Executive summary
For at least four decades, Latin American governments have invested considerable resources in an effort to disrupt and dismantle organised criminal groups. Over that period, countless criminal kingpins have been arrested, and substantial volumes of illicit goods seized or diverted before reaching their destination. Counter-crime interventions can impose substantial costs on an organisation, but these costs are often treated as evidence for a separate, weaker claim: that the organisation’s power and capacities have declined too. This report argues that inflicting damage to a criminal organisation does not necessarily weaken its capacity to operate. Most often, it transforms it.
The distinction emerges across very different countries in the region. In Colombia, armed groups draw resources from legal economic activity and regulate civilian life in areas under their influence. Brazil’s Primeiro Comando da Capital (PCC) developed around a prison-centred organisational system, whilst Mara Salvatrucha created internal links intended to preserve coordination when senior leaders were isolated. In Mexico, leadership removal affected extortion and kidnapping differently, and stronger enforcement redirected trafficking along alternative routes. The experience of military occupation in Rio de Janeiro also shows that substantial reductions in violence and visible gang presence can disappear once armed pressure recedes. In Guatemala, the experience of the International Commission against Impunity in Guatemala (CICIG) threw light on the relationships between criminal networks and state institutions that can underpin criminal power.
The report’s findings point to the need to rethink how the success of interventions against organised crime in Latin America is assessed. Operational indicators may continue to showcase the ‘achievements’ of state agencies, but claims about their wider impact require direct evidence of changes in an organisation’s criminal activity or coercive power. Interventions targeting leaders, for instance, would benefit from a prior understanding of how power is distributed inside the organisation. Likewise, assessments of territorial pressure need evidence drawn from civilian experience after armed pressure recedes, whilst drug market interventions ought to follow where illicit activity moves afterwards. Moving beyond the resilience or adaptability of criminal actors, the report introduces the idea of ‘reconnection’ to describe the restoration of an impaired criminal activity using surviving or substitute routes and relationships. In practice, this means tracking how long the disruption lasts, where and how the activity reappears, and the networks that allow it to resume. These questions should be kept separate, since they need not produce the same answer.
Introduction
In February 2015, the Colombian government launched a full-fledged military offensive against the country’s largest armed group, the Clan del Golfo. Dubbed Operation Agamenón, the offensive led to the arrest of over 1,000 members of the Clan in only two years, and to the killing of 66 of its key commanders.1 During that period, Colombian authorities also seized 94 tonnes of cocaine associated with the Clan and identified 517 assets for forfeiture.2 State forces eventually arrested Dairo Antonio Úsuga, ‘Otoniel’, the organisation’s foremost leader, in October 2021.
But despite these leadership and economic losses, the Clan del Golfo expanded considerably over the following years. According to a recent estimate, the group exercised some form of influence in 292 Colombian municipalities in 2025, up from 200 in 2022 and 179 in 2018.3 Its membership more than doubled over the same period, from 3,632 members in 2018 to 9,840 by late 2025.4
That expansion unfolded against a changing conflict landscape in the South American nation, which makes any simple causal reading of Operation Agamenón or ‘Otoniel’s’ arrest difficult to make. Across Colombia, the 2016 demobilisation of the largest guerrilla group at the time—the FARC—altered long-standing territorial orders and was followed by renewed competition in which the Clan progressively consolidated its coercive power.5 By all accounts, Agamenón inflicted substantial damage on the Clan del Golfo, but the arrests, deaths, and seizures generated by the operation concealed a longer-term reconfiguration and growth of the organisation’s territorial reach and operational capacity.
