection that had facilitated the cartels’ operations under one-party rule. Without protection, cartels created their own private militias to defend themselves from rival groups and from incoming opposition authorities. Thus, political competition during Mexico's democratization process undermined bargains between drug trafficking organizations and politicians, creating incentives to resort to violence.
Dube et al. (2013) also emphasize the role of electoral competition in shaping criminal violence dynamics. The authors demonstrate that the impact of US gun trafficking on violence in Mexico is conditioned by political factors associated with Mexico's democratic transition. Concretely, there was a disproportionate rise in killings in Mexican municipalities where local elections had become more competitive before 2004, particularly in regions with high levels of drug trafficking activities, indicating that increased electoral competition eroded informal agreements between drug cartels and local governments. Similarly, Rios (2015) provides empirical evidence suggesting that organized crime can be controlled more effectively when the same political party is in power at all levels of government.
Durán-Martínez's (2018) study of drug violence sheds light on the strategic use of violence by criminal groups. Examining the cases of Colombia and Mexico, the author argues that political fragmentation influences the visibility and frequency of violence employed by drug trafficking organizations. Criminal organizations adjust their violent tactics in response to shifting state responses toward criminal activities, challenging the notion that decreased high-profile violence indicates diminished cartel power. The strategic use of violence employed by OCGs includes public communication strategies, as shown by Johnson & Gillooly (2023) and Phillips & Ríos (2020). Public displays of extralethal violence (Fujii 2013) and publicized threats by criminal groups target specific audiences (either rival OCGs, the state, or society at large) and broadcast an imagined order of the criminal underworld.
The use of violence by criminal organizations against politicians and voters is also influenced by the political environment in which these groups operate. As Albarracín (2018) shows by exploring the case of Rio de Janeiro, criminal violence can be used to sustain and construct local political orders that undermine democratic processes. Blume (2022) argues that corruption, electoral competition, and the politicization of the security apparatus are the key factors determining the relationship between traffickers and the state. This relationship shapes the strategies employed by traffickers, leading to varying levels of violence. For instance, in Honduras, where co-optation strategies are prevalent, violence levels are strikingly high. In contrast, evasion strategies in Costa Rica resulted in moderate violence, while Nicaragua's collusion strategies produced, for some time, low levels of drug-related violence. These findings underline how different political contexts can significantly impact the intensity and nature of criminal violence.
Attacks on politicians and voters often follow strategic patterns designed to maximize criminal organizations’ influence and control. These groups typically target vulnerable subnational actors, especially during election cycles, to establish territorial dominance. The motivations behind such attacks vary, including retaliation for politicians’ acceptance or refusal of illicit funds and elimination of political rivals affiliated with competing criminal groups. This targeted violence serves as a tool for criminal organizations to influence political outcomes and safeguard their interests. For example, Trejo & Ley (2021) show that attacks on politicians and party candidates in Mexico between 2007 and 2012 were motivated by OCGs seeking to gain control over local governments, territories, and people. The attacks were strategically timed around local election cycles and targeted clusters of neighboring municipalities to strengthen their territorial control. The lack of protection from central powers, either political or military, makes subnational actors particularly vulnerable to these violent tactics. Blume (2017) also examines the reasons behind the assassinations of politicians in Mexico, identifying three primary motivations: accepting illicit funds, refusing to cooperate with criminal groups, or having political opponents connected to criminal organizations.
Beyond Latin America, Alesina et al. (2019) and Daniele & Dipoppa (2017) provide insights into how criminal organizations use pre- and postelectoral violence to influence electoral outcomes and politicians’ behavior in Italy. Alesina et al. (2019) show that the Sicilian Mafia strategically increases violence against politicians before elections, particularly when outcomes are uncertain, to reduce support for anti-Mafia parties. This preelectoral violence manipulates electoral results and curtails anti-Mafia efforts. Daniele & Dipoppa (2017) provide evidence that postelectoral violence is used to influence politicians from the start of their term, especially in regions with high organized crime presence. These strategies ensure that criminal organizations can maintain influence over local governments and protect their interests, illustrating the use of violence as a political instrument.
