Latin America and the Caribbean Economic Review, April 2025

Organized Crime and Violence

In Latin America and the caribbean

© 2025 International Bank for Reconstruction and Development / The World Bank 1818 H Street NW, Washington, DC 20433

Telephone: 202-473-1000; Internet: www.worldbank.org

Some rights reserved

1 2 3 4 28 27 26 25

This work is a product of the staff of The World Bank with external contributions. The findings, interpretations, and conclusions expressed in this work do not necessarily reflect the views of The World Bank, its Board of Executive Directors, or the governments they represent. The World Bank does not guarantee the accuracy, completeness, or currency of the data included in this work and does not assume responsibility for any errors, omissions, or discrepancies in the information, or liability with respect to the use of or failure to use the information, methods, processes, or conclusions set forth. The boundaries, colors, denominations, links/footnotes and other information shown in this work do not imply any judgment on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. The citation of works authored by others does not mean the World Bank endorses the views expressed by those authors or the content of their works.

Nothing herein shall constitute or be construed or considered to be a limitation upon or waiver of the privileges and immunities of The World Bank, all of which are specifically reserved.

Rights and Permissions

This work is available under the Creative Commons Attribution 3.0 IGO license (CC BY 3.0 IGO) http://creativecommons.org/licenses/by/3.0/igo. Under the Creative Commons Attribution license, you are free to copy, distribute, transmit, and adapt this work, including for commercial purposes, under the following conditions:

Attribution—Please cite the work as follows: Maloney, William, Marcela Melendez, and Raul Morales. 2025. Organized Crime and Violence in Latin America and the Caribbean. Latin America and the Caribbean Economic Review (April). World Bank, Washington, DC. doi:10.1596/978-1-4648-2235-3. License: Creative Commons Attribution CC BY 3.0 IGO

Translations—If you create a translation of this work, please add the following disclaimer along with the attribution: This translation was not created by The World Bank and should not be considered an official World Bank translation. The World Bank shall not be liable for any content or error in this translation.

Adaptations—If you create an adaptation of this work, please add the following disclaimer along with the attribution: This is an adaptation of an original work by The World Bank. Views and opinions expressed in the adaptation are the sole responsibility of the author or authors of the adaptation and are not endorsed by The World Bank.

Third-party content—The World Bank does not necessarily own each component of the content contained within the work. The World Bank therefore does not warrant that the use of any third-party-owned individual component or part contained in the work will not infringe on the rights of those third parties. The risk of claims resulting from such infringement rests solely with you. If you wish to re-use a component of the work, it is your responsibility to determine whether permission is needed for that re-use and to obtain permission from the copyright owner. Examples of components can include, but are not limited to, tables, figures, or images.

All queries on rights and licenses should be addressed to World Bank Publications, The World Bank, 1818 H Street NW, Washington, DC 20433, USA; e-mail: pubrights@worldbank.org

ISBN (electronic): 978-1-4648-2235-3

DOI: 10.1596/978-1-4648-2235-3

Cover design: Alejandro Espinosa/sonideas.com

Edition: Nancy Morrison

Acknowledgements

This report is a product of the Chief Economist Office for Latin America and the Caribbean at the World Bank. The preparation of the report was led by William Maloney (Chief Economist), Raul Morales Lema (Economist), and Marcela Meléndez Arjona (Deputy Chief Economist). Excellent research and editorial assistance were dispensed by Pilar Ruiz Orrico (Research Analyst), Gaston Marinelli (Research Analyst), José Andrée Camarena Fonseca (Research Analyst), Laura Daniela Tenjo Galvis (Research Analyst), Laura Victoria Gonzalez Solano (Research Analyst), Nicolas Peña Tenjo (Consultant) and Jacqueline Larrabure (Senior Program Assistant). Chapter 2 used a first draft of a policy document written by Marcela Meléndez Arjona, Juan Fernando Vargas, and Ernesto Schargrodsky as an input.

Substantive inputs were offered by the regional teams:

Finance, Competitiveness and Innovation: Federico Alfonso Diaz Kalan (Financial Sector Specialist, ELCFN) and Guillermo Fernandez Zubia (Consultant, ELCFN).

Poverty and Equity: Carlos Rodriguez Castelan (Practice Manager, ELCPV), Hernan Winkler (Senior Economist, ELCPV), Diana Marcela Sanchez Castro (Research Analyst, ELCPV), and Karen Yiseth Barreto Herrera (Consultant, ELCPV).

Country-specific macroeconomic estimates and write-ups were produced by country economists in the Economic Policy Global Practice, under the leadership of Shireen Mahdi (Practice Manager, ELCMU) and the coordination of Elena Fernández Ortiz and Adriane Landwehr. Contributors included Daniel Barco, Rafael Barroso, Paola Brens, Luigi Butron, Natalia Campora, Bledi Celiku, Paulo Chavez, Fabiano Colbano, Antonio Cusato, Barbara Cunha, Anton Dobronogov, Aygul Evdokimova, Cornelius Fleischhaker, Julian Folgar, Sebastian Franco, Fernando Giuliano, Christian Gonzalez, Bernard Haven, Fernando Im, Evans Jadotte, Melise Jaud, Santiago Justel, Andres Lajer, Woori Lee, Karen Muramatsu, Rafael Ornelas, Fausto Patiño, Raphael Pinto Fernandes, Maryan Porras, Daniel Reyes, Natasha Rovo, Gabriela Schmidt, Heron Teixeira, Vasileios Tsiropoulos, Hulya Ulku, Constanza Vergara, Erik von Uexkull, Pui Shen Yoong, and Gabriel Zaourak (all ELCMU).

Communications and Dissemination support was provided by Ana Elisa Luna Barros (Manager, ECRLC), Ruth Idalina Gonzalez Llamas (Senior External Affairs Officer, ECRLC), Analia Martinez (Online Communications Officer, ECRLC), Yuri Szabo Yamashita (External Affairs Officer, ECRLC), Belkis Delcid Diaz (Program Assistant, ECRLC), Carlos Alberto Cortes Galavis (Consultant, ECRLC), Francisco Seminario (Consultant, ECRLC), and Leandro Juan Hernandez (Consultant, ECRLC). Nancy Morrison (Consultant, LCRCE) provided editorial support, Leonardo Padovani (Consultant, LCRCE) translated the document, and Alejandro Espinosa (Sonideas) contributed on design.

The cutoff date for this report was April 10, 2025.

Box

Box

Figure 1.28. Public Transfers Have Mostly Returned to Pre-Pandemic Levels Amidst Limited Fiscal Space

Figure 1.29. Employment Rates Were Mostly Back to Pre-Pandemic Levels by Mid-2024, with a Few Exceptions

Figure 1.30. The Three Largest Economies in the Region—Brazil, Mexico, and Argentina—Were Already Operating at Historically Low Unemployment Rates by Mid-2024

29

29

30

Figure 1.31. Labor Income Was Stagnant in Most Countries, with Gains in Lower Income Deciles in Brazil, Colombia, and Mexico 30

Figure 1.32. Employment in Most LAC Countries Is Highly Concentrated in Small Firms with 1–4 Workers

Figure 1.33. LAC’s Low Growth, Lack of Skills and Tough Labor Rules Perpetuate the Large Informal Sector

Figure 1.34. Most of the Informal Labor Force Has Basic or Less-than-Basic Education

Figure 1.35. Labor Productivity Growth Leads to Real Labor Earnings Growth across the Region

Figure 1.36. The Potential for Labor Productivity Improvement Is Significant

Figure 1.37. Poverty Reduction Gains in LAC over the Past Two Decades Have Been Impressive

31

31

32

32

33

33

Figure 1.38. Advances in Reducing Income Inequality Have Been Moderate 34

Figure 1.39. LAC’s Net Migration Flows Continue to Increase

Figure 1.40. Other LAC Countries Have Emerged as the Top Destination for Venezuelan Migrants

Figure 1.41. Most Emigrates from LAC Countries Live in the United States

Figure 1.42. As Deportations Increase, Net Migration into the United States Continues to Be Positive for Most LAC Countries

Figure 1.43. Countries in Central America and the Caribbean Are Highly Dependent on Remittances

Figure 1.44. The Recent Spike in Deportations Has Helped Lower the Growth of Remittances

Figure B1.1.1. Employment Exposure to GenAI is Higher in the More Developed LAC Countries

Figure B1.1.2. LAC Countries Currently Use Fewer Digital Technologies in Jobs with GenAI Augmentation Potential Compared to Advanced Counterparts

Figure B1.1.3. Households in Higher Income Quintiles Have a Higher Share of Workers Exposed to GenAI Augmentation and Who Use a Computer at Work

34

35

35

Figure B1.2.1. Green Occupations and Green Sectors Are at Relatively Low Levels in LAC 41

Figure B1.2.2. The Poorest Quintiles in LAC are Most Vulnerable to Job Loss during a Green Transition

Figure 2.1. Organized Crime is Increasingly Appearing in Google Searches in LAC

Figure 2.2. Twelve LAC Countries Are Among the Top 50 (of 193) for Organized Crime

Figure 2.3. LAC Presence Is Significant in Many Criminal Markets 47

Figure 2.4. LAC Has the Highest Average Homicide Rate Worldwide 53

Figure 2.5. LAC Countries Suffer More Lethal Violence than Others with Similar Development and Inequality Levels

Figure 2.6. The Homicide Rates Varies Greatly across LAC Countries

Figure 2.7. Victimization is More Evenly Distributed than Homicides in LAC

Figure 2.8. LAC Is the World Region with the Highest Excess Lethal Criminal Violence

Figure 2.9. LAC Countries Have Relatively High Incarceration Rates

Figure 2.10. 10 LAC Prisons Are Overcrowded, and Many Prisoners Are Unsentenced

Figure B2.2.1 Homicides Declined Sharply after the Rosario Prison Intervention

55

55

56

57

59

60

Figure 2.11. The Size of Police Forces Varies across Countries 61

Figure 2.12. Most Latin American Countries Rank Poorly on the Effectiveness of their Criminal Justice Systems

65

Overview

The outlook for the Latin America and the Caribbean (LAC) region has become more uncertain as modest advances on the internal front are being compounded by a more challenging external scenario. Progress continues in the fight against inflation, although the last mile is proving long and the pace of interest rate declines has slowed. Consumer and business confidence continue to rise or remain stable, tourism in the Caribbean has fully recovered, and job markets in the largest economies continue to tighten, facilitating modest poverty reduction.

The external environment, however, has changed substantially in the six months since the October 2024 Latin America and Caribbean Economic Review (LACER), with both short-term and long-term consequences.  Inflation in the advanced countries, instead of being nearly vanquished as previously expected, looks to persist, delaying further interest rate cuts and constraining regional authorities’ room to loosen monetary policy further.  The apparent shift toward higher tariffs by the United States casts uncertainty on the nearshoring project, the practice of bringing offshore operations to nearby or friendly countries, and global market access more generally. LAC’s other principal market—China— continues to show sluggish growth. Recent cuts to overseas development assistance by world’s advanced economies will affect some countries acutely. Though over the last decade, growing migratory flows have shifted away from the United States and have increased from one LAC country to another, increased return migration from the United States will strain local labor markets and resources for reintegration.  Finally, the growing expansion of transnational organized crime has elevated crime and violence to be a dominant source of civic discontent and poses acute challenges to governance and development within LAC.

These challenges will complicate efforts to reignite growth, redress fiscal imbalances and reduce debt, and regain the advances in reducing poverty of the previous decade. LAC’s growth rates continue to be one of the slowest of any world region. Investment, both public and private, remains subdued and even before the increased uncertainty introduced by the rising US tariffs, there was evidence that the region was potentially missing the boat on “nearshoring”. The modest results in both growth and trade underline the need for progress on the long-standing reform agenda to prepare regional economies for new challenges and potential opportunities.

Chapter 1 of this report lays out the recent macroeconomic and social evolution of the region and the near-term challenges it faces, particularly in the realm of growth, trade, and balancing the fiscal accounts. Chapter 2 explores the implications for growth, social well-being, and governance of the continued influence of organized crime.

Chapter 1. Continued Recovery Amidst New Challenges

The region continues to make progress reducing inflation although the convergence to the inflation targets has slowed, largely due to increased labor costs pushing up the prices of services and continued increases in international food prices. Inflation expectations remain anchored as all major countries expect to hit their targets by 2026 and monetary policies continue to ease; both nominal and real rates have begun to fall, except for Brazil’s. In the Caribbean, because many currencies are broadly pegged, the initial inflation shock was more modest and manageable, while both the Dominican Republic and Jamaica’s targeting regimes have led to more inflation and slower reductions.

Over the medium term, international headwinds are expected to be more challenging. Less certain decreases in near-term interest rate declines in the United States and Europe may limit the room for further local reductions as countries defend against capital outflows and weak currencies. Higher rates will also contribute to modest growth in Group of Seven (G-7) countries, while China continues to struggle with a contracting real estate sector and declining consumer confidence. All these factors imply that commodity prices will soften.

Growth in LAC is forecast to reach 2.1 percent in 2025, driven by Argentina’s recovery, with other major economies showing little dynamism. Tourism-based Caribbean countries will do better, with growth in St. Vincent and the Grenadines and Dominica likely to surpass 4 percent, although continued expansion will probably require increased connectivity and growth in hotel capacity. Commodity exporters such as Trinidad and Tobago and Suriname will see slight improvements in growth this year.

On the financial front, the continued high interest rates will maintain the pressure on households and firms that has resulted in increases in non-performing loans, although recent rate cuts have provided relief in some countries. These risks must continue to be monitored, although to date, banks appear to be well provisioned and international markets remain sanguine.

Of more general concern are the ongoing global trade tensions that threaten market access and which are likely to have a dampening effect on global growth. Higher tariffs, and the highest levels of trade uncertainty in a decade, impede further integration of the region into US supply chains, as well as imperiling jobs in export-related industries. The recently signed Mexico and Mercosur agreements with the European Union represent a step toward diversification of markets and a continued relatively open stance toward the global markets. But the emerging challenges require addressing a now decades-old agenda in infrastructure, education, regulation, competition, and tax policy to increase both productivity and the nimbleness of the region’s economies in the face of new uncertainty. In addition, cuts to overseas development assistance will have varied impacts including weakening relief efforts in Haiti, weakening conservation of the Amazon in South America, and reducing humanitarian assistance to Venezuelan migrants in host countries.

On the fiscal front, government spending remains elevated, and deficits remain substantial. Debt service continues to be high, averaging 10.9 percent of government spending in the largest economies (LAC-6), importantly, driven by continued high interest rates on the expanded stock of post-pandemic debt. Transitory transfers to vulnerable individuals and businesses during the COVID-19 pandemic are continuing to recede—albeit incompletely, while in many countries, other spending has not fallen or has increased. Overall, progress on debt reduction remains limited: the debt-to-GDP ratio increased in 2024 to 63.3 percent and remains above the 2019 level of 59.4 percent. Overall, debt levels in the Caribbean remain stable, although several countries—Barbados, Belize and Jamaica in particular—have engineered impressively large declines of 26 percentage points to 40 percentage points of GDP in recent years. The entire region still needs to generate more fiscal space through gains in efficiency, spending reductions, and better designed tax systems.

Some reduction in current account imbalances has occurred, though mostly due to stagnation in investment and increased private sector savings. This low public and private capital accumulation, combined with low productivity gains over the longer term, bodes ill for long-term growth.