State responses to organised crime tend to be organised around identifiable organisations, their leadership structures, and the illicit markets most closely associated with them. These are important objects of intervention, but they only capture part of how criminal power is organised in Latin America. In São Paulo, Brazil, for instance, prisons have historically been a key part of the organisational infrastructure through which the PCC manages illicit flows and enforces its internal rules.6 In Michoacán, Mexico, criminal organisations have also drawn resources and power from legal sectors, extracting rents from mining and export agriculture as part of efforts to consolidate their territorial influence.7
The sections that follow test four of the most important assumptions informing state responses to organised crime in Latin America. The first assumption is that criminal power derives mainly from illicit markets, overlooking the ways in which armed groups also regulate legal economic activity and civilian life in their areas of influence. The second is that criminal organisations have a clear structure and hierarchy, and that their operational capacities rely on specific leadership figures. The third assumption flows from the second one and holds that removing those leaders fundamentally alters the criminal activities and coercive power of an organisation. The last is that operational results such as arrests, seizures, and territorial gains offer sufficient indication of the broader effects of a counter-crime intervention on an organisation and its networks.
Examining these four assumptions, the report argues that state interventions can substantially weaken an organisation without necessarily producing a corresponding decline in the activities or forms of control associated with it. As such, the report proposes separating operational results from the outcomes they are intended to produce, and following the latter over time and space. The report also introduces the idea of ‘reconnection’ to describe the restoration of an impaired criminal activity using surviving or substitute routes and networks.
Assumption 1: Criminal power derives mainly from illicit markets
The first assumption concerns the sources of criminal power. State responses across Latin America generally view illicit markets, especially drug trafficking, as the economic basis of armed organisations. But the survival of criminal groups today cannot be separated from their ability to infiltrate legal economic activities and regulate civilian life in their areas of influence.
In the Colombian department of Guaviare, FARC dissident groups have been compiling information on land ownership and measuring the size of farms to enforce an illicit taxation system.8 Charges can be set according to the number of hectares owned and, in some cases, the number of cattle on the property.9 In areas where coca used to be the dissidents’ main source of funds, these structures have expanded into other sources of revenue and even developed their own land registry to calculate what residents should pay.10 Independent reporting has corroborated these practices and documented attempts to restrict the access of state agencies responsible for official cadastral work.11
These forms of armed governance extend beyond economic extraction and into everyday civilian life.12 Across large swathes of Colombian territory, state agencies have documented systematic threats, restrictions on mobility, controls over communications, and rules imposed on civilians by armed groups.13 Research on the Clan del Golfo in southern Córdoba likewise describes local informants who perform surveillance and mediation activities, and participate in the regulation of everyday life.14
In addition to direct taxation, criminal power in the region is also exercised through legal economic activities and the movement of people and goods, sometimes through arrangements that involve state actors themselves. In the Mexican state of Michoacán, criminal organisations have permeated the legal economy, profiting from mining and agriculture.15 Their involvement in these industries has been closely tied to efforts to consolidate control over the communities and territories in which production takes place.16 Along the Colombia–Venezuela border, organised criminal actors regulate mobility and economic practices linked to flows of people and goods across the border.17 And, in southern Venezuela, non-state armed groups are known to operate protection rackets around illegal mining, whilst members of state forces have been accused of protecting, participating in, and profiting from illicit gold production and trafficking.18
These forms of criminal governance vary considerably across the region. A representative survey covering 18 Latin American countries found evidence of criminal governance practices in every country included in the study, but with substantial differences in their prevalence and reach.19 The same study distinguishes the presence of criminal organisations from their exercise of control over civilian life, finding that many groups do not impose rules on the populations around them.20 It also finds that some forms of criminal governance are positively associated with state presence, although the direction of that relationship cannot be established from the data.21
These examples complicate any account of criminal power centred exclusively on illicit markets and flows, and suggest that interventions designed exclusively around those markets may disrupt a source of revenue without necessarily weakening the wider economic and social networks on which a criminal organisation relies. The concept of ‘reconnection’ proves useful here because it draws attention to the relationships and sources of support that can survive the initial disruption and allow criminal activity to resume afterwards.
Assumption 2: Criminal organisations have clear structures and hierarchies
The second assumption is that criminal organisations have a set structure and hierarchy, and that their operational capacities rely on specific leadership figures. Across Latin America, however, criminal power is distributed very differently within and across countries and organisations.