The strategies states adopt in response to criminal violence can set societies on divergent trajectories, shaping their long-term violence patterns. Governments and policymakers face a spectrum of approaches in their efforts to curb criminal violence. On one end, punitive measures emphasize aggressive tactics such as government crackdowns on criminal groups, kingpin strategies, militarizing public security, and intensifying drug interdiction efforts. In contrast, alternative strategies prioritize institutional reform or community-oriented solutions that focus on rebuilding trust, implementing community policing initiatives, enhancing education and employment opportunities, and supporting development programs that facilitate transitions to legal economic activities. The choice between these approaches—or the balance struck between them—significantly impacts security outcomes.
There is ample evidence that punitive approaches to reducing criminal violence, despite having popular support and significant human and financial resources, are largely ineffective and sometimes counterproductive when focused solely on deterrence, retribution, or incapacitation. Studies from various regions indicate that punitive measures often exacerbate violence and infringe on human rights, and, if they do manage to reduce crime, the success is usually short-lived and achieved at the cost of freedom and democracy. These findings suggest that such strategies not only fail to address the root causes of crime and violence but also can lead to greater societal harm in the long run.
Kingpin strategies, which target criminal group leadership, often result in unintended consequences. Castillo & Kronick (2020) argue that these strategies can make criminal bosses more shortsighted and prone to breaking peaceful agreements without significantly disrupting the illicit drug trade. Osorio (2015) shows that aggressive law enforcement tactics increase violence among criminal organizations in Mexico, particularly in areas with a high concentration of these groups. Similar findings for the case of Mexico have been documented by Dell (2015) and Calderón et al. (2015). Phillips (2015) presents more nuanced evidence, suggesting that while leadership removal may weaken criminal organizations temporarily, it leads to fragmentation and increased violence in the longer term.
In many Latin American countries, the militarization of public security is part of a new militarism that threatens the democratic rule of law (Ríos-Figueroa 2019). These militarized strategies to combat OCGs have produced particularly grim results. Magaloni et al. (2020a) and Willis (2015) find that in Rio de Janeiro and São Paulo, Brazil, militarized police tactics often escalated violence and eroded community trust. Blair & Weintraub (2023) experimentally evaluate a military policing intervention in Cali, Colombia, concluding that such intervention failed to reduce crime or improve human security. Flores-Macías & Zarkin (2021) examine the constabularization of the military in Latin America, arguing that this blurring of lines between civilian and military law enforcement has undermined citizen security, human rights, police reform, and the legal order.
More carefully tailored law enforcement strategies have shown promise in certain contexts. In this regard, recent research highlights the importance of conditional repressive strategies from the government to discipline criminal organizations’ use of violence (Castillo-Quintana 2025, Kronick 2020, Lessing 2017, Snyder & Duran-Martinez 2009). The logic of conditionality is that government crackdowns on criminal groups and their leaders can be conditioned on their behavior as retaliatory measures to misconduct. For example, Lessing (2017) argues that policies conditioning repression on cartel violence are more likely to effectively deter violence than brute-force repression. Trudeau (2022) demonstrates that reforms implementing internal, nonpolice oversight can be effective in reducing both police and civilian violence, particularly in areas with widespread organized crime. Cruz & Durán-Martínez (2016) examine the potential for negotiations between or with criminal organizations to reduce violence under certain conditions. However, as Castillo-Quintana (2025) illustrates, only far-sighted, patient governments can commit to conditional policies, and the violence-reducing effects of conditional policies seem implausible in most policy scenarios.
Further research is needed for a more comprehensive understanding of the unintended consequences of these law enforcement approaches to combat criminal violence. The negative consequences of punitive policies can be amplified by patronage politics. For instance, Ravanilla et al. (2022) found that in the Philippines, mayors excluded from establishment patronage networks executed President Rodrigo Duterte's drug war more aggressively, illustrating a trade-off between traditional patronage politics and violent democratic backsliding. Likewise, Cruz (2016) shows how state agents contribute to the escalation of criminal violence through various mechanisms in Latin America. Moreover, research by Gentil-Fernandes et al. (2022) and Flores-Macías (2018) on the impacts of militarized policing on local economies and fiscal capacity underscores the need to evaluate the externalities of punitive approaches.