Poverty and Inequality

Progress on poverty reduction continues to advance, albeit at a slower pace. Estimates for 2024 indicate that monetary poverty is expected to decrease marginally to 24.4 percent of LAC’s population from 25.0 percent in 2023 (based on an upper-middle-income poverty line of $6.85 per day in 2017 purchasing power parity terms), while inequality is expected to remain high by global standards, with a Gini coefficient of 49.9 percent. About 65 percent of the improvement in poverty from 2018 to 2023 was driven by a combination of transfers and stronger labor markets, in particular higher inflation-adjusted earnings. The brightest spots were in Brazil, Colombia, and Mexico, but given that these countries were already operating in a historically low unemployment rates by mid-2024, and transfers are retrenching in many countries, additional gains are more likely to be achieved by improving the quality of jobs through productivity growth. In Colombia, for example, a 1 percent rise in productivity can lead to almost a half point rise in wages with attendant falls in poverty. At present, labor productivity hovers between 20 percent and 60 percent of levels in Organisation for Economic Co-operation and Development (OECD) member states, suggesting room for gains with appropriate reforms. Raising modern sector productivity, along with improving the skills of the region’s workers, are also critical to lowering the stubbornly high informality rate.

Creating High Productivity Jobs: The Challenge of Artificial Intelligence and the Energy Transition

Though raising minimum wages from very low levels can increase wages and lower poverty, the long-term solution to social progress is to promote growth in the number of modern sector high-productivity jobs. Two shifts in the global economy, the emergence of artificial intelligence (AI) and the evolution of the green economy, both offer challenges and opportunities to creating such jobs.

While Generative AI (GenAI) has not been widely adopted in LAC, its potential to augment jobs (increase productivity) but also automate (and displace) workers has raised concerns in the advanced countries. However, the technology is diffusing more slowly to LAC: at present, 26 percent to 38 percent of employment is estimated to be exposed to GenAI , with 1 percent to 6 percent of employment the most vulnerable—especially jobs such as customer support agents and data entry clerks. Between 7 percent to 14 percent of jobs in LAC could become more productive through GenAI augmentation, particularly in sectors such as education, health, and personal services. However, to maximize the benefits of GenAI and minimize its potential to increase inequality, complementary policies are necessary, such as developing digital infrastructure, providing income support for displaced workers, and equipping workers with foundational skills.

The energy transition has significant implications for labor demand and income inequality as well, by displacing jobs in traditional sectors while creating opportunities in new ones. Employment in sectors with relatively low greenhouse gas emissions is relatively low at about 10 percent, while employment in sectors with high greenhouse gas emissions, such as agriculture, is widespread. As with AI, emerging sectors will potentially require a different set of skills. As in most emerging sectors, providing workers with the education and skills to compete is a critical complementary policy.

The Impact of Higher Return Migration

Over the medium term, announced deportations from several advanced economies—but especially the United States—will require integrating return migrants into local labor markets. Though total migration rose slightly over the 2021–24 period to 1.5 million per year from the historically high 2016–20 period of 1.4 million, both destination and origin countries have changed substantially in the past decade. More stringent migration norms in the United States have redirected migrant flows from the United States to Latin America and Southern Europe, with approximately 20 percent of migrants destined for the United States, 61 percent to other countries in the region, and the remainder to parts of Europe. That said, the largest share of total LAC migrants (55 percent) are concentrated in the United

States, although that level is significantly down from a decade ago (68 percent). The humanitarian crisis in República Bolivariana de Venezuela has made the country the source of 57 percent of total migrant flows from LAC, while net flows from Mexico have become negative over the past decade. Because of the overall expansion of migration, even with the historically high rates of deportation in the 2021–22 period, net flows to the United States remain positive. Remittance flows remain high in some countries—amounting to more than 15 percent of GDP in Nicaragua, Honduras, El Salvador, Guatemala, Haiti, and Jamaica. These countries will see a slowdown in growth—if not a continued fall in the case of Haiti.

Finally, the rising tide of violence in the region driven by organized crime is leading to a deterioration in living conditions of the region’s citizens, increasing uncertainty that impedes the necessary investment in capital and technology for productivity growth, as well as weakening the states’ capacity to respond to their needs.

Chapter 2. Organized Crime and Violence in Latin America and the Caribbean

LAC’s traditional development challenges are increasingly compounded by the expansion of organized crime in the region beyond the countries where drug trafficking and criminal groups have been present for decades—like Colombia, Brazil, or Mexico—to new countries and industries. This expansion is not only a regional problem: LAC’s criminal groups are players in a vast international network where transnational organized crime groups are involved in a wide range of illegal activities from drugs to migrant trafficking to illegal mining and forestry. These groups are characterized by control and governance of territories; widespread extorsion, and state capture. Twelve LAC countries rank among the top fifty in 2023 by the Global Initiative Against Transnational Organized Crime (GI-TOC)’s Criminality Score.

Organized crime and violence obstruct the path to development

Organized crime poses numerous challenges, not only to citizen wellbeing but also to growth: uncertainty about property rights reduces and distorts investment; extortion and insecurity raise business transaction costs and reduce competitiveness; unproductive public security expenses divert resources away from health, education, and infrastructure investments that can improve people’s lives; victims of violence experience reductions in their capacity to accumulate human capital; crimes targeting assets or property deplete physical capital; drug trafficking, illegal mining, and flora and fauna crimes deplete natural capital through deforestation and water contamination; communities living under organized crime rule see their basic freedoms compromised; control of state institutions weakens the quality of government and the delivery of critical services.

Why is organized crime rising in LAC?

While it is not easy to determine what is behind the escalation of organized crime in the region, a series overlapping factors may be contributing. During the 2010s, global demand for cocaine, illegal gold, and migrant smuggling boomed attracting new entrants. Government crackdowns after 2005 resulted in a reorganization of LAC organized crime groups and a new landscape of actors, old and new, competing for influence. The COVID-19 pandemic allowed criminal groups to gain legitimacy and power in areas where the state was not present, provided opportunities for groups to upgrade their businesses, and facilitated the recruitment of impoverished populations. Expanded production and availability of arms have facilitated the growth of organized crime into new areas and markets. Technological innovation, including cryptocurrencies, have permitted the expansion of transnational groups.

Excess lethal

violence distinguishes organized crime in LAC from other regions

Lethal violence in LAC is higher than in any other region of the world. LAC accounts for 9 percent of the global population but records one-third of all homicides. In the first decade of this century, the average homicide rate in LAC was 5.4 times larger than the world’s (22.0 homicides per 100,000 people versus 4.1) and the gap has widened in the last 20 years to 8.0. Homicide rates in LAC countries are very high relative to those with similar per capita income and inequality levels so while social conditions explain part of the high rates, other factors are at play. Survey-based victimization rates that capture “general criminality” (both lethal and non-lethal violent crimes as well as property crimes) are also high at 30 percent, on average, although only three times higher than the world’s compared to eight times for homicides. There is an “excess” level of homicides in LAC, given what victimization rates would predict, arguably attributable to organized crime being more lethal in LAC.

Weak institutions in critical areas fuel the problem instead of mitigating it

The fight against organized crime is hampered by deficiencies in prison, police, and justice systems.

Prisons: In addition to the persistent levels of overcrowding, high recidivism, and frequent imprisonment without sentencing, in many places in LAC organized crime groups have taken control of prisons and they manage outside crime from within. As a result, prison objectives of incapacitation, deterrence, and resocialization become unattainable while prisons serve as recruitment posts for new gang members. Authorities need to regain control of prisons, improve conditions, and seek alternatives to incarceration. Further, the success of “mano dura” (firm hand) strategies is strongly context-specific, and these strategies often backfire.

Police forces: The size of police forces in the region as a share of population varies widely across countries. However, it is impossible to say whether LAC countries are over- or under-policed because factors other than size determine police effectiveness. Efficacy could be improved through better deployment across the territory, shifting of resources from patrolling to investigation, improved training and working conditions, and greater cooperation across agencies at the local and national levels and across countries. Trust in the police is among the lowest in the world and reforms need to be taken to cultivate it.

Justice systems: The region’s weak judicial systems imply that countries cannot detect, investigate, and dismantle complex criminal organizations with multiple actors and layers or, at the very least, punish the criminals and activities that generate the worst welfare externalities for the region’s citizens. Impunity is a significant problem. Part of this weakness may be due to ineffective performance indicators that do not provide the correct incentives for strategic effectiveness for law enforcement and officials across the various links in the criminal justice chain. Judicial authorities are also central to reducing incarceration. Regaining state control in areas where criminal organizations provide justice services requires bringing well-functioning justice institutions. Supporting innovative dispute resolution mechanisms can be a first step.

State action to fight organized crime requires a better understanding of how it operates

Part of the institutional weaknesses results from a lack of data or capacity to use the available data to inform the course of action. A challenge is that organized crime has both visible elements—like lethal violence—and hidden aspects that are much harder to quantify. There are no systematic comparable official surveys that can be used to target policies. This chapter emphasizes the need for specialized recurrent surveys and the smart use of administrative data, and calls attention to the role long-term, rigorous research can play in guiding state action.

Organized crime breeds in the absence of opportunities

The lack of opportunities—especially for young men, who are the primary victims and main perpetrators of criminal violence—facilitates recruitment into criminal organizations. In many cases, recruits are not forced into crime, but criminal groups offer significantly better income opportunities than those available in legal markets. They also offer career prospects and elements of “emotional mobility” (respect and recognition)—aspects the precarious jobs in the region’s licit markets often fail to provide.

Policies to prevent recruitment targeted at high-risk youth must be prioritized in the fight against organized crime because they can help break criminal organizations’ access to their potential workforces. Actions can include targeted educational interventions, community-based prevention and protection, and resocialization of former criminals. Also, interventions addressing harmful masculinities and gender inequality can help replace norms that underlie the acceptability of different forms of violence with those that construct peaceful societies. Mentorship and emotional support programs, where young people can find role models outside the criminal context, have also proven effective. Prevention cannot replace security policies but is a critical complement. Similarly, a good security policy without long-term prevention measures is likely to be insufficient.

International coordination and collaboration are critical to face organized crime

LAC states face clear challenges in building the strategic capacity they need to manage organized crime, protect their citizens from their most harmful expressions, and give them equal opportunities to build the lives they want for themselves. However, they are unlikely to succeed on their own because organized crime is a problem that transcends national borders and requires coordinated solutions. International collaboration needs to occur on several fronts, including cross-country platforms for exchanging information—across customs and tax authorities, financial intelligence units, and criminal investigation units, standardization of probatory protocols, and presenting a united voice in the global conversation around drugs, firearms and financial flows.

The scale and breadth of organized crime in LAC is arguably greater than ever. However, the region has faced this scourge before and, with the necessary reforms of governance, creation of opportunities for youth, and international collaboration, the current challenges can be met, and their threat to development reduced.

Growth Outlook for the Region

Real GDP Growth Rates

Source: World Bank staff calculations.

Note: The cut-off date for the data is April 10, 2025. “e” stands for estimate; “f” for forecast.

CHAPTER 1

The State of the LAC Region

Many assumptions about the direction of movement of the global economy have changed in the six months since the October 2024 Latin America and the Caribbean Economic Review (LACER)1. Inflation in the advanced countries, instead of being vanquished as previously expected, looks to persist, delaying further interest rate cuts and constraining room for central banks in Latin America and the Caribbean (LAC) to further loosen monetary policy. The apparent shift toward higher tariffs by the United States casts uncertainty on the nearshoring project in LAC and global market access more generally, while LAC’s other principal market, China, continues to show sluggish growth. In addition, recent cuts to overseas development assistance by world’s largest economies could be consequential for several countries. Though over the last decade, most migratory flows have been within LAC, increased return migration from the United States may stress some local labor markets and resources for reintegration, while dampening remittances. The global order will likely be in a state of flux for some time.

Economic Activity: Growth Continues to Lag

Once again in 2024, as in previous years, LAC was one of the slowest growing regions in the world (figure 1.1, panel a).

Output growth continued to moderate, largely reflecting the continued convergence to potential output of Mexico, and the adjustment process in Argentina. Growth in Brazil remained robust. The remaining large economies—Chile, Colombia, and Peru—experienced a moderate uptick in growth, narrowing the output growth gap from their negative cyclical position. For 2025, growth is expected to slightly decelerate, with only Argentina and Colombia showing stronger performance than in 2024 (figure 1.1, panel b).

FigUre 1.1.

growth Continues to lag relative to Other regions a. Growth by region

Source: World Bank Macro-Poverty Outlook (Spring Meetings 2025).

Note: The values are based on projections (as of April 10, 2025). In panel b, the LAC-6 includes Argentina, Brazil, Chile, Colombia, Mexico, and Peru. e = estimate; f = forecast; GDP = gross domestic product.

1 World Bank (2024a).

Across the region in 2024, growth continued to be propelled by consumption, except for Argentina and Chile, where growth has been led by the external sector. In Mexico, public investment supported the expansion of aggregate investment and output growth during 2023, but as it lost momentum during 2024, this stimulus weakened. In the rest of the region, investment remained subdued, due to high interest rates and expectations of persistently low economic growth (figure 1.2).

FigUre 1.2.

Consumption Continued to propel growth in most laC-6 Countries

Sources: Haver Analytics; Banco Central de Chile; World Bank Macro-Poverty Outlook (Spring Meetings 2025); and World Bank staff calculations. Note: The values are based on projections (as of April 10, 2025). f = forecast; GDP = gross domestic product. a. Inventories include statistical discrepancies.

The slowdown in internal demand has contributed to correcting imbalances in the external sector. While fiscal deficits persist across the region, increases in net savings by the private sector have helped improve the external position in most countries (figure 1.3).

FigUre 1.3.

as Budget deficits remain, increased private savings Has led to improvements in the Current account

Source: World Bank Macro-Poverty

FigUre 1.4.

Consumer Confidence Continues to rise, while Business Confidence remains stable

b. Business confidence index

Sources: For panel a, OECD (Consumer Confidence Index (CCI) (indicator) (doi: 10.1787/46434d78-en). For panel b, OECD Business Confidence Index (BCI) (indicator) (doi: 10.1787/3092dc4f-en).

Note: OECD = Organisation for Economic Co-operation and Development.

Households’ expectations continue to improve, supported largely by decreasing inflation and dynamic labor markets (figure 1.4, panel a). Business confidence remains stable around historical values, consistent with sluggish investment rates (figure 1.4, panel b). This pattern suggests that low investment, rather than being only a cyclical phenomenon, is partially explained by LAC’s economic fundamentals.

External Environment—External Headwinds Remain Challenging

The external environment continues to be challenging and is likely to remain so. On the positive side, in the third quarter of 2024, the US Federal Reserve (Fed) began to slowly normalize interest rates, a process that is expected to continue into 2025, easing financial conditions around the world. However, uncertainty remains about the speed of the interest rate decrease as the Fed delays further cuts, as well as the end point, which is likely to remain higher than in the pre-inflationary period (figure 1.5, panel a). Additionally, while commodity prices are below their 2022 peak, they remain high compared to pre-pandemic levels (figure 1.5, panel c). On the negative side, output growth of LAC’s main trading partners continues to moderate. After a slowdown in 2024, growth in the G-7 is expected to continue moderating and reaching 1.2 percent in 2025, reflecting a moderation of the economy of the United States and the anemic growth in Europe (figure 1.5, panel b). At the same time, China’s prospects remain subdued and uncertain (figure 1.5, panel d).

FigUre 1.5. external Uncertainty will remain in 2025 a. Fed funds rate and real yield on 10-year US Treasury Note

c. Commodity prices

China growth

Sources: For panel a, US Federal Reserve Economic Data (FRED), Federal Reserve Bank of St. Louis; for panel b, Organisation for Economic Co-operation and Development (OECD) Quarterly GDP (indicator) (doi: 10.1787/b86d1fc8-en) and International Monetary Fund (IMF) World Economic Outlook (WEO) database (https://www.imf.org/en/Publications/WEO/ weo-database/2024/October); for panel c, World Bank Commodity Prices (Pink Sheets) (https://www.worldbank.org/en/research/commodity-markets) and IMF WEO database; for panel d, Haver Analytics and IMF WEO database. Note: f = forecast; G-7 = Group of Seven.