Brazil’s PCC, one of the country’s largest criminal organisations, grew out of São Paulo’s prison system and later became involved in the Brazilian drug trade. Prisons have played a key role in its structure and internal discipline.22 The organisation’s prison origins and subsequent expansion are well documented in research on Brazilian organised crime, although the most detailed evidence on its conduct comes from documents seized directly from the PCC.23 Those documents describe a decentralised drug business organised using detailed records and an internal sanctions system to enforce debts and other commitments, both internally and externally.24 The same research contrasts such arrangements with the more territorially concentrated organisation of drug markets associated with the Comando Vermelho in Rio de Janeiro.25 Far from removing its members from criminal activity, incarceration and prison transfers created the conditions that helped the PCC to reproduce its networks and expand across Brazil.
In El Salvador, the Mara Salvatrucha gang created an internal structure known as La Federación during the period in which its senior leaders were negotiating with the state. La Federación was intended to preserve leadership continuity if the gang’s imprisoned leaders were later removed or isolated.26 It connected otherwise localised structures via intermediaries, and intercepted communications among its members indicate that these connections helped to maintain the organisation’s cohesion after tighter prison restrictions were imposed.27
The Clan del Golfo in Colombia similarly combines a recognisable leadership structure with a much larger network beyond its armed component. Even at the height of Operation Agamenón in 2016, security agencies estimated that the organisation had 1,958 armed members and 3,756 individuals in its support networks.28 By late 2025, those figures had increased to approximately 3,328 armed members and 6,512 people in support networks.29 Subsequent research continues to document the importance of these non-armed structures to the organisation’s territorial presence.30 Other Colombian armed groups are known to give their mid- and low-level commanders considerably greater autonomy and have gradually fragmented under a common organisational label.31
These cases problematise the assumption that criminal organisations are organised around a clear hierarchy. Senior leaders may enjoy significant power, but important parts of an organisation can also operate with considerable local autonomy or maintain connections that do not rest on a single chain of command.
Assumption 3: Removing leaders alters an organisation’s criminal activity and power
The third assumption follows closely from the second and holds that removing senior leaders—the so-called ‘beheading’ strategy—fundamentally alters the criminal activities associated with an organisation. Still, evidence from across the region suggests that the effects of leadership removal can vary in important ways depending on the activity being measured.
In Mexico, violence associated with criminal organisations was shown to increase during the first six months following the arrest or killing of a senior leader, whereas the effects on homicides affecting the wider population lasted longer.32 Extortion also increased during the first six months after leadership removal, but kidnapping showed no comparable statistically significant change. There was also little evidence that either offence shifted persistently into neighbouring municipalities.33
Also in Mexico, Dell follows trafficking activity after enforcement action intensified at the municipal level.34 Close municipal victories by the Partido Acción Nacional (PAN) political party increased enforcement against trafficking organisations already operating in those municipalities, after which drug flows shifted away from affected municipalities and violence increased along alternative trafficking routes.35 The idea of ‘reconnection’ helps us to distinguish changes in criminal activity from changes in the organisation or route originally targeted. Dell cannot establish whether the same trafficking organisation recreated the disrupted route, but does show that trafficking continued along alternative parts of the network after enforcement intensified.36
From 2011 onwards, Guatemalan authorities intensified the arrest and extradition of major drug traffickers, working closely with the United States and drawing on criminal justice reforms introduced in previous years.37 The first major arrests included Juan Ortiz López, described at the time as the country’s highest-ranking drug trafficker, and senior members of the Lorenzana network.38 But despite these arrests, there is little evidence that the volume of cocaine moving across Guatemala declined, even as drug-related violence appears to have fallen during this time.39
Leadership removal can thus impose substantial costs on an organisation without producing the same effect across the activities associated with it. Changes in violence, extortion, or trafficking need to be established separately rather than inferred from the removal of a criminal leader.