Institutional reform is at the heart of reimagining how states address criminal violence, particularly in contexts of weak rule of law and widespread corruption. Reforming institutions such as police forces, judicial systems, and penal institutions has the potential to enhance state legitimacy, rebuild trust between the state and its citizens, and ultimately reduce criminal violence. Many of these reforms aim to replace or soften mano dura policies. However, we still lack a clear understanding of what works, what does not, and why certain reforms succeed in some contexts but fail in others.
Several studies highlight the challenges and trade-offs of institutional reform. Police reform, in particular, encounters resistance from entrenched interests and organizational inertia as well as resource constraints and the persistence of informal practices. González (2023) finds that operational reforms such as body-worn cameras and community policing may improve short-term performance but can paradoxically increase police autonomy to resist more ambitious structural reforms, compromising long-term violence reduction efforts. The electoral disadvantages associated with police reform further complicate its implementation. As González (2019) notes, enhancing state capacity in security provision, while urgently needed, often conflicts with political incentives. Leaders may shy away from comprehensive reforms due to their potential unpopularity or the lengthy timelines required to achieve visible results.
The literature has also explored judicial and penal reform strategies, with an emphasis on initiatives that seek to reduce impunity and improve criminal justice system efficiency. Brinks (2008) and Helmke & Ríos-Figueroa (2011) emphasize the need to confront the systemic inequalities and procedural weaknesses that perpetuate violence. For example, abolishing inquisitorial criminal justice institutions can help restrain police brutality in some contexts, as demonstrated by Magaloni & Rodriguez (2020) in the case of Mexico. Procedural safeguards are essential for effectively combating criminal violence. As Huebert & Brown (2019) argue, how states fight crime significantly influences their success. Due process not only fosters legitimacy but also enhances the deterrent effect of a well-functioning criminal justice system.
Thus, despite implementation challenges, public security and criminal justice reforms seem to be a promising path to move forward. When well-designed and implemented inclusively, such reforms can help affirm citizens’ rights and build or restore trust between the public and state institutions. Cruz (2011) shows how public security reforms during critical political periods help explain why some Central American countries have controlled violence more effectively than others. Similarly, recent work by Alcocer (2024) provides important evidence that police reforms fostering intergovernmental coordination can weaken criminal organizations and reduce violent crime, even if violence may increase in the short term during reform implementation.
Effective institutional reform requires a comprehensive approach that extends beyond the immediate security sector. This includes revising constitutional frameworks, establishing accountability and civilian oversight mechanisms, promoting human rights, and addressing the structural conditions that allow not only criminal organizations but also state actors to become perpetrators of violence (Pellegrino 2024). Central to these efforts is enhancing the legitimacy and integrity of state institutions.
Community-based approaches to criminal violence have emerged as promising alternatives or complements to traditional law enforcement policies. The menu of interventions includes a diverse range of strategies that aim to strengthen state–society relations, prevent individuals from engaging in criminal and violent behavior, promote reintegration into society, and enhance the overall effectiveness of public safety efforts (Beaudry et al. 2021, Farrington & Welsh 2008, Kondo et al. 2018, Magaloni & Diaz-Cayeros 2018). Rather than rely solely on arrests and incarceration, these approaches recognize that sustainable violence reduction requires engaging communities as partners in creating safer neighborhoods.
One widely discussed strategy is community policing, which emphasizes proactive engagement and partnership between police and local residents. This approach has been particularly popular in Latin America (Arias & Ungar 2009), where impunity rates are extremely high and states often struggle to establish legitimacy in marginalized communities. However, recent evidence from field experiments conducted in several low- and middle-income countries challenges the effectiveness of community-oriented policing, specifically when implemented in isolation. For example, Blair et al. (2021) conducted six coordinated field experiments in Brazil, Colombia, Liberia, Pakistan, the Philippines, and Uganda, collaborating with local police to increase community policing practices, and found that community policing initiatives did not significantly improve citizen–police relations or reduce crime. These findings underscore the need to integrate such strategies with broader institutional reforms and interventions to address deeper systemic issues.