Heightening Uncertainty over Trade

The less dynamic external scenario is complemented by increased uncertainty about trade policies in the region’s main trading partners, from protective measures undertaken or under study by the largest economies (United States, China, and the European Union) to realignments consistent with the nearshoring phenomenon. The Trade Policy Uncertainty Index shows that over the last 10 years uncertainty has increased, staying persistently above the levels observed before 2015 (figure 1.6).

This increased uncertainty in the world’s largest economies is especially relevant for LAC, as most countries have steadily made progress in integrating themselves into global value chains. Over the last two decades, LAC has increased its exports from 14.6 percent to 21 percent of GDP, with both the United States and China growing more important as export markets. While the United States remains LAC’s main export destination, countries in LAC have sought to diversify their trading partners. Some countries like Chile and Peru have expanded their exports to China significantly (figure 1.7), and recently Mexico and Mercosur have pursued Free Trade Agreements with the

FigUre 1.6.

trade policy Uncertainty Has increased over the past decade

Trade policy uncertainty index Index

Tari s on Chinese goods go into e ect

Announced tari s on steel and aluminum

Tari threat on Mexico

New tari threats on Mexico

New tari s on Chinese goods

Source: Caldara et al. 2020.

Note: The index represents the proportion of news articles that contain references to trade policy uncertainty (TPU). A value of 100 in the index corresponds to 1 percent of articles mentioning TPU.

FigUre 1.7.

Has Been integrating into the world

Sources: World Bank staff using data from the United Nations Department of Economic and Social Affairs (UN DESA) Statistics Division, National Accounts Main Aggregates and Commodity Trade (Comtrade) Database.

Note: The figure focuses on the LAC-6 (Argentina, Brazil, Chile, Colombia, Mexico, and Peru).

European Union. While this increased integration has created ample opportunities for firms and workers, it also means that a significant share of workers is employed in exporting industries, and consequently, exposed to changes in international trade (figure 1.8).

The sensitivity to changes in the global trade regime varies with the differing exports baskets to destination countries. While LAC exports mostly primary goods, 39 percent of exports are manufactured goods, mostly from Brazil and Mexico to the United States. In contrast, exports to China are focused almost exclusively on unprocessed food and mining. Thus, changes in global trade policies would be felt unevenly across countries and industries (figure 1.9). Even primary goods sectors will be challenged.

FigUre 1.8.

workers in export industries are exposed to Changes in international trade

Domestic employment embodied in gross exports in 2019 in LAC-6 countries

of total employment Agriculture, forestry and shing

and quarrying

Source: World Bank staff using data from OECD, Trade in Employment (TiM) database (https://www.oecd.org/en/data/datasets/trade-in-employment.html).

Note: The figure focuses on the LAC-6 (Argentina, Brazil, Chile, Colombia, Mexico, and Peru).

In addition to disruptions in international trade, recent cuts in overseas development assistance (ODA) by some of the largest economies could be consequential for the region— not only because some economies, like Haiti, are highly dependent on it, but also because it plays an important role for key policy challenges, such as Amazon conservation and humanitarian assistance to migrants from the República Bolivariana de Venezuela.

FigUre 1.9.

laC export patterns Vary by trade partner Exports by destination and sector in 2023

Source: World Bank staff using WITS (World Integrated Trade Solution).

Note: Manufacturing category includes machinery and electrical, rubber and chemicals, textiles and related, and transportation. Mining and fuels category includes metals, minerals and fuels. LAC = Latin America and the Caribbean.

It is impossible to know where the new tariff regime will settle. However, trade will remain an engine of growth for LAC. While the United States will remain a major export market worldwide, its share of global imports of goods has fallen to 13 percent, while LAC is increasingly diversifying its trade in a world that continues to pursue Free Trade Agreements and increased connection. Both Mercosur and Mexico have signed agreements with the EU, and presidents from the LAC region have recently visited India and Japan and discussed trade. The nearshoring project, although it is certainly much more uncertain, is not necessarily moribund. With major Asian competitors under heavy tariffs, Mexico and LAC look more attractive.

Further, the future of trade is increasingly in services, which have grown as manufactures trade has slowed. For instance, India and China export both low-end and high-end (engineering) services, and this will be increasingly feasible with advances in translation and communication software. LAC remains underrepresented in this area: 14 percent of LAC’s exports are in services compared to 18 percent globally and, instructively, 44 percent in India. With increases in services exports, on average, of 4.4 percent per year, LAC is growing slightly above the global average, but India, again, shows the possibilities, growing at 13.4 percent.

Building resilience against changes going forward, and taking advantage of new opportunities, will require undertaking long-delayed reforms in education, infrastructure, behind the border regulations, and the overall enabling environment that disincentive investment and trade even before the recent changes in the global landscape. Between the likely countermovement of exchange rates and the fact that certain policies, such as managerial extension programs, have been shown to achieve gains in firm productivity of 10 percent in a year, the currently discussed base level of tariffs need not be an insuperable barrier to access. Such productivity growth has proven elusive so far. Despite LAC’s comparative advantage in producing primary goods, the exporting sector is affected by the region’s lack of competitiveness in the same way as the rest of the economy. The labor productivity embedded in LAC’s export’s basket significantly lags that of the United States (figure 1.10).

FigUre 1.10.

exporters’ productivity in laC lags Behind that of the United states

Relative Export Productivity Index in 2019 as percent of US productivity in gross exports

Percent

Sources: World Bank staff using data from OECD, Trade in Employment (TiM) database (https://www.oecd.org/en/data/datasets/trade-in-employment.html); Trade in Value Added (TiVA) (https://www.oecd.org/en/topics/sub-issues/trade-in-value-added.html); and The Conference Board (https://data-central.conference-board.org/).

Note: The Relative Export Productivity Index measures the domestic value added per job in gross exports for each LAC-6 country (Argentina, Brazil, Chile, Colombia, Mexico, and Peru), expressed as a percentage of the United States (US) level.

It may be, as some have argued, that globalization has changed in such a way that LAC cannot follow the paths of the Asian miracles. But it is also the case that the region has not engaged in the building of human capital, quality universities, and research institutes linked to the private sector, or more generally embraced the goal of getting to the technological frontier that the success stories have pursued. These steps will be essential to leverage both new and traditional sectors, as well as high-end services. What is clear is that access to new technologies, and taking advantage of scale economies, will require maintaining an open stance toward the eventual system that emerges.

Inflation is Slowly Receding, while Monetary Policy Has Loosened

After the decisive response of monetary authorities and the normalization in global supply chains, inflation receded significantly throughout the region from its peak in 2022. However, during 2024, the disinflation process slowed as core inflation remained high, largely due to increased labor costs affecting the prices of services and rising international food prices (figure 1.11).

FigUre 1.11.

inflation is moderating, but more slowly than anticipated

a. Annual headline inflation (percent, year-on-year)

Source: World Bank Macro-Poverty Outlook (Spring Meetings 2025).

Note: LAC = Latin America and the Caribbean. LAC (median) includes Argentina, Belize, Bolivia, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Trinidad and Tobago, and Uruguay.

Still, inflation expectations remain anchored as markets expect inflation to fall within central bank targets for all countries by the end of 2026. However, convergence to the target will require further efforts, necessitating a slower decline in interest rates than previously anticipated. A rise in Brazilian inflation in the middle of 2024, associated with increased food prices and labor costs, led to a reversal of previous interest rate reductions (figure 1.12).

In Chile, inflation also increased during 2024, due to a combination of cost factors: depreciation of the peso, and heightened real wages and electricity rates. However, during 2025, inflation is expected to moderate as internal demand remains weak, allowing the central bank to forgo possible interest rates increases.

Considering the initially rapid decrease in inflation, the major central banks of the region began normalizing their monetary policy rates more aggressively than most advanced economies in 2024, a trend that is expected to continue during 2025 (figure 1.13, panel a). However, market expectations suggest that at least in the short to medium term, real interest rates will not reach their relatively low levels of the 2010s for most countries, staying on average 2.4 percentage points higher, reflecting higher neutral rates for the region and for major economies (figure 1.13, panel b).

inflation expectations remain anchored

Inflation forecasts and central banks targets

Percent, year-on-year

In ation target range In ation in 2024

Long-term forecast Forecast for December 2025

Forecast for December 2026

Source: Consensus Economics.

Target

Note: The survey date was April 2025 for the December 2025 and December 2026 forecasts, and April 2025 for the long-term forecasts. Long-term forecasts are period averages at the 6–10-year horizon.

The major currencies of the region have depreciated against the dollar during 2024 due to global factors, such as the relative strength of the dollar, and domestic ones, such as diminished interest rate differentials caused by monetary loosening (figure 1.14, panel b). The slower monetary policy relaxation in the United States puts further bounds on the ability of central banks in LAC to decrease interest rates if they wish to avoid further capital outflows and depreciations. While interest rate differentials for Brazil, Colombia, and Mexico are consistent with their historical levels, they are close to the minimum for Chile and Peru (figure 1.14, panel a).

monetary policy Has slowly loosened in the laC-5, except for Brazil a. Monetary policy rates, January 2021 to March 2025 and projected (percent)

Figure 1.13. Monetary Policy Has Slowly Loosened in the LAC-5, Except for Brazil (continuation)

b. Real policy rates, January 2019 to March 2025 and projected (percent)

Sources: World Bank Macro-Poverty Outlook (Spring Meetings 2025); Consensus Economics; central banks databases.

Note: For panel b, real interest rates are calculated as the difference between the nominal monetary policy rate and the expected inflation rate for the next 12 months, based on data from central banks. For forecasted values, the same calculation is applied, using national central bank forecasts for the nominal policy rate. For inflation rates, the 12-month forecast uses expected inflation for 2026 from Consensus Forecasts, while the 24-month forecast uses the inflation target for each country. The series for Mexico contains breaks due to the unavailability of inflation expectations data in December of each year since 2020, based on data from Bank of Mexico. LAC-5 includes Brazil, Chile, Colombia, Mexico, and Peru. Fed = US Federal Reserve.

FigUre 1.14.

major Currencies in the region Have depreciated due to a strong dollar and monetary loosening

a. Interest rates differentials, January 2010–March 2025

Difference between monetary policy rate and federal funds rate

b. Exchange rates, January 2024–March 2025

Local currency per US dollar (USD)

Index, 2024:M1 = 100

Sources: World Bank staff calculations based on Haver Analytics and US Federal Reserve Economic Data (FRED), Federal Reserve Bank of St. Louis.

Note: For panel a, each box plot provides a summary of the distribution of the interest rate differentials for the 2010–25 period. The central box represents the interquartile range (IQR), spanning from the first quartile (Q1, 25th percentile) to the third quartile (Q3, 75th percentile). The “whiskers” extend to the minimum and maximum values. For panel b, the nominal broad USD index measures the value of the US dollar (USD) against a trade-weighted basket of foreign currencies.

Fiscal Sector: Fiscal Deficits and Debt Remain High

Fiscal deficits remain stubbornly high in the largest economies of Latin America. While most countries have undertaken measures to address the primary deficit, such as phasing out the pandemic-era stimulus and proposed tax reforms to increase revenue, they have been unable to completely balance the fiscal budget as debt service requires increasing resources (figure 1.15).

FigUre 1.15.

Fiscal deficits persist significantly due to rising interest payments

Source: World Bank Macro-Poverty Outlook (Spring Meetings 2025).

Note: The values are based on projections (as of April 10, 2025). e = estimate;

Hence, while the likelihood of a debt crisis along the lines of those of the 1980s and 1990s in Latin America is unlikely, the relatively high public debt is still proving costly. Debt service currently represents 10.9 percent of government spending in the LAC-6, crowding out investment in items such as infrastructure to meet their citizens’ demands for public goods, as well as support for pension payments (figure 1.16). As discussed in the World Bank flagship report Public Spending Policies in Latin America and the Caribbean: When Cyclicality Meets Rigidities 2, the rigidities inherent in fiscal spending harm the ability of countries to navigate cyclical elements such as marked increase in interest rates.

FigUre 1.16.

service weighs Heavily on public spending, limiting investment

Source: World Bank Macro-Poverty Outlook (Spring Meetings 2025).

Note: The values are based on projections (as of April 10, 2025). e = estimate; GDP = gross domestic product.

2 Riera-Crichton and Vuletin (2024).

While higher interest rates are partly responsible for higher debt service, the persistent increment in government debt stock observed throughout the region over the last 15 years is just as important. After a surge during the pandemic, debt-to-GDP ratios have stabilized due to smaller deficits and the strong GDP growth during the reopening. However, in the absence of further fiscal consolidation, they are not expected to revert to the levels observed before the pandemic for most countries (figure 1.17). The October 2022 LACER, New Approaches to Closing the Fiscal Gap, proposes a framework for improving fiscal accounts by increasing revenue mobilization in growth-friendly ways3

1.17.

Source: World Bank Macro-Poverty Outlook (Spring Meetings 2025).

Note: The values are based on projections (as of April 10, 2025). e = estimate; GDP = gross domestic product.

The latest increase in debt stocks occurs in the context of relatively high interest rates for LAC, which has historically faced abnormally high interest rates on government debt (figure 1.18, panel a). The cycle of monetary policy tightening increased borrowing costs for most LAC-5 governments (figure 1.18, panels b and c). The normalization of interest rates will help alleviate this cost, but it is unlikely to return to the levels observed during the past decade. For Colombia and Mexico, sovereign risk premiums have also contributed to increase borrowing costs.

Exposure to interest rate hikes depends on how indebted countries are. While most LAC-6 economies have increased their debt interest payments, for the less indebted countries, Chile and Peru, this increase is the result of increasing debt stocks. For the other countries, increased interest rates played an important role (figure 1.19).

FigUre 1.18.

Borrowing Costs remain High

a. EMBI Global Diversified Subindexes, strip spread, regions

Sources: For panel a, World Bank staff calculations based on J.P. Morgan; for panel b, World Bank staff calculations based on Haver Analytics, Central Bank of Peru and US Federal Reserve Economic Data (FRED), Federal Reserve Bank of St. Louis; for panel c, World Bank staff calculations based on World Bank Macro-Poverty Outlook (Spring Meetings 2025).

Note: EM = emerging market; LAC = Latin America and the Caribbean.

a. For panel a, the J.P. Morgan EMBI Global Diversified Index (EMBIGD) tracks liquid, US dollar-denominated emerging market fixed, and floating-rate debt instruments issued by sovereign and quasi-sovereign entities. The Emerging Markets Bond Index (EMBI) spread reflects the difference between the yield on these instruments and the yield on US Treasury bonds of comparable maturity. Aggregates are simple averages. EM Asia includes India, Indonesia, Malaysia, Philippines, Thailand. EM Europe includes Bulgaria, Hungary, Poland, Romania, Serbia. LA7 includes Brazil, Chile, Colombia, Mexico, Paraguay, Peru, Uruguay.

b. For panel b, the spread shown in the chart represents the difference between the 10-year government bond yields of LAC-5 countries and the yield on the 10-year U.S. Treasury bond. The LAC-5 includes Brazil, Chile, Colombia, Mexico, and Peru.

c. For panel c, the implicit interest rate is measured as the ratio of interest payments to the debt stock from the previous period. Converted to real terms with consumer price index (CPI) from the U.S. Bureau of Labor Statistics. Values for implicit interest rates in 2024 are based on estimations (as of April 10, 2025). 2024e = 2024 estimate.

FigUre 1.19.

Source: World Bank Macro-Poverty Outlook (Spring Meetings 2025).

Note: Converted to real terms with consumer price index (CPI) from the U.S. Bureau of Labor Statistics. The values are based on projections (as of April 10, 2025). e = estimate.

Caribbean Focus

The Caribbean economies outperformed Latin American countries during 2024, albeit with marked differences between tourism-dependent countries and commodity exporters (table 1.1). Tourism-dependent countries have recovered their pre-pandemic levels of GDP, boosted by the rebounding tourism sector. Some countries have already reached their pre-pandemic levels of tourist arrivals, suggesting a moderation in growth rates of the services sector, while others still have room to catch up. Further efforts in improving connectivity and hotel capacity would contribute to sustaining the momentum (figure 1.20, panel a).