Assumption 4: Operational results indicate a decline in a criminal organisation’s activities and power
The final assumption is that operational results, such as arrests and seizures, are a good proxy for an intervention’s wider effects. Indeed, these conventional indicators speak to the immediate effects of state action far more clearly than they do the longer-term trajectory of the criminal activity or group being targeted.
Complexo da Maré is a complex of favelas in Rio de Janeiro, Brazil, that had long been contested by rival criminal factions. In 2014, the Brazilian government deployed the military to occupy the area as a temporary measure before the planned installation of four Police Pacification Units.40 During the first year of the occupation, homicide fell from 21 to 6 per 100,000 and criminal factions reduced much of their open presence, with drug-selling points becoming less visible.41 The planned transition never took place. The Police Pacification Units were not established, and the military remained for fifteen months rather than the four originally envisaged.42 When troops finally withdrew in 2016, the change was immediate. Within hours, the three largest gangs had re-established a street presence and dozens of open-air drug markets were operating again.43 The reduction in visible criminal control therefore lasted little beyond the occupation itself.
The duration of these effects can also vary in important ways. In Mexico, Guanajuato’s municipal police forces were placed under the operational command of state-level authorities.44 This reform reduced cartel presence and cartel wars, but was followed by an initial increase in overall violence and theft.45 These effects also lasted for different periods. The increase in homicides had largely disappeared after three and a half years, and the effect on theft after around four. Estimated reductions in cartel strength and the number of cartels disappeared after approximately five and six years, respectively.46 The reduction in cartel wars, however, persisted for at least five years.47
In Guatemala, the CICIG and domestic law enforcement institutions dismantled more than 70 criminal structures entrenched within the state between 2008 and 2019.48 Over the same period, it is estimated that the wider process of internationalised prosecution substantially reduced homicide, partly by weakening collusion between state actors and criminal networks.49 This suggests that interventions may also need to tackle the relationships that protect and sustain criminal power.
The effects of state action can also materialise beyond the territory in which it takes place. For instance, reductions in Colombian cocaine supply during the Mexican drug war (2006–2009) created measurable scarcity in Mexico and increased the value of the cocaine still in circulation.50 Competition over that scarcer commodity fed into higher levels of violence, with consequences in strategically important trafficking areas and regions where several cartels operated.51 Between 2006 and 2009, declining Colombian supply accounted for an estimated 10 to 14 per cent of the increase in Mexican violence.52
Across these cases, conventional operational results establish the immediate effects of an intervention far more clearly than they do the subsequent trajectory of the activity or mode of power being targeted. However, they provide little indication of how long those effects last or whether they reappear elsewhere.
Rethinking how state agencies measure success
The four assumptions examined above point to a common problem in how counter-crime interventions in Latin America are assessed. Operational results, showcased by governments in the region as evidence of success in the fight against organised crime, are a testimony that pressure was indeed applied to an organisation. Yet, they give a much weaker basis for judging the extent to which the illicit activity or coercive power being targeted actually changed.
They ought therefore to be recorded separately from the outcomes an intervention is intended to produce. Where an intervention also seeks to reduce a criminal activity or weaken criminal power, the outcome should be followed separately once the operation itself has ended. Recent work on cocaine market interventions makes a similar criticism of reliance on seizures and arrests as short-term measures, assessing disruption according to its effects on criminal markets and the societies around them.53
Before targeting its leadership, enforcement agencies could benefit from establishing more clearly how power is distributed inside a given criminal organisation. Existing intelligence on communications and finances can help to identify whether instructions and transactions are concentrated around a small number of people, and can also show whether local structures follow central directives and how otherwise separated parts of the organisation are interconnected.
The criminal activity targeted by an intervention also needs its own baseline and follow-up. If an operation is intended to reduce extortion, subsequent extortion victimisation and payments may offer evidence of whether that objective was achieved that the number of commanders arrested cannot. Trafficking interventions can similarly follow changes in routes and whether flows recover after disruption.