Community-based approaches also place emphasis on addressing the socioeconomic drivers of criminal violence. Youth violence prevention programs are a key component of these initiatives. Evidence from field experiments in Liberia suggests that a combination of employment programs and cash transfers (Blattman & Annan 2016) or cognitive behavioral therapy and cash transfers (Blattman et al. 2017) can reduce individual engagement in crime and violence. Furthermore, early interventions targeting potential recruits to criminal organizations—through vocational training, education, and public health programs—can help disrupt pathways into delinquency and the labor supply that sustains illegal markets. Such interventions must be holistic, combining economic, social, and psychological support to tackle the root causes of violence comprehensively. Addressing both individual behaviors and the structural conditions that fuel criminal violence is essential for long-term success (Magaloni & Diaz-Cayeros 2018).
Examining patterns of criminal violence from a public health perspective has gained notoriety among academics, practitioners, and policymakers, emphasizing prevention and intervention rather than traditional enforcement. For instance, the Cure Violence model in the United States was presented as a useful framework for treating crime and violence as preventable diseases. These interventions have yielded mixed results (Butts et al. 2015). Existing research provides promising evidence that community response approaches to mental health and substance abuse help reduce crime (Dee & Pyne 2022) and inconclusive findings regarding the causal impact of psychological interventions in prison on recidivism (Beaudry et al. 2021).
Overall, community-based approaches to criminal violence represent a critical shift in how states fight crime and insecurity. By focusing on prevention, social cohesion, and cooperation between institutions and community members, these interventions have the potential to not only reduce violence but also rebuild the social fabric and restore confidence in criminal justice institutions. Their success, however, depends on integrating them with broader efforts to address the socioeconomic roots of violence. Limited resources, corruption, resistance from powerful interest groups, and the lack of political will often hinder their effectiveness and scalability. Furthermore, their success varies across contexts, making it crucial to adapt these strategies to local conditions.
A recent study conducted in Medellín by Blattman et al. (2022) offers important insights into the effectiveness of noncoercive, service-oriented approaches to state-building in high-crime areas. This intervention deployed community liaisons to help address public order issues, improve communication with government agencies, and connect residents to social services. The impacts on security and legitimacy diverged across neighborhoods by initial state presence, underscoring the importance of tailoring interventions to local conditions and revealing potential challenges and unintended consequences of expanding state presence in poorly governed areas.
As evidence grows, it becomes increasingly clear that the long-term success of interventions against criminal violence hinges on coordinated efforts between governments, civil society, and international organizations in some cases. An obvious yet often overlooked dimension of this fight is the necessity of reducing the size of illegal markets. Reducing the demand for goods and services provided by criminal organizations is arguably the most decisive step toward sustainable reductions in criminal violence. How states can effectively shrink illegal markets to undermine criminal organizations remains a subject of ongoing research and debate.
Societal responses to criminal violence can take many forms, as evidenced by recent political science research. Drawing on Hirschman's (1970) seminal framework, these responses can be conceptualized through the lenses of exit, voice, and loyalty. Exit, for example, could entail migration, political disengagement, or complete withdrawal from public spaces. Voice may include increased political participation, protest, or collective resistance against violence. Loyalty refers to strategic compliance with either state authorities or criminal organizations.
In high-crime contexts, citizens can typically opt to exit in two main ways: migrating (either internally or internationally) to safer areas or withdrawing from politics and public life more generally. Although there is a vast literature on migration and displacement in traditional conflict zones, such as areas affected by civil wars or interstate conflicts, scholarly work on migration driven by criminal violence remains scarce. Notable contributions include Atuesta & Paredes's (2015) research, which uses census and survey data to document internal migration related to criminal violence in Mexico. They show that individuals fleeing criminal violence are willing to sacrifice economic opportunities for safety. According to their estimates, migrants moving to violent states require a 25% salary increase to raise migration odds by 10%, whereas those moving to nonviolent states require only a 15% salary increase for the same effect. Similarly, drawing on intensive participant observation and interviews in Colombia, Marston (2020) argues that the urban residents most likely to remain despite being threatened with displacement by gang violence are the well-connected, that is, those who can leverage ties to a community figure or a member of the criminal group to stay.