Among the commodities exporters countries, Trinidad and Tobago and Suriname were severely affected by the sharp decline in commodity prices during the pandemic, leading to a commensurate decline in output. As prices have rebounded since 2021, even surpassing their pre-pandemic levels, GDP growth has accelerated but has not yet reached its 2019 level (figure 1.20, panel b). In Guyana, the exploitation of recently found oil fields has led to an impressive and sustained expansion of GDP since 2020. taBle 1.1 growth prospects in the Caribbean

the regional average but Vary Real

Source: World Bank Macro-Poverty Outlook (Spring Meetings 2025).

Note: The values are based on projections (as of April 10, 2025). e = estimate. f = forecast.

1.20. while tourism-dependent Countries Have recovered from pre-pandemic levels, Commodity exporters Have not a. Tourism-dependent countries

Figure 1.20. While Tourism-Dependent Countries Have Recovered from Pre-Pandemic Levels, Commodity Exporters Have Not (continuation)

b. Commodity exporters

Sources: For panel a, World Bank Macro-Poverty Outlook (Spring Meetings 2025); Tourism Analytics; World Travel & Tourism Council (WTTC); and World Bank staff

panel b, World Bank Macro-Poverty Outlook (Spring Meetings 2025); UN Comtrade; and International Monetary Fund (IMF). Note: The values are based on projections (as of April 10, 2025). For panel a, the ratio of tourism to GDP (2023) is shown in brackets; for panel b, the ratio

GDP (2023) is shown in brackets. e = estimate; f = forecast; GDP = gross domestic product.

Like the rest of Latin America, countries in the Caribbean experienced an important increase in inflation during 2022, associated with the dynamics of food and fuels in international markets. However, because most of the Caribbean follows a currency peg monetary regime, the inflation peak was less severe than in the rest of Latin America. Jamaica and the Dominican Republic, two countries with inflation-targeting monetary policies, experienced the largest inflation surge, with considerable persistence. The normalization of prices since 2023 has contributed to bringing down inflation throughout the region (figure 1.21).

Sources: World Bank Macro-Poverty Outlook (Spring Meetings 2025), Guyana’s Bureau of Statistics, Barbados Statistical Service, International Monetary Fund (IMF) and World Bank staff calculations.

Note: The values are based on projections (as of April 10, 2025). e = estimate.

path of public debt Varies across the Caribbean and Central america regions

Sources: World Bank Macro-Poverty Outlook (Spring Meetings 2025). Note: The values are based on projections (as of April 10, 2025). e = estimate;

The path of public debt varies considerably across the Caribbean and Central America regions. Economic growth and fiscal discipline have contributed to some countries making significant progress in lowering debt-to-GDP ratios, reaching levels significantly below the pre-pandemic situation. As noted in the October 2023 LACER4. Jamaica’s use of the Economic Programme Oversight Committee (EPOC) to ensure transparency and build societal consensus around fiscal discipline serves as a model for the LAC region. Similarly, Barbados and Belize, that have struggled with high debt over the past decade, have engaged in ambitious fiscal consolidation, managing to lower debt-to-GDP ratios substantially. In contrast, several highly indebted Caribbean countries still need to undertake measures to bring debt dynamics to more sustainable patterns (figure 1.22).

Financial Sector: Non-Performing Loans Moderate, While Credit Growth Diverges across Subregions

The monetary tightening process that started in 2022 has increased borrowing costs for households and businesses across the region, leading to a moderate rise in non-performing loans (NPLs). As central banks relax monetary policy, borrowing costs are expected to fall, leading a moderation in NPLs, a trend already observable in Brazil, Colombia, and Mexico. However, slower reductions in interest rates in advanced economies will limit the downward trajectory of domestic rates in LAC. Further, in Chile, despite interest rate declines, NPLs have steadily increased since 2022, albeit from a low level (figure 1.23).

Credit growth continues to diverge between South America and Central America and the Caribbean. Most South American countries have exhibited weak credit growth, having been unable to reach their pre-pandemic trend (figure 1.24, panel a). A noteworthy exception is Paraguay, where credit growth has recently accelerated to more than 9 percent per year. In contrast, Central America and the Caribbean have exhibited significantly faster growth, consistent with their rapid economic growth (figure 1.24, panel b).

4 World Bank (2023). FigUre 1.22.

FigUre 1.23.

despite moderation in non-performing loans, Vigilance is still necessary

Share of non-performing loans to total gross loans, LAC-5

Source: World Bank staff calculations based on national statistics.

Note: Household debt is composed of mortgages and consumption loans. Non-performing loans are loans for which the contractual payments are

ratio being overdue for more than 90 days in the case of Brazil, Chile, Mexico and Peru, and 30 days in the case of Colombia.

FigUre 1.24.

Credit growth diverges among laC economies

as the

Source: World Bank staff calculations based on Haver Analytics.

Labor Markets and Social Conditions: Recent Trends and Future Prospects

Economic Growth and Poverty Reduction Have Slowed

Social progress in key variables, such as monetary poverty, has closely followed economic performance across emerging regions. In LAC, the poverty rate dropped by 17.4 percentage points during the 2003–13 “Golden Decade” of high commodity prices and robust growth. However, as economic growth slowed, poverty reduction also decelerated, with a modest 7.4 percentage point decrease from 2013 to 2024 (figure 1.25).

FigUre 1.25.

the pace of poverty reduction tracks economic growth across emerging regions

Poverty rates by world region, 2000–24

Sources: World Bank staff calculations based on Regional Poverty and Inequality Update, Latin America and the Caribbean, October 2024 (World Bank 2024b). Data for the LAC region are drawn from the World Bank Group LAC Equity Lab Database, while data for other regions are sourced from the World Bank Group Poverty and Inequality Platform (PiP). Data for China are obtained from the World Bank Group Macro Poverty Outlook (last updated: April 10, 2025).

Note: The figure uses a poverty line for upper-middle-income countries of $6.85/day in 2017 PPP terms. The LAC regional aggregate is based on 18 countries in the region for which microdata were available at national level. In cases where data were unavailable, values have been estimated using a combination of methods, including microsimulations, and then pooled to create regional estimates. Due to substantial methodological revisions in Mexico’s official household survey in 2016, which resulted in a discontinuity in the poverty series, a break has been introduced in the LAC-18 series from 2015 onward. For other regions, the 2023 and 2024 values were estimated by the nowcasting model implemented by PiP. The estimate for East Asia and the Pacific (without China) is derived by subtracting China’s data on the poor population and total population from the East Asia and Pacific aggregates published by PiP. e = estimate; f = forecast; PPP = purchasing power parity; w/o = without.

In contrast, regions with more sustained GDP and labor productivity growth, such as East Asia and the Pacific (EAP), reduced poverty more aggressively in both periods (figure 1.26). EAP, led by China’s strong performance, caught up to LAC’s poverty rate.

FigUre 1.26.

Source: World Bank staff calculations based on SEDLAC (CEDLAS-World Bank), Macro Poverty Outlook, Poverty and Inequality Platform (PiP), and Word Development Indicators (WDI).

Note: GDP measured in real terms. Labor productivity is GDP (in constant 2021 international dollars, PPP) per person employed. The figure uses a poverty line for upper-middleincome countries of $6.85/day in 2017 PPP terms. pp = percentage points; PPP = purchasing power parity; w/o = without.

Key Drivers of Poverty Reduction

LAC exhibits diverging trends in poverty reduction (figure 1.27). While most countries reduced poverty at a slower pace during the 2018–23 period than during the 2013–18 period, in Brazil, Colombia, and Mexico, poverty reduction accelerated, largely as result of improved labor markets.

FigUre 1.27.

laC’s poverty reduction trends improved in Brazil and mexico, while most Other Countries Faced setbacks in 2018–23

Contribution of household income sources to changes in poverty rate

Source: World Bank staff calculations based on SEDLAC (CEDLAS-World Bank) (last updated: October 2024).

Note: The figure uses a poverty line for upper-middle-income countries of $6.85/day in 2017 purchasing power parity (PPP) terms. The “Others” category includes remittances, retirement and pensions, demographics (share of individuals aged

and

The dynamics of labor markets and public transfers explain most (more than 65 percent) of poverty reduction in LAC across this period. The performance of real earnings—primarily determined by the ability of nominal labor earnings to keep pace with inflation—was the most crucial driver, surpassing the impact of employment levels. Public transfers

also played a vital role, both in increasing gains (as in Brazil, through the Bolsa Familia program) and in cushioning the setbacks in poverty reduction. As the pandemic-era public transfers return, albeit incompletely, to their pre-2020 levels (figure 1.28), and governments grapple with fiscal concerns, poverty reduction is expected to become even less reliant on this channel in the upcoming years. Thus, moving forward, the evolution of labor markets will be increasingly critical in the fight against poverty.

1.28.

public transfers Have mostly returned to pre-pandemic levels amidst limited Fiscal space

Average public transfer amount per household and average share of public transfer in total family income

Source: World Bank staff calculations based on SEDLAC (CEDLAS-World Bank).

Note: The bars for LAC present a simple/unweighted average of seven countries: Argentina, Bolivia, Brazil, Costa Rica, Dominican Republic, Ecuador, and Peru. LAC = Latin America and the Caribbean; PPP = purchasing power parity.

Employment and Labor Income Trends

By mid-2024, employment rates had largely returned to pre-pandemic levels, with some exceptions among specific groups (figure 1.29). Moreover, the three largest economies in the region—Brazil, Mexico, and Argentina—were already operating at historically low unemployment rates by mid-2024 (figure 1.30). This makes it unlikely that substantial social gains will be achieved, at least in the short term, by increasing employment. Thus, social policies should focus on job quality rather than quantity.

FigUre 1.29.

employment rates were mostly Back to pre-pandemic levels by mid-2024, with a Few exceptions

Employment-to-population ratio

Employment-to-population ratio in 2024:Q2 (Indexed to 2018:Q2 =

A. Largest economies

B. Southern Cone

C. Andean Region

D. Central America

E. Caribbean

Source: World Bank staff calculations using ILO (International Labour Organization) Modelled Estimates Database, ILOSTAT [database] (https://ilostat.ilo.org/data/).

Note: The employment-to-population ratio is total employment divided by working-age population. Due to data availability, for Bolivia, the compared periods are 2023:Q4 and 2018:Q4; for El Salvador, Honduras, and Panama, they are 2023 and 2018.

FigUre 1.30.

the three largest economies in the region—Brazil, mexico, and argentina—were already Operating at Historically low Unemployment rates by mid-2024

Unemployment rate, 2019–24

Average 2015-19

A. Largest economies

B. Southern Cone

C. Andean Region

D. Central America

E. Caribbean

Source: World Bank staff calculations using ILO (International Labour Organization) Modelled Estimates Database, ILOSTAT [database] (https://ilostat.ilo.org/data/). The data in the plot are primarily from the second quarter of each year. However, for Bolivia, the data correspond to the fourth quarter. For El Salvador, Honduras, Panama, and Paraguay, the data are reported annually. Note: The unemployment rate is total unemployment divided by the labor force.

Labor income presents a mixed picture. Up to 2023, average real individual labor income was mostly stagnant or decreasing throughout the region (see dotted lines in figure 1.31). This trend was particularly evident among the middle class in multiple countries, where labor income struggled to keep pace with inflation. However, in Brazil and Mexico, relatively better performance in the lower income deciles allowed these countries to still reduce poverty. In addition, the relative tightness of labor markets after the pandemic has accelerated the pace of real wage growth for most countries, most notably in Mexico, Colombia, and Brazil.

FigUre 1.31. labor income was stagnant in most Countries, with gains in lower income deciles in Brazil, Colombia, and mexico

a. Annualized real individual labor income growth by decile, 2018–23

b. Annualized real wage growth, 2021–23

Source: World Bank staff calculations based on SEDLAC (CEDLAS-World Bank).

Note: For Argentina, data have only urban coverage. In panel a, for Chile, the 2017 value is used for 2019, and the 2022 value for 2023; for Mexico, the 2018 value is used for 2019 and the 2022 value for 2023. In panel b, for Chile and Mexico, the 2020 value is used for 2021, and the 2022 value for 2023.

FigUre 1.32.

employment in most laC Countries is Highly Concentrated in small Firms with 1–4 workers

Employment composition by firm size

Brazil Mexico Argentina Chile Paraguay Uruguay Bolivia Colombia Ecuador Peru Costa Rica Dominican Rep.

Not stated

50+ persons

5-49 persons

1-4 persons

A. Largest economies

B. Southern Cone

C. Andean Region

D. Central America

E. Caribbean

Source: World Bank staff calculations using ILO (International Labour Organization) Modelled Estimates Database, ILOSTAT [database] (https://ilostat.ilo.org/data/).

Note: Due to data availability, for Bolivia, the compared periods are 2023:Q4, 2017:Q4, and 2015:Q4; for Uruguay, they are 2014:Q2, 2018:Q2, and 2021:Q2.

Increasing the availability of jobs in higher-productivity, better-paying firms is one of the main channels through which the poverty gains of the previous decade can be extended. However, jobs available in LAC have largely stagnated since 2018 in dimensions such as employer size and informality (figures 1.32 and 1.33), suggesting that, in accordance with low productivity growth, jobs offer no better opportunities today than they did in 2018. Recent global trends on artificial intelligence and the green transition present opportunities to create high-productivity and high-paying jobs (boxes B1.1 and B1.2).

1.33.

laC’s low growth, lack of skills and tough labor rules perpetuate the large informal sector Employment breakdown by institutional sector and formality

Share in total employment (%)

Private informal

Public

Private formal

A. Largest economies

B. Southern Cone

C. Andean Region

D. Central America

E. Caribbean

Source: World Bank staff calculations using ILO (International Labour Organization) Modelled Estimates Database, ILOSTAT [database] (https://ilostat.ilo.org/data/).

Note: Due to data availability, for Bolivia the compared periods are 2023:Q4 and 2018:Q4; for El Salvador and Panama, they are 2023 and 2018.

A substantial body of literature indicates that informal self-employment often emerges as a desirable (and often higher-paying) alternative to low-productivity formal jobs that poorly educated workers can access.5 In the case of LAC, the persistently high informality rates arise in the interplay between low formal sector growth, costly labor market regulations, and lack of skills (figure 1.34), which limit the supply of attractive formal sector alternatives.

FigUre 1.34.

most of the informal labor Force Has Basic or less-than-Basic education

Informal employment breakdown by educational attainment

Less than basic Basic Intermediate Advanced Not stated

A. Largest economies

B. Southern Cone

C. Andean Region

D. Central America E. Caribbean

Source: World Bank staff calculations using ILO (International Labour Organization) Modelled Estimates Database, ILOSTAT [database] (https://ilostat.ilo.org/data/).

Note: Due to data availability, for Bolivia the compared periods are 2023:Q4 and 2018:Q4; for El Salvador and Panama, they are 2023 and 2018.

Growth-enhancing reforms that increase labor productivity are the main long-term channel to revitalize real labor earnings. As well-established in the literature and evidenced across the region, there is a positive association between labor productivity growth, employment, and real labor earnings growth (figure 1.35). Moreover, there is significant potential to improve labor productivity in the region. Pre-pandemic estimates by economic sector place overall productivity at just 20 percent to 60 percent of levels in Organisation for Economic Co-operation and Development (OECD) member states, with considerable variation across sectors (figure 1.36). Despite the positive association between productivity and labor earnings, it is an open question whether there are labor market institutions that prevent the pass-through from being higher, and what reforms might increase it.

1.35. labor productivity growth leads to real labor earnings growth across the region Correlation between annual growth rates of employee’s real monthly earnings and labor productivity, 2000–22

Source: World Bank staff calculations using ILO (International Labour Organization) Modelled Estimates Database; ILOSTAT [database] (https://ilostat.ilo.org/data/); National Accounts–Analysis of Main Aggregates (AMA) Database; UN-DESA [database] (https://unstats.un.org/unsd/snaama/); and World Economic Outlook (WEO).