Moving from adaptation and resilience to reconnection means following the criminal activity beyond the initial disruption, including the length of that disruption and the place and moment in which activity resumes. The relationships or infrastructure that allow it to resume can be examined separately. As seen in Brazil and Mexico, illicit activities may return at a lower level across a larger portion of a country’s territory, or continue under actors who were not part of the structure originally targeted. This goes on to show why these dimensions must be followed beyond the initial point of enforcement action.
Operational statistics tend to also gloss over modes of criminal power that are experienced mainly by civilians. The multiple examples discussed in this report show that groups can lose members, from leaders to rank-and-file, and continue imposing rules on the population around them. Repeated local surveys or sentinel monitoring could establish whether civilians continue to face restrictions on movement or armed intervention in dispute resolution. Extortion victimisation can be tracked separately from formal complaints, especially where coercion discourages reporting. Crucially, Uribe et al. show that criminal governance can be measured using reported practices and discuss more localised questions concerning whether respondents themselves are subject to criminal rules or obligations.54
Agreements reached by the state with senior leaders provide equally limited evidence of implementation where local structures retain significant autonomy. Monitoring could follow the behaviour covered by an agreement in the territories where compliance is expected. An agreement on extortion, for instance, can be assessed against subsequent changes in extortion rates in those areas. Commitments affecting civilian mobility can be compared with what communities later report. The evidence presented from criminal groups in El Salvador and Colombia gives good reason to examine the extent to which directives issued centrally are implemented by local structures.
Rethinking how ‘success’ is measured in counter-crime interventions in Latin America thus entails following the activity or mode of power being targeted beyond the point of enforcement, and using indicators suited to that specific outcome.
Conclusion
Across Latin America, state responses to organised crime tend to use operational results to assess the deeper effects of a counter-crime intervention. The cases examined here expose the limits of relying on conventional indicators. In Mexico, trafficking flows moved away from municipalities exposed to stronger enforcement, and extortion followed a different trajectory from kidnapping after leadership removals. In Rio de Janeiro’s favelas, homicide fell and criminal factions reduced their visibility during the state’s military occupation, but open drug markets returned soon after troops withdrew, exposing a gap between the immediate results of enforcement and the subsequent trajectory of the illicit activities being targeted.
Interventions most often target illicit markets and senior leaders, even though criminal power often extends beyond either of them. In Michoacán, criminal organisations have drawn resources from legal economic activity, and along the Colombia-Venezuela border they regulate mobility and economic practices linked to the movement of people and goods. The PCC developed around Brazil’s prison system, and Mara Salvatrucha created links intended to preserve coordination when senior leaders were isolated, revealing that crucial links underpinning both organisations’ power extended beyond the leadership structures targeted by the respective states.
Damage to an organisation and change in the criminal activity associated with it do not always move together. Leadership removal in Mexico and Guatemala produced different effects across the activities associated with the organisations being targeted, whilst the cases from Rio de Janeiro and Guanajuato show that those effects can also last for very different periods.
Operational indicators hence provide evidence of the intervention itself. Claims about longer-term changes in criminal activity or power require the assessment to continue after the initial disruption and, where relevant, beyond the territory in which the operation took place. The idea of ‘reconnection’ helps to follow that longer-term trajectory by examining the extent to which criminal activity survives or resumes, and the networks or routes that allow it to do so.
Endnotes
1 G. Arias and C. A. Prieto Herrera, La última negociación del Clan: Antecedentes, contexto y lecciones de la negociación del gobierno colombiano con el Clan del Golfo (2015–2018) (Fundación Ideas para la Paz, 2024).
2 Ibid.
3 K. Johnson, Á. Gómez, and Á. Olaya, El poder armado en expansión: Entre el fortalecimiento y la fragmentación de los grupos armados en Colombia (Fundación Conflict Responses, 2026).
4 Ibid.
5 A. F. Aponte, L. M. Asprilla, and J. Benavides T., “Gobiernos de autodefensa territorial: De cómo los actores armados gobiernan durante el tercer ciclo de violencia en Colombia,” in Transiciones posibles de la guerra y la paz en Colombia a casi una década del Acuerdo de paz, ed. C. Niño Guarnizo and S. Tolosa Bello (Friedrich-Ebert-Stiftung Colombia and El Espectador, 2025), 57–104.