The impact of violence on migration patterns extends to child migrants as well. Clemens (2021) examines the relationship between violent crime and child migration from Honduras, El Salvador, and Guatemala to the United States. Based on a study of 178,825 unaccompanied child migrants apprehended in the United States from 2011 to 2016, the author reveals that in an average municipality, 10 additional homicides led to 3–6 additional cumulative child migrant apprehensions in the United States. Furthermore, this important study finds that migration experience diffusion through peer and family networks creates self-reinforcing, city-specific waves of migration that can persist even after violence subsides.
It is important to note that individuals escaping criminal violence in Latin America and other regions frequently encounter restricted options for protection. Albuja (2014) contends that this situation arises primarily due to the infrequent characterization of criminal violence as a humanitarian crisis and the reluctance of affected nations to adopt protection and humanitarian assistance strategies. Consequently, the responses usually emphasize punitive measures aimed at sanctioning offenders, diminishing crime rates, and reinstating public order rather than extending support to victims.
Scholars have also examined the conditions under which criminal violence deters individuals from engaging in the political life of their communities. Córdova (2019) explains how gang dominance limits the ability of citizens to engage politically and effectively resist gang violence in El Salvador. In neighborhoods with a strong presence of organized crime, gangs are incentivized to deter, monitor, and punish electoral and community participation, particularly that which involves close contact with political operatives. Trelles & Carreras (2012) examine municipal-level crime and turnout rates in Mexico and contend that individuals in municipalities with higher levels of crime have increased levels of fear and political dissatisfaction, resulting in significant declines in electoral participation. Similarly, Ley (2018) shows that voters can be significantly demobilized by the strategic use of violence by OCGs during elections. High-profile attacks induce anxiety among voters, reduce their perceived benefits and satisfaction from voting, and increase the costs and risks of voting, all resulting in decreased electoral participation.
Particularly concerning is the mounting evidence that exposure to criminal violence and insecurity at both the individual and community levels can erode political trust and support for democracy (Blanco & Ruiz 2013, García-Ponce & Laterzo 2023, Malone 2010). Crime victimization has been shown to decrease overall horizontal trust in one's community, family, and friends (Corbacho et al. 2015). Preexisting social inequality can further exacerbate vulnerability to violence (Barham 2024), and the erosion of social cohesion in combination with increased political mistrust in such contexts can translate into diminished faith in the importance and efficacy of democracy and civic engagement. When governments and institutions repeatedly fail to provide safety and security and the rule of law is weak or inconsistently applied, formal political institutions can be seen as ineffective and unworthy of trust or participation.
Existing work highlights a diverse spectrum of strategies to cope and adapt to criminal violence. Using survey data from five continents, Bateson (2012) provides evidence of a positive impact of crime victimization on political participation. Rather than becoming withdrawn or disempowered, crime victims tend to become more engaged in civic and political life. This echoes findings from a number of studies that have found that exposure to civil war violence tends to increase political participation and prosocial behavior (Bauer et al. 2016). However, as noted above, recent academic research contests this perspective by uncovering considerable variation in individual responses to criminal violence. For example, Skigin (2024) finds that narratives characterizing people as responsible for their misfortune reduce prosocial behavior, but citizens are more likely to aid victims perceived to have little control over their situation, suggesting that elite and media discourses crucially shape public responses to violence.
A growing strand of literature sheds light on how communities mobilize to resist criminal violence and demand state protection. Ley et al. (2019) show that indigenous communities in Mexico have leveraged their autonomous governance institutions to resist narco rule. They argue that a history of social mobilization, translocal networks, and regional ethnic autonomy help explain successful resistance. Similarly, Osorio et al. (2021) find that historical experiences of armed resistance make communities more likely to mobilize against contemporary criminal violence. Dorff (2017) contends that social connectivity is crucial for community resilience when confronting violence, since it can increase a sense of purpose among victims, incentivizing them to participate more actively in the political life of their communities.