Note: Labor productivity is

“contemporaneous”

5 See Perry et al. (2007).

FigUre 1.36.

the potential for labor productivity improvement is significant Labor productivity as a percentage of the OECD average, 2018

Finance and business services

Other services

Total

productivity (% of OECD average)

Source: World Bank staff calculations using harmonized series from the WBG Aggregate and Sectoral Productivity Database [database] (https://prosperitydata360.worldbank.org/ en/dataset/WB+ASPD) (last updated: March 27, 2024).

Note: Labor productivity is measured as real GDP per worker (at PPP-adjusted 2011 international dollars). OECD average excludes countries from Latin America and the Caribbean. OECD = Organisation for Economic Co-operation and Development; PPP = purchasing power parity.

Acknowledging Social Progress and Future Challenges

Overall, it is important to acknowledge the impressive gains in poverty reduction in LAC over the past two decades, while recognizing that the pace of this progress has significantly slowed and that, for most countries in the region, much more remains to be done—especially when compared to the average OECD poverty rate (figure 1.37). In terms of reducing income inequality, more moderate advances have been made in lowering the Gini coefficient. However, by the end of 2024, most of the region remained above the World Bank’s high inequality threshold, defined by a Gini coefficient greater than 40 (figure 1.38).

FigUre 1.37.

poverty reduction gains in laC over the past two decades Have Been impressive

Evolution of poverty rates

Poverty headcount ratio (% of population)

2008 2019 2023 2024e 2025f

OECD average

A. LAC aggregate

B. Largest economies

C. Southern Cone D. Andean Region

E. Central America F. Caribbean

Source: World Bank Macro Poverty Outlook (April 2025, forthcoming).

Note: The figure uses a poverty line for upper-middle-income countries of $6.85/day in 2017 PPP terms. For Argentina, data have only urban coverage. For Brazil, Costa Rica, the Dominican Republic, Honduras, Mexico, and the LAC aggregate, due to survey and methodological changes, the 2008 value is not entirely comparable with the rest of the series. For Colombia, Paraguay, and Uruguay, due to survey and methodological changes, values from 2023 onward are not entirely comparable with the rest of the series. For Chile, the 2009 value is used for 2008 and the 2017 value for 2019. For Guatemala, the 2006 value is used for 2008 and the 2014 value for 2019. For Mexico, the 2018 value is used for 2019. Mexico 2023, 2024, and 2025 figures are a projection using neutral distribution (2022) with pass-through = 0.87, based on GDP per capita in constant local currency units (LCU). The LAC regional aggregate is based on 18 countries in the region for which microdata were available at national level. In cases where data were unavailable, values have been estimated using a combination of methods, including microsimulations, and then pooled to create regional estimates. Updated April 10, 2025. 2024e = 2024 estimate. 2025f = 2025 forecast. LAC = Latin America and the Caribbean; PPP = purchasing power parity.

FigUre 1.38.

advances in reducing income inequality Have Been moderate

Evolution of Gini coefficient

2008 2019 2023 2024e 2025f

OECD average

A. LAC aggregate

B. Largest economies

C. Southern Cone

D. Andean Region

E. Central America

F. Caribbean

Source: World Bank Macro Poverty Outlook (April 2025, forthcoming).

Note: For Argentina, data have only urban coverage. For Brazil, Costa Rica, the Dominican Republic, Honduras, Mexico, and the LAC aggregate, due to survey and methodological changes, the 2008 value is not entirely comparable with the rest of the series. For Colombia, Paraguay, and Uruguay, due to survey and methodological changes, values from 2023 onward are not entirely comparable with the rest of the series. For Chile, the 2009 value is used for 2008 and the 2017 value for 2019. For Guatemala, the 2006 value is used for 2008 and the 2014 value for 2019. For Mexico, the 2018 value is used for 2019. Mexico 2023, 2024, and 2025 figures are a projection using neutral distribution (2022) with pass-through = 0.87, based on GDP per capita in constant local currency units (LCU). The LAC regional aggregate is based on 18 countries in the region for which microdata were available at national level. In cases where data were unavailable, values have been estimated using a combination of methods, including microsimulations, and then pooled to create regional estimates. Updated April 10, 2025. 2024e = 2024 estimate. 2025f = 2025 forecast. LAC = Latin America and the Caribbean.

Migration Diversion

Just as they have tightened their trade policies, several advanced economies have shifted toward more restrictive migration regimes in recent years. This change is especially relevant for LAC because 7.3 percent of all citizens of the region are migrants, and the flow of migrants has steadily risen over the last 15 years (figure 1.39). Moreover, several countries in Central America and the Caribbean are highly dependent on the remittances their expats send back home, which in some cases surpass 20 percent of domestic GDP.

The more stringent migration norms in the United States have generated a shift of migrant flows from the United States to Latin America and, albeit less severely, to Southern Europe (mostly to Spain, Italy, and Portugal). While during 2006–15 most new migrants moved to the United States (56 percent) and only 26 percent to other countries in LAC, over the last decade migration within the region has represented 61 percent of all new migrants, and migration to the United States, 20 percent (figure 1.40). However, the largest stock of migrants from LAC remains in the United States (55 percent) (figure 1.41), although the share is significantly lower than a decade ago (68 percent). Over the same period, the share in other LAC countries has risen from 17 percent to 29 percent.

FigUre 1.39. laC’s net migration Flows Continue to increase LAC net migration flow, annualized, by destination -200,000 0 200,000 400,000 600,000 800,000 1,000,000 1,200,000 1,400,000 1,600,000 1,800,000

Number of new migrants USA LAC Southern Europe Western Europe Eastern Europe

Others

Source: World Bank staff calculations using International Migrant Stock database (UN-DESA, Population Division, 2024).

Note: LAC = Latin America and the Caribbean; USA = United States of America.

Other laC Countries Have emerged as the top destination for Venezuelan migrants LAC net migration flow composition a. By origin b. By destination

Source: World Bank staff calculations using International Migrant Stock database (UN-DESA, Population Division, 2024).

Note: LAC = Latin America and the Caribbean.

The shift in destinations has been accompanied by a change in the countries of origin. The humanitarian crisis in the República Bolivariana de Venezuela has fostered a diaspora among its citizens, which, in addition to the relatively high growth of Mexico and the Caribbean economies, have decreased the incentives of locals to emigrate. Countries such as Colombia, Peru, Brazil, Chile, and Ecuador have been especially targeted by the emigrates from the República Bolivariana de Venezuela because of their cultural and geographical proximity, in addition to their migration laws (figure 1.41).

Source: World Bank staff calculations using International Migrant Stock database (UN-DESA, Population Division, 2024).

Note: LAC = Latin America and the Caribbean; USA = United States of America.

FigUre 1.41. most emigrates from laC Countries live in the United states

LAC migration stock (thousands of migrants), by destination, 2024

Costa Rica 564

Dominican Rep 682

Ecuador 690

Italy 719

Mexico 745

Brazil 904

Canada 983

Venezuela 1,032

Chile 1,406

Argentina 1,574

Peru 1,738

Colombia 2,973

Spain 4,101

Others 3,384

Panama 368

Source: World Bank staff calculations using International Migrant Stock database (UN-DESA, Population Division, 2024).

Note: USA = United States of America.

Deportations to LAC nations from the United States have increased in recent years and are expected to accelerate starting in 2025. In addition, the region has agreed to receive repatriations from the United States and extra regional flows. Before 2025, Mexico, Guatemala, and Haiti had the highest returnee numbers, but the most dramatic increases have been in Colombia and Ecuador, with 369.5 percent and 366.1 percent growth in returnee flows since 2019, respectively. However, while deportations target a significant number of migrants, across the 2019-24 period net migration to the United States continued to be positive because the share of newcomers was larger than the share of deportees (figure 1.42).

This increase in returnee flows is challenging in different dimensions. First, it has been accompanied by higher vulnerability. Several countries in LAC report that a significant share of deportees are minors, and/or in need of state services. In addition, reintegration of returnees is likely to be challenging. These migrants return to countries without roots, jobs, or social networks. Thus, they are likely to migrate again (secondary migration) to places such as Chile, Brazil, and Europe, and their integration into local labor markets is likely to be costly. For example, a substantial

FigUre 1.42.

as deportations increase, net migration into the United states Continues to Be positive for most laC Countries

a. Total deportations from the United States

Total number of returned per

Source: World Bank staff calculations using International Migrant Stock database (UN-DESA, Population Division, 2024) and statistics on the number of returned, processed by the World Bank’s Social Development Global Practice.

Note: Includes returns from migrants of Argentina, Bolivia, Colombia, Cuba, Ecuador, El Salvador, Guatemala, Haiti, Honduras, Mexico, Nicaragua, and Venezuela, RB.

b. Evolution of returned migrants and total migrant stock living in the United States, per country of origin

Percent change in migrant stock Returned (as % of the period's initial migrant stock)

Source: World Bank staff calculations using International Migrant Stock database (UN-DESA, Population Division, 2024) and statistics on the number of returned, processed by the World Bank’s Social Development Global Practice.

number of children and adolescents would require education and training to integrate effectively.6 Second, several countries in Central America and the Caribbean are highly dependent on remittances (figure 1.43), which are likely to decrease should deportations increase significantly enough to decrease the number of migrants in the United States (figures 1.43 and 1.44). Recent news about the United States government planning to revoke the legal status of a significant number of migrants from several LAC countries could prove consequential, as in some of these economies, received remittances represent a large portion of their GDP.

6 For a thorough analysis on the challenges posed by returned migration, please refer to Latin America and the Caribbean Regional Strategy for Human Mobility (World Bank, forthcoming).

FigUre 1.43.

Countries in Central america and the Caribbean are Highly dependent on remittances

Personal remittances received, 2023 (percent of GDP)

Personal remittances, received (% of

A. Largest economies B. Southern Cone C. Andean Region D. Central America

Caribbean Mean

Source: World Bank staff calculations based on World Development Indicators (WDI).

FigUre 1.44. the recent spike in deportations Has Helped lower the growth of remittances

Annual growth rate in real personal remittances received

Source: World Bank staff calculations based on World Development Indicators (WDI) and World Economic Outlook (WEO). Note: Real personal remittances received are calculated as nominal remittances (in millions of US dollars), adjusted for inflation using the US Consumer Price Index (CPI).

Rising Sources of Social Insecurity

Despite the continued—albeit slowing—progress on the poverty and inequality fronts, crime and violence have climbed to be the number one concern in many public opinion polls. Indeed, both the number of incidents (raw quantity) of crime and the nature of crimes have intensified, becoming more associated with organized crime previously not experienced in many parts of the region. This new source of social insecurity not only lowers the quality of life but also threatens the growth dynamism that is necessary for progress on all social fronts. This is the subject of the next chapter.

BOX 1.1.

Can artificial intelligence Be the long-awaited remedy for income inequality in latin america?

Public interest in Generative AI (GenAI) has significantly increased in recent years.a The emergence of conversational models such as ChatGPT has raised concerns in various areas, particularly regarding the impact on employment. Most US adults are more worried than excited about AI in daily life, mainly due to fears of job lossb.

The broad adoption of this new technology is still pending, making it premature to gauge its impacts on jobs. However, we can evaluate the extent to which jobs are exposed to GenAI. When a job is exposed to GenAI, it means many tasks within that job are likely to be automated by GenAI. Yet automating some tasks does not mean the entire job will be replaced. In fact, exposure to GenAI could make some jobs more productive if workers can delegate tasks to the technology and focus on other tasks.

In LAC, 26 percent to 38 percent of employment is exposed to GenAI. Exposure is higher in more developed countries such as Costa Rica and Uruguay, and it is lower in Bolivia, Ecuador, and Barbados (figure B1.1.1). This overall level of GenAI exposure can be divided into three categories. First, some jobs are at risk of being automated by GenAI, representing about 1 percent to 6 percent of employment across countries in the region. Examples of these occupations include customer support agents and data entry clerks. Second, some jobs can be augmented or become more productive through GenAI. About 7 percent to 14 percent of employment across LAC countries fall into this category. In other words, the fraction of jobs that could benefit from GenAI is larger than the fraction facing the risk of automation.

FigUre B1.1.1.

Source: Gmyrek, Winkler, and Garganta 2024. Note: “The Big Unknown” refers to jobs for which it is unclear which is higher: the risk of

or the potential for augmentation. HICs = high-income countries; LAC = Latin America and the Caribbean.

Examples of occupations exposed to GenAI augmentation include teachers, health professionals, and coaches. Finally, for a significant portion of jobs (13 percent to 22 percent), it is not yet clear whether they face the risk of automation or have the potential for augmentation.

To understand the impacts of GenAI on income inequality, we need to identify the profile of exposed workers. In LAC, those at risk of automation are typically female, urban, young, and well-educated, have high incomes, and work in formal jobs in banking, finance, insurance, or the public sector. The profile for GenAI augmentation is similar but less defined by age and gender, with higher exposure in education, health, and personal services. These trends suggest that while automation risks could increase inequality across gender and age groups, the benefits of GenAI, which favor higher-income and well-educated households, may also lead to greater income inequality.

These estimates of exposure do not consider that the rate of GenAI adoption in LAC may differ from that in wealthier regions. For example, while the same occupation might have a similar potential for augmentation in both New Zealand and Peru, the adoption pace of GenAI might be slower in Peru. Consequently, the advantages of this technology may take longer to materialize in developing regions.

When considering the distance to the technology frontier, the potential for job augmentation shows a significant disparity between wealthier and poorer countries (figure B1.1.2). Specifically, although the proportion of jobs that could benefit from GenAI is similar (about 14 percent) in wealthy countries like New Zealand and Slovenia as well as in selected LAC countries, the fraction unable to utilize GenAI due to technological limitations is notably higher in the latter. For example, only 5.1 percent of Peruvian workers have roles exposed to GenAI augmentation and simultaneously use digital technologies at work, compared to more than 11 percent in wealthier nations.

laC Countries Currently Use Fewer digital technologies in Jobs with genai augmentation potential Compared to advanced Counterparts

Source: Gmyrek, Winkler, and Garganta 2024, based on PIAAC data.

Note: PIAAC data refers to microdata from the Programme for the International Assessment of Adult Competencies (PIAAC) collected by the Organisation for Economic Co-operation and Development (OECD). LAC = Latin America and the Caribbean.

Similar patterns are observed when evaluating the proportion of jobs exposed to GenAI across different income levels within countries. In both Brazil and Mexico, the percentage of workers who can benefit from GenAI significantly increases with their income levels (figure B1.1.3). Individuals in the highest income quintile are at least twice as likely to have access to such positions compared to those in the lowest quintile. This disparity is further amplified when considering access to digital technologies. In Mexico, for instance, workers in the top income quintile are 5.6 times more likely to hold jobs that involve GenAI augmentation and utilize computers than those in the bottom quintile. Across the LAC region, there are approximately 17 million jobs that could potentially benefit from GenAI but are hindered by inadequate access to digital technologies. These missed opportunities disproportionately affect lowerincome countries and poorer segments of the workforce.

FigUre B1.1.3.

Households in Higher income Quintiles Have a Higher share of workers exposed to genai augmentation and who Use a Computer at work

Brazil

Source: Gmyrek, Winkler, and Garganta 2024.

Note: p.p. = percentage points.

Mexico

In summary, while GenAI can benefit many workers in the region, complementary policies are needed to ensure these benefits. This includes developing digital infrastructure, offering income support for displaced workers, and providing foundational skills to adapt to job changes. These policies should aim to help all workers, not just those at the top, to use GenAI effectively and thus mitigate a potential increase in inequality.

Notes

a. This box is based on Gmyrek, Winkler, and Garganta (2024). b. Rutgers (2024).

BOX 1.2. are latin american workers ready for an energy transition?