6 B. Lessing and G. D. Willis, “Legitimacy in criminal governance: Managing a drug empire from behind bars,” American Political Science Review 113, no. 2 (2019): 584–606.
7 J. S. Herrera and C. B. Martínez-Alvarez, “Diversifying violence: Mining, export-agriculture, and criminal governance in Mexico,” World Development 151 (2022): Article 105769.
8 Johnson, Gómez, and Olaya, El poder armado en expansión.
9 Ibid.
10 Ibid.
11 J. Vélez Falla, “El catastro de tierras de las disidencias en el Guaviare avanza más rápido que el del Estado,” El País, 20 October 2024, https://elpais.com/america-colombia/2024-10-20/el-catastro-de-tierras-de-las-disidencias-en-el-guaviare-avanza-mas-rapido-que-el-del-estado.html.
12 C. Voyvodic, “The legacy of rebel order: Local (in)security in Colombia,” Conflict, Security & Development 21, no. 2 (2021): 177–197.
13 Defensoría del Pueblo, “Escalada de la confrontación armada en Guaviare, Caquetá y Amazonas exige medidas urgentes para proteger a las comunidades del sur del país,” 29 May 2026, https://www.defensoria.gov.co/-/la-defensoria-del-pueblo-alerto-sobre-la-grave-escalada-de-la-confrontacion-armada-en-guaviare-caqueta-y-amazonas.
14 Aponte, Asprilla, and Benavides, “Gobiernos de autodefensa territorial.”
15 Herrera and Martínez-Alvarez, “Diversifying violence.”
16 Ibid.
17 V. García Pinzón and J. Mantilla, “Contested borders: Organized crime, governance, and bordering practices in Colombia–Venezuela borderlands,” Trends in Organized Crime 24, no. 2 (2021): 265–281.
18 International Crisis Group, Gold and Grief in Venezuela’s Violent South, Latin America Report No. 73, 28 February 2019, https://www.crisisgroup.org/latin-america-caribbean/venezuela/073-gold-and-grief-venezuelas-violent-south.
19 A. Uribe, B. Lessing, N. Schouela, and E. Stecher, “Criminal governance in Latin America: Prevalence and correlates,” Perspectives on Politics 24, no. 1 (2025): 124–142.
20 Ibid.
21 Ibid.
22 Lessing and Willis, “Legitimacy in criminal governance”; R. C. Berg, The Tussle for the Amazon: New Frontiers in Brazil’s Organized Crime Landscape (Jack D. Gordon Institute for Public Policy, Florida International University, 2021).
23 Berg, The Tussle for the Amazon.
24 Lessing and Willis, “Legitimacy in criminal governance.”
25 Ibid.
26 C. Ponce and M. Bouchard, “Still standing: Proactive resilience mechanisms in the MS-13 in El Salvador,” The British Journal of Criminology 66, no. 1 (2026): 168–187.
27 Ibid.
28 Arias and Prieto Herrera, La última negociación del Clan.
29 Johnson, Gómez, and Olaya, El poder armado en expansión.
30 Aponte, Asprilla, and Benavides, “Gobiernos de autodefensa territorial.”
31 Johnson, Gómez, and Olaya, El poder armado en expansión.
32 G. Calderón, G. Robles, A. Díaz-Cayeros, and B. Magaloni, “The beheading of criminal organizations and the dynamics of violence in Mexico,” Journal of Conflict Resolution 59, no. 8 (2015): 1455–1485.
33 P. R. Estévez-Soto and R. Lecona Esteban, “The impact of the kingpin strategy on extortion and kidnapping,” Criminology & Criminal Justice 25, no. 1 (2025): 200–227.