Community mobilization varies widely across different contexts, leading to varying degrees of effectiveness. Moncada (2021) examines collective resistance to criminal extortion in Colombia, El Salvador, and Mexico, showing how the success of these efforts depends on the underlying social and economic conditions. In some cases, resistance can cause a violent backlash from criminal groups, creating cycles of retributive violence. For example, García-Ponce et al. (2023b) show that anger induced by exposure to criminal violence increases the demand for retribution and punitive justice, including vigilantism. When criminal violence and impunity are high, outrageous and persistent forms of violence against civilians can lead individuals to prioritize harsh punishments carried out by community members, such as lynchings or vigilante actions, even at the expense of the rule of law or possible retaliation (Baron et al. 2024, Bateson 2021, Jung & Cohen 2020, Zizumbo-Colunga 2017).
Several scholars have turned their attention to the politics of vigilantism (Bateson 2021), highlighting that extralegal responses to crime are driven by enabling factors such as inequality (Phillips 2017) or denser community organization that facilitates collective action (Berg & Carranza 2018). Recent micro-level research on the logic of vigilante violence emphasizes the role of moral reasoning in responses to crime. As Nussio (2023) explains, moral beliefs can both restrain and motivate collective violence. In this regard, Baron et al. (2024) find that two types of moral reasoning are more likely to be used to justify pro-vigilante actions: consequentialist reasoning (weighing costs and benefits) and dehumanization of accused criminals. Interestingly, deontological justifications about the right or just action are commonly used but less strongly linked to support for punitive violence. These findings suggest that—regardless of the legality and long-term effectiveness of vigilantism and other forms of collective violence—citizens are not only motivated by a desire for retribution but also take into account cost–benefit considerations when deciding to take the law into their own hands.
The literature has identified various forms of loyalty as responses to criminal violence, including cooperation with state authorities and adaptation to criminal governance. Several empirical studies have examined the factors that influence citizens’ willingness to cooperate with law enforcement in high-crime contexts (Córdova & Kras 2020, Haim et al. 2021, Nanes et al. 2023, Tankebe 2009). This work underscores that effective citizen–state cooperation depends not merely on security outcomes, but on whether enforcement institutions are viewed as fair, responsive, and worthy of trust—even amid persistent threats from criminal actors.
In some contexts, communities may cooperate or coexist with criminal groups as a survival strategy. For instance, drawing on ethnographic fieldwork in Costa Rica, Nicaragua, and Honduras, Blume (2021) argues that rural and marginalized communities with widespread corruption and limited state capacity are more inclined to be supportive of and form collaborative relationships with criminal groups. Furthermore, traffickers who are native to the community, limit their use of violence, and invest economically in the community are the most likely to gain popular support.
Using survey data and list experiments to elicit truthful responses about potentially illegal behavior, Magaloni et al. (2020b) shed light on why criminal groups sometimes provide assistance to communities. Their findings suggest that collaborative relationships between criminals and community members depend on the extent to which the criminal group maintains control over a specific territory. Collectively, this body of research offers novel insights into the complex dynamics of criminal governance and community interactions in areas where state authority is limited or contested.
Citizens affected by criminal violence also express their loyalty through preferences for justice and criminal punishment. Militarized policing and extralegal harsh punishments by state authorities, such as torture or extrajudicial killings, are often popular when criminal violence and impunity are high (Flores-Macías & Zarkin 2022, García-Ponce et al. 2023b, Laterzo 2024). There is evidence that individuals are willing to make trade-offs between security, corruption, human rights, and the rule of law in violent contexts. For instance, García-Ponce et al. (2021) show that individuals who have been victims of crime in Mexico are more likely to report both higher levels of fear and willingness to accept corruption if it lowers violence.
Visconti (2020) provides evidence from 18 Latin American countries indicating that crime victims are more likely to support mano dura or iron-fist policies to reduce crime, such as state repression. Affected citizens value democracy less, which might explain their willingness to accept the erosion of human rights in favor of harsh measures to combat criminal violence. Looking at the cases of Argentina and Brazil, Laterzo (2024) provides experimental evidence that the effect of ideology on preferences for punitive candidates is moderated by three instrumental factors, including perceptions of insecurity, ineffective social policy, and gang-related crime.
Perhaps the most prominent contemporary example of loyalty to a regime in its fight against criminal violence is El Salvador. President Bukele's government achieved unprecedented success in reducing violent crime through repressive measures, including suspending constitutional rights, undermining democratic norms, and conducting mass arrests that significantly curtailed gang activity. While domestically popular, as Meléndez-Sánchez & Vergara (2024) highlight, similar crackdowns in other countries have largely failed. Moreover, the success of Bukele's model relies heavily on eroding democratic principles, prompting serious concerns about its long-term implications for governance and the rule of law.