While “green” technologies can promote a healthier planet, their adoption can have unintended consequences.a These technologies have the potential to reduce fossil fuel reliance, boost energy efficiency, and increase renewables. Adopting them transforms production processes, changing input types and quantities while mitigating negative impacts on the environment. Importantly, they affect labor demand, making some skills essential and others obsolete, potentially leading to increased income inequality. The impact of green technologies on workers will depend on the balance of skill demand and supply. Inclusive growth is more likely when the needed skills are plentiful, as displaced workers can find jobs more easily and wages for these skills may rise.

What are the expected impacts of the green transition on workers in the LAC region?

The concentration of employment in green occupations and sectors is relatively small in several LAC countries (figure B1.2.1). Green occupations include various jobs such as those intensive in specifically (pure) green skills (for example, climate change specialists); those that may require some re-skilling (for example, an architect needing training to use green materials); and existing occupations that may see increased demand even if they are not distinctly “green” (for example, carpenters). Green sectors are characterized by having relatively low greenhouse gas emissions (GHGE) per worker. Workers in these sectors face lower displacement risk during the green transition compared to their counterparts in non-green sectors. Approximately 90 percent of workers in each country in the region are employed in either a non-green sector or a non-green occupation. Non-green sectors account for a significant portion of jobs, ranging from 37 percent in Argentina to 59 percent in Guatemala. This share is influenced by the size of the agricultural sector, which has one of the highest levels of GHGE per worker and predominantly employs individuals in non-green occupations—about 90 percent or more in most countries.

Source: Winkler et al. 2024.

Note: Data are as 2019. LAC = Latin America and the Caribbean.

The profile of workers in green jobs within countries highlights the potential distributional impacts of the green transition. In the LAC region, men and urban workers dominate green occupations and have better re-employment prospects if displaced. However, overall employment in green sectors is higher for women and those with higher education levels, who face a lower risk of displacement. This gender bias in green occupations reflects the male dominance in roles like chief executives, electricians, and construction workers. Conversely, women’s higher presence

B1.2.2. the poorest Quintiles in laC are most Vulnerable to Job loss during a green transition

Green Occ. in Green Sectors

Non-green Occ. in Green Sectors

Green Occ. in Non-green Sectors

Non-green Occ. in Non-green Sectors

Source: Winkler et al. 2024. Note: Data are as of 2019.

in green sectors is due to their concentration in services such as the care economy and retail. Differences in the share of green occupations across household income quintiles is not particularly large (figure B1.2.2), but the share of the most vulnerable workers during a green transition (that is, workers in non-green occupations and in non-green sectors) is significantly higher in the poorest quintiles.

In conclusion, these findings underscore the necessity for complementary policies that address the potential impacts of green transitions on income inequality. Countries with a lower share of green occupations often face challenges such as lower levels of human capital and higher rates of labor market informality. These factors raise concerns regarding the preparedness of their workforce to acquire the skills required for a green economy and to safeguard themselves against the risk of job displacement during this transition.

Notes a. This box is based on Winkler et al. (2024).

References

Caldara, D., M. Iacoviello, P. Molligo, A. Prestipino, and A. Raffo. 2020. Journal of Monetary Economics, Vol. 109: 38-59.

Gmyrek, P., H. Winkler, and S. Garganta. 2024. “Buffer or Bottleneck? Employment Exposure to Generative AI and the Digital Divide in Latin America.” World Bank Policy Research Working Paper 10863. Washington, DC: World Bank.

Perry, G. E., W. F. Maloney, O. S. Arias, P. Fajnzylber, A. D. Mason, and J. Saavedra-Chanduvi. 2007. Informality: Exit and Exclusion

World Bank Latin American and Caribbean Studies. Washington, DC: World Bank.

Riera-Crichton, D., and G. Vuletin. 2024. Public Spending Policies in Latin America and the Caribbean: When Cyclicality Meets Rigidities. Latin American Development Forum Series. Washington, DC: World Bank.

Rutgers. 2024. U.S. Workers Assess the Impacts of Artificial Intelligence on Jobs: Topline Survey Results. New Brunswick, NJ: Heldrich Center for Workforce Development.

Winkler, H., V. Di Maro, K. Montoya, S. Olivieri, and E. Vazquez. 2024. “Measuring Green Jobs: A New Database for Latin America and Other Regions.” Policy Research Working Paper 10794. Washington, DC: World Bank.

World Bank. 2022. Latin America and the Caribbean Economic Review, October 2022. New Approaches to Closing the Fiscal Gap.  Washington, DC: World Bank.

World Bank. 2023.  Latin America and the Caribbean Economic Review, October 2023. Wired: Digital Connectivity for Inclusion and Growth. Washington, DC: World Bank.

World Bank. 2024a. Latin America and the Caribbean Economic Review, October 2024. Taxing Wealth for Equity and Growth. Washington, DC: World Bank.

World Bank. 2024b. Regional Poverty and Inequality Update, Latin America and the Caribbean, October 2024 . Washington, DC: World Bank.

World Bank. 2025. Macro-Poverty Outlook (Spring Meetings 2025). Washington, DC: World Bank.

World Bank. Forthcoming. Latin America and the Caribbean Regional Strategy for Human Mobility. Washington, DC: World Bank.

CHAPTER 2

Organized Crime and Violence in Latin America and the Caribbean

LAC Must Prioritize the Fight Against Organized Crime

The traditional development challenges in Latin America and the Caribbean (LAC) are increasingly compounded by the expansion of organized crime in the region. The regions’ news outlets document not only rising homicide rates but killings ordered from within prisons, sometimes involving civilians; politicians and government officials under threat; candidates for office assassinated, altering elections; murders involving people with alias names and kilograms of lost cocaine; businesses that must pay extortion fees to operate; and parts of the territory—neighborhoods, cities, and rural municipalities—under criminal control.7 The grip of organized crime has gotten tighter not only in the countries where drug trafficking and criminal groups, in one form or another, have been present for a while, such as Colombia, Brazil, and Mexico, but also in new countries; each year, another country seems to be affected.8 The growing spread and intensity of crime is matched by growing queries by the public on Google Trends—a revealing source of data about topics that were not in the local mindset a few years back but have gained prominence recently (figure 2.1). The frequency with which LAC countries appear among those with more queries on “organized crime” has grown from 3 among the top 20 in 2005 to 13 among the top 20 in 2024 and 2025.

FigUre 2.1.

Organized Crime is increasingly appearing in google searches in laC Web searches of “organized crime” on Google, selected countries, 2011/12–2025

Source: World Bank staff calculations based on Google Trends data.

Note: Numbers represent search interest relative to the highest interest point in each country. 100 is the topic’s peak popularity. A value of 50 means the topic is half as popular as it was at its peak. LAC = Latin America and the Caribbean.

This chapter argues that organized crime is one of the region’s most pressing problems and must be at the center of any conversation about development. While it is a problem present in many countries and regions around the world, and eradicating it will require international collaboration and coordinated solutions beyond national borders, its particular grip on the region demands urgent action. In LAC, organized crime tends to be more violent than in other places. It also flourishes locally through territorial control, coercion, and extortion; the capture of state institutions; and criminal governance—sometimes with rules that limit individual freedoms, including the right to move, work, and vote freely. The impediments it poses to the region’s development are myriad: uncertainty about property rights reduces and distorts investment; extortion and insecurity raise business transaction costs and reduce competitiveness; unproductive public security expenses divert resources that could otherwise go into health, education, and infrastructure, improving people’s lives; victims of violence experience reductions in their capacity to accumulate human capital; communities living under organized crime rule see their basic freedoms compromised; and control of state institutions weakens the quality of government. The channels through which organized crime contributes to low productivity, low growth, poverty, and high inequality in LAC are countless.

7 Abi-Habib, Correal, and Nicas (2024); Economist (2024b); Freeman (2024); Rodríguez Mega (2024).

8 Abi-Habib (2024); Nicas, Bonnefoy, and Bartlett (2025).

Organized crime feeds on an absence of opportunities, insufficient state presence, and institutional weakness. In the medium and long term, the best public security policy is building more functional states that can offer equal opportunities to all, including better education systems and labor markets that work well and can offer quality jobs. In the short term, however, LAC must prioritize strategic capacity building in prisons, police, and justice systems to mitigate the most harmful expressions of organized crime and control its further expansion. It must also prioritize prevention interventions targeting youth at risk of joining criminal groups. Both efforts require better data and studies to inform policy design. Finally, coordination among institutions, as well as collaboration and coordination across levels of government and internationally among countries, will be critical to managing a problem no single country can tackle alone.

Organized Crime Is a Global Phenomenon Extending Its Reach Throughout LAC

The growing presence of organized crime is not a problem exclusive to the region. On the contrary, the criminal groups operating in LAC are players in a vast international network where transnational organized crime groups are involved in a wide range of illegal activities, including drug trafficking, migrant smuggling, human trafficking, money laundering, firearms trafficking, extortion, illegal mining, and cybercrime, among others, interacting with one another at different levels and in various capacities.

There is a complete absence of hard data on organized crime comparable across countries, partly due to its illicit nature. The Global Initiative against Transnational Organized Crime (GI-TOC) has recently produced an Organized Crime Index based on quantitative data and qualitative assessments that attempts, however imperfectly, to offer internationally comparative measures. The index is a composite of two scores, one on criminality and the other on resilience. The first score assesses “the scale, scope, and impact” of 15 criminal markets and “evaluates the structure and influence” of five types of criminal actors.9 The second score measures countries’ preparedness to face organized crime.

The criminality score shows organized crime is a serious matter in LAC, but not exclusively (figure 2.2). Twelve LAC countries ranked among the top fifty by this score in 2023, alongside countries such as Myanmar, the Democratic Republic of Congo, Nigeria, South Africa, the Russian Federation, and Italy. 10 Notably, while organized crime is a

twelve laC Countries are among the top 50 (of 193) for Organized Crime Criminality Score, top 50 in the sample, 2023

Criminality score [1-10]

Source: World Bank staff calculations based on Global Initiative against Transnational Organized Crime data.

Note: LAC countries are highlighted in yellow. The Criminality Score in the Global Organized Crime Index measures the extent and impact of organized crime in a country on a 0–10 scale, with higher scores indicating more criminal activity. The figure uses International Organization for Standardization (ISO) country codes. LAC = Latin America and the Caribbean.

9

financial crimes. Criminal actors are mafia-style groups, criminal networks, state-embedded actors, foreign actors, and private sector actors.

10 They are, in order, Colombia, Mexico, Paraguay, Ecuador, Honduras, Panama, Brazil, Venezuela, Guatemala, Peru, Guyana, and Haiti.

significant problem in several LAC countries, including some of the region’s larger economies, not all countries are equally affected yet, and those where signals of organized crime presence are incipient may have a head start in protecting themselves.

Considering the score’s components independently offers insights about the criminal markets in which each country participates and in how many of these markets it participates. Overall, the rankings suggest different business specialization patterns across countries and country specialization along the vertical production chains of the various criminal businesses (in LAC and elsewhere). For example, not all countries that rank among the top 50 for cocaine trade (production, distribution, and sales) show up among the top 50 for financial crimes. LAC has a dominant presence in the cocaine trade, where 17 LAC countries (out of 33 in the data set) rank among the top 20 by the GI-TOC scores. And, while the number among the top 20 in other criminal market scores is lower, LAC countries still rank high for cannabis trade (15), extortion and protection racketeering (11), arms trafficking (4), and financial crimes (4), among others. Mexico appears in the most categories (13) among the top 20 countries across the 15 criminal markets in the GI-TOC index, followed by Brazil and Colombia, each with 8 (figure 2.3).

laC presence is significant in many Criminal markets

Number of LAC countries among the top 50 and the top 20 by criminal market, 2023

Cocaine trade

Cannabis trade

Extortion and protection racketeering

Arms tra cking

Fauna crimes

Flora crimes

Human tra cking

Financial crimes

Non-renewable resource crimes

Illicit trade in excisable goods

Trade in counterfeit goods

Human smuggling Cyber dependent crimes

Heroin trade

Synthetic drug trade

Number of countries In the top 50

Source: World Bank staff calculations based on Global Initiative against Transnational Organized Crime data.

Note: Multiple countries can have the same score and be tied with others within the top 20 or 50. As a result, the top group may include more than 20 or 50 countries. LAC = Latin America and the Caribbean.

Little is known about where and how precisely these activities materialize, how these groups are structured and interconnected, the extent to which they specialize or diversify into multiple markets, or how the vast sums of money involved in these activities flow across hands. Much of this activity remains invisible. Homicides and corruption related to organized crime can be observed when uncovered, but other criminal activities are harder to observe. However, what is known from cases explored by journalists, researchers, and investigative units in state agencies is enough to understand that addressing organized crime is an urgent priority for development.

There Are Four Core Characteristics of Organized Crime in LAC

Territorial control. Territorial control is a central characterizing element of how organized crime groups operate in the region. These groups control territories that range from a few blocks in urban contexts to large rural areas, where they monopolize illicit (and sometimes licit) markets through coercion and violence. Large parts of the territory in several LAC countries are controlled by criminal groups, notably Colombia, Mexico, Brazil, Ecuador, and República Bolivariana de Venezuela, but not exclusively. Indeed, 17 LAC countries rank among the top 50 in the GI-TOC index for having “mafia-style” groups characterized, among other things, by the control they exercise over the territory in which they operate.

In many cases, achieving territorial control involves high initial levels of lethal violence. This type of violence is most often related to disputes between groups over territories or markets, so once control is consolidated, it tends to decrease, and the violence that remains is targeted to punish deviations from agreements or send signals to potential competitors. There are places in which women and their bodies have been strategically targeted as a way to control communities.11 In Colombia, the ceasefire and disarmament of the rebel group FARC under the 2016 peace agreement, combined with a lack of state presence in former FARC territories, resulted in organized crime groups once again fighting for dominance. This led to systematic killings of local leaders in disputed areas—a practice that has persisted to this day.12

Criminal governance. In the territories they control, criminal organizations dictate the rules of the game. They provide services to the community, substituting for the government, particularly in security and justice, often responding to citizens’ demand. Sometimes, they limit the freedom of citizens, subjecting their behavior and mobility to their rules. For example, in Puerto Leguizamo, a small port town on Putumayo River, on the triple border of Colombia, Ecuador, and Peru, the group Comandos de la Frontera (Colombia), an ally of Comando Vermelho (Brazil) and Ecuadorian groups in the cocaine trade, dictates the hours in which civilians may transit the river and who they may carry on their boats, prohibits their consumption of alcohol, and punishes noncompliance with sanctions that range from shoveling dirt to torture and murder. Under this group’s governance, there are no more drug addicts in Puerto Leguizamo and no more speedboat thefts.13

Organized crime is a “competitor” of the state in many places. In others, however, criminal governance coexists with the state in a mutually convenient equilibrium (sometimes stable, sometimes not). For example, it is not unusual for the police to indirectly delegate control of certain criminal activities to these groups or allow them to dominate certain territories. This is the case in some neighborhoods in Medellín (Colombia), where locals are known to be instructed by the police to resort to the criminal group in control of their neighborhood in case of theft.14 Another example is Lara (Venezuela), where Colectivos controls public transportation, water pipes, and gas stations operating under government-subsidized prices. They charge public buses a fee to enter social housing projects to pick up passengers, decide which streets and houses get water, and charge drivers for letting them skip long queues at these gas stations.15

Extortion. Collecting “taxes” from businesses is a widespread practice of organized crime groups in the territories under their control. Extortion is often disguised as a sale of protection services, local justice, or even social assistance. It is, however, a protection against the threat of violence that the groups themselves can exercise against those who do not pay for their services. Extortion falls on top of the taxes established by governments, so many businesses that pay formal taxes are subject to double taxation (to the detriment of their ability to compete and invest) if they pay formal taxes. Extortion affects smaller businesses more than larger ones. A paradigmatic case is El Salvador, where MS-13 and Barrio 18 were involved in extortion and protection racketeering throughout the country for decades. Approximately

11 Insight Crime (2023).

12 Prem et al. (2021).

13 Vélez (2025).

14 Blattman et al. (2024c).

15 Venezuela Investigative Unit (2023).

79 percent of businesses, including high-end restaurants and shopping malls, paid extortion fees. The process often began with a local youth connecting the business owner to a criminal leader, who called from prison to discuss “la renta”—a protection fee. The total cost of extortion in El Salvador, including lost income from household security expenses and people deterred from working, was estimated at 16 percent of GDP in 2014.16 In Ecuador, extortion cases increased by more than 65 percent from 2022 to 2023 and have nearly quadrupled since 2021. Businesses are being forced to pay extortion fees in areas controlled by three major drug-trafficking groups: Choneros, Lobos, and Tiguerones.17 In 2023, a grenade was thrown at a school in Guayaquil following a $10,000 extortion demand, and a small pharmacy in Pascuales was bombed after the owner refused to pay a $5,000 protection fee.18

State capture. Criminal activity is made possible by the ability of criminal groups to manipulate state actors at the national and subnational levels through a perverse combination of coercion and bribes, outright infiltration of state institutions, and, in some cases, control over electoral processes through campaign financing or elimination of candidates by murder or pressure to drop out. In several prominent cases, criminals have occupied high-level government positions. Perhaps even more worrisome for democracy and the rule of law, there is the long chain of murders of candidates for office, ranging from the assassination of Ecuadorian presidential candidate Fernando Villavicencio in 2023, just days before the election, to those of the ten candidates for local office in Chiapas and Michoacan during the 2024 elections in Mexico. Organized crime has become a parallel power at the local level across parts of Brazil, Colombia, Mexico, and Ecuador.