34 M. Dell, “Trafficking networks and the Mexican drug war,” American Economic Review 105, no. 6 (2015): 1738–1779.
35 Ibid.
36 Ibid.
37 C. D. Brockett, “The Drug Kingpin Decapitation Strategy in Guatemala: Successes and Shortcomings,” Latin American Politics and Society 61, no. 4 (2019): 47–71.
38 Ibid.
39 Ibid.
40 N. Barnes, “The logic of criminal territorial control: Military intervention in Rio de Janeiro,” Comparative Political Studies 55, no. 5 (2022): 789–831.
41 Ibid.
42 Ibid.
43 Ibid.
44 M. Alcocer, “Increasing intergovernmental coordination to fight crime: Evidence from Mexico,” Political Science Research and Methods 13, no. 3 (2025): 745–754.
45 Ibid.
46 Ibid.
47 Ibid.
48 G. Trejo and C. Nieto-Matiz, “Containing large-scale criminal violence through internationalized prosecution: How the collaboration between the CICIG and Guatemala’s law enforcement contributed to a sustained reduction in the murder rate,” Comparative Political Studies 56, no. 9 (2023): 1328–1364.
49 Ibid.
50 J. C. Castillo, D. Mejía, and P. Restrepo, “Scarcity without Leviathan: The violent effects of cocaine supply shortages in the Mexican drug war,” The Review of Economics and Statistics 102, no. 2 (2020): 269–286.
51 Ibid.
52 Ibid.
53 J. Collins, B. Pantaleão de Oliveira, A. Aponte, M. McLaggan, E. Glantz, and A. Greig, Evaluating Cocaine Market Interventions: How External Shocks and Disruption of Criminal Networks Impact the Cocaine Trade and Social Outcomes (Global Initiative Against Transnational Organized Crime, 2025).
54 Uribe et al., “Criminal governance in Latin America.”
Bibliography
Alcocer, M. (2025). Increasing intergovernmental coordination to fight crime: Evidence from Mexico. Political Science Research and Methods, 13(3), 745–754.
Amnesty International. (2026). “Security” without rights: The impact of mass arbitrary detentions and denial of due process on the possible commission of crimes against humanity in El Salvador. https://www.amnesty.org/en/documents/amr29/0967/2026/en/
Aponte, A. F., Asprilla, L. M., & Benavides T., J. (2025). Gobiernos de autodefensa territorial: De cómo los actores armados gobiernan durante el tercer ciclo de violencia en Colombia. In C. Niño Guarnizo & S. Tolosa Bello (Eds.), Transiciones posibles de la guerra y la paz en Colombia a casi una década del Acuerdo de paz (pp. 57–104). Friedrich-Ebert-Stiftung Colombia and El Espectador.
Arias, G., & Prieto, C. A. (2024). La última negociación del Clan: Antecedentes, contexto y lecciones de la negociación del gobierno colombiano con el Clan del Golfo (2015–2018). Fundación Ideas para la Paz.
Barnes, N. (2022). The logic of criminal territorial control: Military intervention in Rio de Janeiro. Comparative Political Studies, 55(5), 789–831.
Berg, R. C. (2021). The tussle for the Amazon: New frontiers in Brazil’s organized crime landscape. Jack D. Gordon Institute for Public Policy, Florida International University.
Calderón, G., Robles, G., Díaz-Cayeros, A., & Magaloni, B. (2015). The beheading of criminal organizations and the dynamics of violence in Mexico. Journal of Conflict Resolution, 59(8), 1455–1485.
Castillo, J. C., Mejía, D., & Restrepo, P. (2020). Scarcity without Leviathan: The violent effects of cocaine supply shortages in the Mexican drug war. The Review of Economics and Statistics, 102(2), 269–286.
Collins, J., Pantaleão de Oliveira, B., Aponte, A., McLaggan, M., Glantz, E., & Greig, A. (2025). Evaluating cocaine market interventions: How external shocks and disruption of criminal networks impact the cocaine trade and social outcomes. Global Initiative Against Transnational Organized Crime.