The study of criminal violence within political science has produced a vast literature in a relatively short period of time, yet several critical areas remain understudied, presenting opportunities for future research. While much of the existing work focuses on Latin America, Italy, and the United States, there is a pressing need to broaden the geographic coverage of this research. Many countries in the global South, which face unique challenges related to organized crime and violence, have been largely neglected. Additionally, expanding the substantive scope to include criminal activities in illicit markets beyond drug trafficking and protection rackets—such as human trafficking networks or organized environmental crime—would contribute to a more comprehensive understanding of how criminal violence operates and evolves across diverse political, economic, and cultural contexts.
The cross-fertilization between civil war studies and criminal violence research has fundamentally reshaped our understanding of the porous boundaries separating criminal and political violence. An important area for future inquiry in this regard is examining the processes through which former rebel and insurgent organizations transform into OCGs. While some studies have explored this phenomenon (Daly et al. 2020, Vásquez-Cortés 2024), more systematic analyses are needed to uncover the factors driving these processes of organizational adaptation and transformation. Moreover, research on how these transitions from rebel to criminal groups impact conflict resolution and postconflict governance could provide valuable insights for policymakers grappling with the challenges of rebuilding communities after civil war and reintegrating former combatants into society.
Two interrelated dynamics of organized criminal violence demand further scholarly attention. First, more research is needed to understand the processes of diversification and fragmentation within OCGs. Understanding why and how these groups expand into new criminal activities or splinter into smaller factions is crucial to inform more effective law enforcement strategies. Second, additional empirical research on how different types of victimization map onto various forms of collective resistance is necessary to design more effective interventions to curb violence and support victims. Several studies have explored community responses to criminal violence, but relevant questions remain unanswered about the precise conditions that determine whether communities mobilize against criminal groups, strategically cooperate with them, or remain passive.
The field would also benefit from increased engagement with security studies scholars. In a recent article, Daly & Barham (2024) argue for the inclusion of criminal war in security studies and adapt a bargaining framework to shed light on why criminal groups fight or agree to peace. Despite the clear relevance of criminal violence to national and international security, the topic has not gained significant traction within traditional security studies. Bridging this gap could provide valuable insights into how criminal violence intersects with other security threats and how it shapes state responses to security challenges.
New lines of inquiry should also place greater emphasis on articulating the policy implications of academic findings. While many studies offer valuable insights and evidence to inform policymaking, there is often a disconnect between their research findings and the formulation of policy implications. Scholars should strive to translate their findings into actionable recommendations for policymakers, law enforcement agencies, and community organizations when adequate. Furthermore, researchers must approach the ethical considerations inherent in studying criminal violence with clarity and responsibility. Understanding the local context is crucial for conducting rigorous and responsible research about these topics. There has been significant progress in this regard, but it is still the case that most researchers leave out important discussions on the ethical standards of their work. Developing robust ethical guidelines and best practices for conducting field research about criminal violence should be a priority for the academic community. Baron & Young (2022) offer a notable contribution to move forward in this direction.
Finally, a promising avenue for future research could more explicitly bridge the study of criminal violence with contemporary scholarship on democratic backsliding. Organized crime, drug trafficking, gang violence, and other forms of criminal violence pose significant challenges to democratic governance, exacerbating societal tensions and political stability. Citizens’ demands for solutions can enable leaders across the political spectrum to justify authoritarian measures under the pretext of restoring order. As criminal organizations increasingly infiltrate political institutions and manipulate electoral processes, it becomes essential to understand how criminal violence contributes to the erosion of democratic norms and institutions. Pursuing this line of inquiry could provide critical insights for safeguarding democratic rights amid intensifying threats from organized criminal groups.
The author is not aware of any affiliations, memberships, funding, or financial holdings that might be perceived as affecting the objectivity of this review.
I would like to thank an anonymous reviewer and the editorial team for helpful comments and suggestions. The usual caveat applies.