These four characteristics apply in varying degrees to the different organized crime groups in the region. These groups differ in terms of their military capacity, the extent to which they govern the territories under their control, and the form of their criminal governance.19 Not all criminal groups provide services, regulate citizens’ behavior, or control elections in the territories under their control. They also vary in sophistication based on the level of diversification of their business portfolios, the profitability of their businesses, and the degree to which they are interconnected with national and transnational criminal networks. For example, compared to gangs in Medellín (Colombia) or Rio de Janeiro (Brazil), Salvadoran “maras” were less organized, less lucrative, less in control of local violence, and less connected with international networks of organized crime.20

Organized Crime and Violence Obstruct the Path to Development

LAC has long been hobbled by mediocre annual average economic growth, low productivity and poor productivity dynamics, high levels of inequality, and high numbers of people struggling to stay out of poverty. This report argues that the region will remain trapped in this poor equilibrium until it figures out how to contain the worst expressions of organized crime, including the violence it brings along.

There are several channels through which organized crime trumps development.

Reducing and distorting private investment. Uncertainty about property rights resulting from criminal violence and extortion reduces and distorts investment. When there is a chance that criminal activity will impede investors or businesses from recovering returns on their investments, private investments—local and foreign—do not occur or occur in suboptimal ways by type and location.21 Criminal taxes, security, and insurance expenses increase business transaction costs, reducing competitiveness and future investment capacity.22 In some countries, segments of the territory are entirely out of reach for legal business due to the presence of criminal organizations, with immense costs in terms of lost profits. Extortion and protection racketeering may also impair compliance with formal taxes,

16 Economist (2016).

17 Austin (2023); Economist (2024a).

18 Austin (2023); El Universo (2021).

19 Magaloni, Franco-Vivanco, and Melo (2020).

20 Blattman (2024).

21 Bernal et al. (2024).

22 Detotto and Otranto (2010).

particularly for smaller businesses with less space to absorb these costs. And corruption may alter the composition of foreign investment, as investors avoid placing resources in a country where corrupt practices are pervasive or choose to partner with locals to navigate bureaucracy if corruption is not prohibitive.23

Diverting public resources toward unproductive uses. Large shares of national and local government budgets must be devoted to unproductive security and defense expenses to fight crime and violence. The International Monetary Fund (IMF) recently estimated the fiscal cost associated with expenses on public security at 1.9 percent of GDP, on average, for seven LAC countries between 2018 and 2022, counting only national government expenses.24 In the absence of crime and violence, a portion of these resources could be devoted to development investments—education, health, infrastructure, and other state services.

Destroying human, physical, and natural capital. Criminal governance and violence affect human capital formation and productivity. Victims often see their income-generating capacity, learning ability, or labor productivity compromised.25 This can occur through direct victimization or by altering schooling choices and opportunities. Violence can also increase mortality and disability rates by disrupting health care services. Crimes targeting assets or property are the most direct form of depletion of physical capital. Finally, drug trafficking, illegal mining, and flora and fauna crimes are responsible for the depletion of natural capital through deforestation and water contamination, at a large cost to future generations (box 2.1).

Weakening institutions and the quality of government. Organized crime also erodes social capital, threatens democratic institutions, and hinders the provision of public goods in victimized communities. 26 Under criminal governance, people can lose the right to political participation. 27 Criminal groups hamper political representation, democratic institutions, and the rule of law in the areas under their control. 28 Interference in electoral processes reduces political competition and leads to state capture, corruption, and weakened local governments.29

Deepening inequalities. Organized crime and criminal violence disproportionally affect segments of the population already at a disadvantage—those living in the poorest urban neighborhoods, rural areas with limited state presence and strong illicit economies, and border communities. 30 While wealthier people can be targets of extortion and kidnapping, they have more resources to protect themselves. People living in poverty, the less educated, and ethnic and gender minorities are more likely victims of homicide and non-lethal crimes that impact their rights, health— physical and mental, and capacity to accumulate human capital. 31 When violence occurs in contexts of high inequality, it contributes to exacerbating and perpetuating it.32

All these channels add up to significant costs, many of which are hard to quantify. The Inter-American Development Bank (IDB) estimates direct human capital losses and public and private security expenses in LAC in 2022 at 3.4 percent of GDP.33 These are the most recent estimates available of the direct costs of crime. In 2010, the World Bank reported a higher figure of 7.7 percent of GDP for only Central America, considering material and health costs in addition to security expenses.34

23 Arcia (2012); Gaviria (2002).

24 Bisca et al. (2024).

25 Justino (2016); Prem, Vargas, and Namen (2023).

26 Arjona (2021); UNDP (2013).

27 In some contexts, criminal groups establish an order that may facilitate certain forms of civic engagement, transforming the nature of political participation rather than eliminating it.

28 Arias and Barnes (2017); Arjona (2016); Arjona, Kasfir, and Mampilly (2015); Lessing (2021); Mampilly (2012).

29 Albarracín (2018); Arjona, Chacón, and García-Montoya (2025); Blanco and Ruiz (2013); Carreras (2013); Ceobanu, Wood, and Ribeiro (2011); Gallego (2018); Krause (2014); Malone (2010); Pérez (2003); Ponce (2019); Visconti (2020).

30 Arjona and Kalyvas (2012); GMH (2013); Gómez Gallego, Herrera Vergara, and Pinilla Pinilla (2010); Yashar (2018).

31 Doran (2017): Soares (2006).

32 Amnesty International (2016); Cerqueira et al. (2019); ECLAC (2018); González (2019); IACHR (2015); International Crisis Group (2020); Wallace (2020).

33 Perez-Vincent et al. (2024).

34 World Bank (2010) based on Acevedo (2008).

BOX 2.1.

an Overview of environmental Crimes in the Brazilian amazona

The Amazon rainforest is facing an escalating crisis of environmental crimes, including illegal deforestation, mining, logging, wildlife trafficking, land grabbing, and biopiracy. These illicit activities threaten the rich biodiversity of the rainforest and the survival of traditional populations who depend on it. Data from 2024 reveal an alarming increase in environmental crimes within the Legal Amazon. Between the beginning of August 2023 and the end of September 2024, the Brazilian Federal Highway Police (PRF) recorded an 88% increase in the occurrence of environmental crimes in the Legal Amazon compared to the previous 14 months.b The Amazon Institute has pointed out that forest degradation in the nine states that comprise the region is the highest in the last 15 years. Studies by the Institute for Amazon Environmental Research (IPAM) and the World Bank emphasize the urgent need for effective interventions against deforestation.

Illegal mining operations contaminate vital water sources and soil with toxic heavy metals, posing severe health risks to both human and aquatic life. Land grabbing and illegal resource exploitation fuel conflicts between traditional communities, illegal settlers, and powerful logging and mining interests, destabilizing the region. The expansion of sophisticated criminal organizations, notably Comando Vermelho and Primeiro Comando da Capital, into the Amazon region adds a new and concerning dimension to the environmental crime landscape.c These factions exploit the region’s vast, sparsely monitored territory and weak law enforcement presence to finance their operations through environmental crimes.

The intricate connection between drug trafficking and environmental crimes is increasingly apparent, with these organizations using established routes and infrastructure to transport both narcotics and illegally sourced natural resources.d

The World Bank recognizes the critical challenge of environmental crimes in the Amazon. It is assessing its potential contributions while advancing initial actions, including mapping the efforts of various agencies in the region, starting with the Brazilian state of Amazonas. The primary objective is to identify existing threats, operational deficiencies, and systemic weaknesses that impede the effective enforcement of environmental laws. The implementation of the Environmental and Natural Resources Crimes Risk Assessment methodology, alongside the strengthening of cooperation with key organizations such as the Amazon Cooperation Treaty Organization (ACTO) and the Igarapé Institute, and the integration of these efforts with existing World Bank programs, are pivotal components of this strategy.

Notes

a. This box is based on text contributed by Erwin De Nys, Practice Manager of the World Bank’s Climate Finance Mobilization Unit.

b. Agência Brasil (2024).

c. O Globo (2024).

d. Amazon Watch et al. (2023).

Why Is Organized Crime Expanding in LAC?

It isn’t easy to determine what is behind the escalation of organized crime in the region. However, a series of overlapping factors may be contributing.

• During the 2010s, global demand for cocaine, illegal gold, and migrant smuggling boomed.35 Improvements in the quality of cocaine, paired with stable international cocaine prices (since at least 2015), resulted in lower real prices.36 Soaring demand resulted in new entrants into the business. For example, Primeiro Commando da Capita (PCC) (Brazil), whose initial income source was the micro-trafficking of drugs in prisons and then locally in São Paulo, is now involved in moving cocaine shipments abroad, mainly to Europe, becoming a large-scale international distributor.37 Rising commodity prices drove increases in illegal mining.38 Rents from migrant smuggling, fueled by global migration to the United States across the Mexican border, gave groups increased revenues to invest in other businesses.39

• Government crackdowns after 2005 resulted in a reorganization of LAC crime groups and a new landscape of actors competing for influence.40 This, combined with record-breaking drug production and trafficking, altered the geography of lethal violence, as criminal groups used force to protect plantations, transshipment routes, and retail outlets.41 There is evidence that violence in Mexico escalated significantly as Colombia stepped up its counter-narcotics operations and drug seizures between 2006 and 2009.42

• The COVID-19 pandemic created new opportunities and challenges for organized crime worldwide.43 In Colombia and Mexico, the pandemic enabled criminal organizations to gain legitimacy and power by providing social support and governance, sometimes replacing the state in areas with deep-rooted poverty and inequality.44 In response to border closures during the pandemic, Tren de Aragua (Venezuela) evolved from charging smugglers fees to establishing its own comprehensive smuggling operation, including owning transport companies and hostels, thereby profiting significantly from migrant exploitation throughout the continent.45 Sudden drops in income make recruiting people for crime easier, which may have also played a role in the expansion of organized crime after the pandemic.46

• Expanded production and availability of arms have facilitated the growth of organized crime into new areas and markets.47 More than two-thirds of homicides in Latin America involved firearms in 2023, and the share is even higher in the Caribbean, at 78 percent—the highest in the world.48 Despite strict firearm controls across LAC, about 253,000 firearms are trafficked into the region each year.49 Experts assert that national efforts to reduce organized crime will be insufficient unless these flows are addressed.50

35 Economist (2023b); IOM (2021); UNODC (2024).

36 UNODC (2024).

37 International Institute for Strategic Studies (2024).

38 Idrobo, Mejía, and Tribin (2014).

39 Financial Times (2024).

40 Calderón, Robles, and Magaloni (2015); Dell (2015); Guerrero (2011).

41 UNODC (2023a).

42 Castillo, Mejía, and Restrepo (2020).

43 Aziani et al. (2023).

44 Tamayo Gomez (2020).

45 Economist (2023c).

46 Blattman (2023); Chassang and i Miquel (2009).

47 Aguirre and Muggah (2020); Pérez Ricart (2024); Solmirano (2023).

48 World Bank staff calculations based on data from UNODC. For some countries, the latest available data point is 2021 or 2022.

49 Dube, Dube, and García-Ponce (2013); McDougal et al. (2013); Pérez Esparza and Weigend (2015); Pérez Ricart (2021); US Bureau of Alcohol, Tobacco, Firearms, and Explosives (2023); Weigend Vargas et al. (2023); Pérez Ricart (2024).

50 The US Bureau of Alcohol, Tobacco, Firearms, and Explosives (ATF) reports that about 68 percent of guns at Mexican crime scenes come from the United States. In Central America, 55 percent of traced firearms come from the United States; in some Caribbean countries, this figure can reach up to 99 percent, as in the Bahamas (US, Bureau of Alcohol, Tobacco, Firearms, and Explosives, 2022a, 2022b).

• Organized crime groups have become increasingly transnational. The days in which a single group—the Medellín Cartel—controlled the production and distribution of refined cocaine in the US market are over. Today, multiple criminal networks collaborate globally to manage complex supply chains and millions of dollars, and their success depends on working with other criminal organizations and corrupt state officials across borders. Relatedly, the presence of foreign criminal actors is growing in LAC countries. Eleven LAC countries rank among the top 50 by their GI-TOC score for the presence of foreign criminal actors as of 2023, a measure pointing precisely to the international interconnectedness of organized crime businesses. It is not rare to hear about killings or captures involving members of recognized European organized crime groups in Latin American cities.51

• Organized crime groups are increasingly diversifying their criminal portfolios. Diversification enables them to absorb losses. Money laundering and penetration into the legal economy belong in this category.52

• Technological innovation, particularly in digital technologies, broadens the portfolio of criminal businesses and has contributed to improving the efficiency of criminal operations. For example, advanced navigation technologies allow organized crime to create more efficient maritime routes. 53 Technological advances have opened the way to cybercrime and hybrid organized crime groups in which humans and machines collaborate closely.54 Cryptocurrencies, increasingly used in money laundering because they are hard to track, also fall in this category of technological innovation.

Organized Crime Is Much More Lethal in LAC Than in Other Regions

Measured by Homicides, LAC Suffers the Highest Levels of Violence Worldwide

Homicides are not the only manifestation of violence. However, the homicide rate is the most reliable and internationally comparable statistic of violence.55 By this measure, violence in LAC is incomparably high. While accounting for approximately 9 percent of the global population, LAC records one-third of all homicides. 56 Moreover, the gap between the homicide rate in LAC and the rest of the world has widened over the last 20 years. In the first decade of this century, the average homicide rate in LAC was 5.4 times higher than the world’s (22.0 versus 4.1); in the second decade, it was 8 times higher (23.9 versus 3.0) (figure 2.4). The LAC average is informative of the entire region: every subregion of LAC suffers violence. The homicide rates of South America, Central America, and the Caribbean all far exceed those observed elsewhere in the world.

Homicides per 100,000 people, LAC and rest of the world, 2000–09 and 2010–19

Homicides per 100,000 population LAC Rest of the world

Source: World Bank staff calculations based on United Nations Office on Drugs and Crime (UNODC) data.

Note: Regional aggregates are calculated as simple averages across countries. To avoid comparing different sets of countries, missing data were interpolated for up to five consecutive years. Countries with more than five consecutive missing observations (Bolivia, Peru, Haiti) are excluded from the LAC averages. LAC = Latin America and the Caribbean.