Defensoría del Pueblo. (2026, May 29). Escalada de la confrontación armada en Guaviare, Caquetá y Amazonas exige medidas urgentes para proteger a las comunidades del sur del país. https://www.defensoria.gov.co/-/la-defensoria-del-pueblo-alerto-sobre-la-grave-escalada-de-la-confrontacion-armada-en-guaviare-caqueta-y-amazonas
Dell, M. (2015). Trafficking networks and the Mexican drug war. American Economic Review, 105(6), 1738–1779.
Estévez-Soto, P. R., & Lecona Esteban, R. (2025). The impact of the kingpin strategy on extortion and kidnapping. Criminology & Criminal Justice, 25(1), 200–227.
García Pinzón, V., & Mantilla, J. (2021). Contested borders: Organized crime, governance, and bordering practices in Colombia–Venezuela borderlands. Trends in Organized Crime, 24(2), 265–281.
Herrera, J. S., & Martínez-Alvarez, C. B. (2022). Diversifying violence: Mining, export-agriculture, and criminal governance in Mexico. World Development, 151, Article 105769.
Human Rights Watch. (2025, June 27). El Salvador: Policías reconocen detenciones arbitrarias. https://www.hrw.org/es/news/2025/06/27/el-salvador-policias-reconocen-detenciones-arbitrarias
Human Rights Watch. (2026). World Report 2026: El Salvador. https://www.hrw.org/world-report/2026/country-chapters/el-salvador
International Crisis Group. (2019, February 28). Gold and grief in Venezuela’s violent south (Latin America Report No. 73). https://www.crisisgroup.org/latin-america-caribbean/venezuela/073-gold-and-grief-venezuelas-violent-south
Johnson, K., Gómez, Á., & Olaya, Á. (2026). El poder armado en expansión: Entre el fortalecimiento y la fragmentación de los grupos armados en Colombia. Fundación Conflict Responses.
Lessing, B., & Willis, G. D. (2019). Legitimacy in criminal governance: Managing a drug empire from behind bars. American Political Science Review, 113(2), 584–606.
Ponce, C., & Bouchard, M. (2026). Still standing: Proactive resilience mechanisms in the MS-13 in El Salvador. The British Journal of Criminology, 66(1), 168–187.
Rueda, A. M. (2025). Limitaciones del Estado colombiano para una estrategia efectiva frente al narcotráfico. In C. Niño Guarnizo & S. Tolosa Bello (Eds.), Transiciones posibles de la guerra y la paz en Colombia a casi una década del Acuerdo de paz (pp. 105–130). Friedrich-Ebert-Stiftung Colombia and El Espectador.
Trejo, G., & Nieto-Matiz, C. (2023). Containing large-scale criminal violence through internationalized prosecution: How the collaboration between the CICIG and Guatemala’s law enforcement contributed to a sustained reduction in the murder rate. Comparative Political Studies, 56(9), 1328–1364.
Uribe, A., Lessing, B., Schouela, N., & Stecher, E. (2025). Criminal governance in Latin America: Prevalence and correlates. Perspectives on Politics, 24(1), 124–142.
Vargas Cáceres, N. (2026, April 20). Del combate al control territorial: Adaptar la política de seguridad a la violencia contemporánea en Colombia. International Center for Transitional Justice. https://www.ictj.org/es/biblioteca-recursos/del-combate-al-control-territorial-adaptar-la-pol%C3%ADtica-de-seguridad-la
Vélez Falla, J. (2024). El catastro de tierras de las disidencias en el Guaviare avanza más rápido que el del Estado. https://elpais.com/america-colombia/2024-10-20/el-catastro-de-tierras-de-las-disidencias-en-el-guaviare-avanza-mas-rapido-que-el-del-estado.html
Voyvodic, C. (2021). The legacy of rebel order: Local (in)security in Colombia. Conflict, Security & Development, 21(2), 177–197.
Washington Office on Latin America. (2019, August 27). Fact sheet: The CICIG’s legacy in fighting corruption in Guatemala. https://www.wola.org/analysis/cicigs-legacy-fighting-corruption-guatemala/