51 GI-TOC (no date); Saiz (2024).

52 McDermott (2020).

53 UNODC (2023b).

54 Chatterjee (2005); Di Nicola (2022); Grabosky (2007).

55 Homicide rates measure the number of homicides relative to the population of a geographic unit over a specific period.

56 Reporting rates for non-homicide violent crimes (such as robbery, assault, and rape) are usually low and vary significantly depending on the type of crime and individual characteristics (Soares 2004).

Homicide rates in the region in the last available five-year period (2018–22) exceed what the countries’ development level would predict—they are high relative to countries with similar per capita income (figure 2.5, panel a) and inequality levels (figure 2.5, panel b).

2.5.

laC Countries suffer more lethal Violence than Others with similar development and inequality levels

Homicides per 100,000 people versus GDP per capita, 2018–22 average

GDP per capita PPP (constant 2021 international $)

Homicides per 100,000 people versus income inequality, 2018–22 average

index

Source: World Bank staff calculations based on World Development Indicators and United Nations Office on Drugs and Crime (UNODC) data.

Note: LAC countries are shown in blue. Data labels use International Organization for Standardization (ISO) country codes. LAC = Latin America and the Caribbean; PPP = purchasing power parity.

The LAC average hides substantial variation across countries in the intensity of violence measured by homicides.57 The 2018–2022 average homicide rates vary from 49 homicides per 100,000 people in Jamaica and 38 in Honduras (at the top) to 5 in Argentina and 4 in Bolivia (at the bottom) (figure 2.6). The trends in lethal violence also differ significantly across countries. The last available country-year measure of the homicide rate (yellow dot in figure 2.6) shows significant improvement in some countries in recent years, notably El Salvador and Venezuela, while the problem in others has worsened considerably, especially Ecuador and Haiti.

57 Violence is, in fact, a localized phenomenon that varies dramatically within countries, cities, neighborhoods, and even street

FigUre 2.6.

the Homicide rates Varies greatly across laC Countries

Average homicide rate, LAC and comparators, 2018–22 and last year available

2018-22 average Last year available

Source: World Bank staff calculation based on the United Nations Office on Drugs and Crime (UNODC) data.

Note: The dotted line is the LAC 2018–22 simple average. Yellow dots show the last available figure for each country. The figure uses International Organization for Standardization (ISO) country codes. EUR = Europe; LAC = Latin America and the Caribbean.

Crime Victimization is More Evenly Distributed than Homicides across Countries

Harmonized surveys provide an alternative approach to measuring crime and violence that is comparable across countries. One-third of LAC respondents to the World Values Survey (WVS) state that they or a family member were the victims of a crime in the past year. This figure is three times larger than the average for the rest of the world. These levels (and the implied gap) have remained largely unchanged in the last 15 years. 58 Survey-based victimization rates capture both lethal and non-lethal violent crimes as well as property crimes, so they are informative measures of general patterns of crime.

The WVS does not allow comparisons of victimization rates across LAC countries in the same years because only a subset of countries enter each survey wave. However, another public opinion poll (Latinobarómetro) includes a similar victimization question, allowing us to consistently compare a larger sample of LAC countries over time (figure 2.7).

FigUre 2.7.

Victimization is more evenly distributed than Homicides in laC Victimization rates, LAC countries, 2000–23 average and last year available

average Last year available 0

Source: World Bank staff calculations based on Latinobarometro.

Note: The dotted line is the simple average of all LAC countries. The last year available is 2023, except for Nicaragua, for which it is 2020. The figure uses International Organization for Standardization (ISO) country codes. LAC = Latin America and the Caribbean.

58 Because not all countries in the WVS appear in every wave, the average victimization rates in these calculations are based on a reduced sample of the countries that show up in both periods. Specifically, the period from 2010 to 2014 is compared with the period from 2017 to 2023. Each country in the sample appears in both time frames at least once.

Victimization is more evenly distributed than homicides across the region, except for very high levels in Mexico and very low levels in Panama. This means that levels of victimization are high even in countries where homicide rates are relatively low (such as Argentina and Peru).

Lethal Violence from Organized Crime is Predominant in Today’s LAC

Most non-LAC countries ranking among the top 50 by their criminality score have homicide rates under 10 per 100,000 people. The exceptions are Nigeria, South Africa, and South Sudan. In contrast, all LAC countries in the same group, except Paraguay and Peru, have homicide rates exceeding 10 per 100,000 people, and seven countries have homicide rates exceeding 20 per 100,000 people. These figures suggest that organized crime is more lethal in LAC than in most other places, prompting the question of what else in LAC is different.

Figure 2.8 plots the “excess” level of homicides given what would be predicted by the reported level of victimization for all countries in the WVS sample.59 LAC, by far, has the highest average levels of excess lethal violence (horizontal lines), followed by Sub-Saharan Africa, and Europe and Central Asia. Within LAC, Venezuela, Colombia, Brazil, Mexico, and Guatemala are the countries with more excess lethal violence. In contrast, countries like Chile, Bolivia, Peru, and Argentina report lower homicide rates relative to their general levels of crime. This excess lethal violence measure is highly correlated with the GI-TOC criminality score (0.63) in LAC. The correlation is lower globally (0.34), confirming there is something particular about organized crime and violence in LAC.

FigUre 2.8.

laC is the world region with the Highest excess lethal Criminal Violence

Residuals of the homicide rate against victimization rate (averaged 2017–22)

Latin America and the Caribbean Sub-Saharan Africa Europe and Central Asia Middle East and North Africa East Asia and Paci c North America South Asia

Source: World Bank staff calculations based on the United Nations Office on Drugs and Crime (UNODC) and World Value Surveys. Note: The figure uses International Organization for Standardization (ISO) country codes. LAC = Latin America and the Caribbean.

The

are

similar if, instead of the regression residuals, the ratio of homicides to victimization rates is computed. The correlation of the country-level rankings between the two measures is 0.79.

Weak Institutions in Critical Areas Fuel the Problem Instead of Mitigating It

The fact that organized crime results in more lethal violence in LAC than in other places may be explained in part by persistent disputes across organizations for territorial and market control, but it also indicates state weakness in the capacity to respond. Entirely eradicating organized crime is unlikely. But improved state action can help “manage” organized crime and address its more harmful manifestations, as has probably been the case in other places where organized crime presence does not have a high correlation with lethal violence and seems to have a smaller impact on economic development.

LAC countries stand out for having poorly working labor markets and large segments of society that lack the opportunities to pursue lives they can value. They also stand out for weaknesses across the three key pillars of security policy: prisons, police forces, and justice systems. The discussion that follows shows that the region has serious problems in all three dimensions and highlights urgent policy interventions that governments must undertake to improve their law enforcement capacity to fight the scourge of organized crime.

LAC Prisons Must Be Transformed to Fulfill Their Objectives60

Prisons should contribute to reducing crime through incapacitation, deterrence, and resocialization.61 Incapacitation refers to removing offenders from society so they cannot commit crimes while imprisoned. Deterrence involves discouraging potential criminal behavior through the threat of imprisonment. Resocialization aims to improve offenders’ outcomes after they have been released from prison. LAC prison systems fail on all these fronts.

Incarceration levels in the region are higher than in most of the world (figure 2.9). An exception among high-income countries is the United States, where imprisonment rates are historically high. Two extreme cases in the region are El Salvador and Haiti, with significantly high and low imprisonment rates, respectively. High incarceration does not always correspond to lower crime.

laC Countries Have relatively High incarceration rates

People held in prison per 100,000 people, 2021 or 2022

People held in prison per 100,000 population

Source: World Bank staff calculations based on United Nations Office on Drugs and Crime (UNODC) data.

Note: LAC countries are shown in dark blue. Bars for world regions (yellow) show the median for the corresponding indicator, while light blue dots indicate the minimum and maximum values within each region. Country labels use International Organization for Standardization (ISO) country codes. LAC = Latin America and the Caribbean.

60 The discussion in this section refers to both prisons and jails under the label “prisons” because the distinctions between these terms made in the United States is neither as clearcut in LAC nor as comparable across countries in the region. The discussion also refers interchangeably to “prison system” and “penitentiary system”.

61 Becker (1968); Bhuller et al. (2020); Drago, Galbiati, and Vertova (2009); Kuziemko (2013).

The most pressing concern is that in many places, organized crime groups have taken control of prisons and manage outside crime from within.62 Criminals behind bars are no longer prevented from committing new crimes, and imprisonment is no longer a threat that deters criminals. Instead, it has become a step in a gang criminal’s career. Prisons in LAC are not only a “safe place” for gang leaders but also a field for the recruitment and training of new gang members. In Brazil, PCC and Comando Vermelho emerged in prisons as a strategy for inmates to defend themselves from prison violence. These gangs then grew in size and reach, forming transnational criminal organizations.63 Tren de Aragua also originated behind bars in Venezuela.64 The largest and deadliest Salvadoran “maras”—Barrio 18 and MS-13 were, in turn, created in prisons in Los Angeles. Other infamous gangs do not originate inside prisons, but their imprisoned leaders preserve and command outside operations and negotiate and regulate operations with imprisoned leaders from competing criminal groups.65

Authorities in LAC must regain control of prisons by interrupting cellular and other forms of communication, subjecting gang leaders to stricter controls in high-security facilities, and improving the allocation of prisoners to prevent the exposure of minor criminals to members of criminal groups. Of course, these interventions are challenging because prison crackdowns can instigate both internal and external violence. For example, in Rosario (Argentina), the hardening of prison security conditions in March 2024 triggered a series of random assassinations in the streets that were ordered by the imprisoned leaders of the Los Monos gang to protest the operation. However, the authorities responded to this violence, and the homicide rate fell rapidly (see box 2.2).

LAC prison systems also stand out for large numbers of unsentenced prisoners, harsh conditions, and overcrowding (figure 2.10). Out of 24 LAC countries with available data, six have more than twice as many prisoners as capacity, and only three operate under capacity. On average, almost half of those in prison are unsentenced. El Salvador is an extreme case again, with only 23 percent of prisoners reportedly sentenced.

A second priority is improving prison conditions. By providing inmates with better infrastructure and services, lower occupancy levels, and higher guard-to-inmate ratios, some regions of Colombia saw a 36 percent reduction in recidivism rates.66 Similar results have been observed in the United States.67 Improving prison conditions also involves implementing resocialization programs for inmates.68

To reduce overcrowding, it is also critical to consider alternatives to incarceration. These alternatives could be particularly valuable to spare young and low-level offenders from a deleterious prison experience under the influence of criminal organizations.69 Electronic monitoring has proven to be a cost-effective alternative to prison in Argentina, reducing recidivism rates by approximately 50 percent.70 Reducing pre-trial detention and the length of sentences are also approaches that may reduce recidivism.71 Restorative justice and behavioral therapies have also demonstrated promising results.72 Most of these are low-cost alternatives and thus potentially cost-effective relative to standard incarceration.

Finally, the world has been observing the Salvadoran government’s “mano dura” (firm hand) strategy put in place in 2021 to incapacitate MS-13, Barrio 18, and other criminal groups through mass incarceration.73 More than 73,000 young males have been arrested.74 The resulting drop in the homicide rate (from 18 to 2.4 per 100,000 people between

62 Schargrodsky and Tobón (2025).

63 Biondi (2016); Lessing (2010); Lessing and Willis (2019).

64 Sampó and Troncoso (2024).

65 Blattman et al. (2024a, 2024b); Kalsi (2018); Sviatschi (2022).

66 Tobón (2022).

67 Chen and Shapiro (2007).

68 Alsan et al. (2024); Davis, Meares, and Arnesen (2025).

69 Agan, Doleac, and Harvey (2023); Agan et al. (2023); Aizer and Doyle (2015); Bhuller et al. (2020); Eren and Mocan (2021); Kleinberg et al. (2018); Kuziemko and Levitt (2004).

70 Di Tella and Schargrodsky (2013). These findings were replicated in the United States and Australia (Grenet, Grönqvist and Niknami 2024; Williams and Weatherburn 2022).

71 Donohue and Wolfers (2006); Estelle and Phillips (2018); Polinsky and Riskind (2019); Rose (2021); Rose and Shem-Tov (2021).

72 Shem-Tov, Raphael, and Skog (2024).

73 The crackdown was possible under a state of emergency declaration that suspended constitutional rights to fight these groups (Human Rights Watch 2024). Originally set to last 30 days, the state of emergency has been extended 34 times so far (Economist 2023a).

74 Human Rights Watch (2024).

FigUre 2.10.

10 laC prisons are Overcrowded, and many prisoners are Unsentenced

a. Share of people in prison who are sentenced, 2021 or 2022

b. Ratio of people held in prison to official prison capacity, 2021 or 2022

People held in prison / o cial capacity of the prison system

Source: World Bank staff calculations based on United Nations Office on Drugs and Crime (UNODC) data.

Note: LAC countries are shown in dark blue. Bars for world regions (in yellow) show the median country in the region, and light blue dots indicate the minimum and maximum values within each region. The “share sentenced” is calculated as the number of reported sentenced prisoners divided by the total prison population. The sentenced share figure for El Salvador is from the World Prison Studies due to incomplete data from UNODC. Data for prison capacity in the United States are from 2019, and for Haiti, they are from 2015.

LAC = Latin America and the Caribbean.

2021 and 2023) and the regaining of state territorial control translated into strong citizen support.75 The strategy, however, has been questioned for disregarding the separation of powers in government and due process.76 In other LAC countries, even short-term results from mano dura approaches have proven strongly context-specific. Some have strengthened criminal organizations’ prison control and extended their influence across borders by forcing their fragmentation and relocation.77

75 A previous reduction in lethal violence, from 105 to 20 homicides per 100,000 people between 2015 and 2020, had resulted from an agreement with MS-13 and Barrio 18 to decrease violence in exchange for improved prison conditions and early release. Although the truce succeeded in containing lethal violence, it increased extortion (Blattman 2024; Dudley 2020; Insight Crime 2024; Meléndez-Sánchez and Winter 2024; Papadovassilakis 2023). Despite a constitutional ban on immediate re-election, President Nayib Bukele was re-elected on February 4, 2024, winning by a large margin. The government’s approval ratings range between 80 percent and 90 percent, the highest in Latin America. Surveys show that individuals in other LAC countries may view President Bukele even more favorably than Salvadoreans (Economist 2023a, 2024a).

76 Amnesty International (2022); Cristosal Human Rights (2023); Economist (2023a); Human Rights Watch (2024).

77 Akee et al. (2014); Blattman (2024); Davenport (2007); Golcalves et al. (no date); Lessing (2013); Meléndez-Sánchez and Winter (2024).

BOX 2.2.

the rosario prison intervention in 2024a

Since 2013, violent crime has increased significantly in the city of Rosario, Argentina. This phenomenon is almost entirely related to drug trafficking, with most murders associated with territorial disputes between Rosario’s largest gang—Los Monos—and rival organizations. Despite the incarceration of most of Los Monos’ leaders, the murder rate kept increasing, reaching 22 per 100,000 inhabitants by 2023 (five times Argentina’s rate). Corruption within the prison system allowed imprisoned gang leaders to carry on their illicit activities outside. For example, in 2022, Los Monos performed a drive-by shooting on the house of a federal judge who had ordered the prison transfer of one of the criminal leaders.

In early 2024, the federal and provincial authorities launched Operativo Bandera, which included a severe tightening of the prison conditions of gang leaders, enforcing the prohibition on the use of cellphones, implementing sudden searches, reallocating prisoners, grouping members of each gang in separate wings, and controlling visits. Gangs responded by starting a series of random assassinations of civilians in the streets. However, the government persisted with the tighter imprisonment conditions for gang leaders and responded with the deployment of 1,400 additional federal security officers in the city in coordination with provincial police. Operativo Bandera had almost immediate success, lowering Rosario’s murder rate by 65 percent (figure B2.1.1).

Homicides declined sharply after the rosario prison intervention Homicides per 100,000 people, Rosario (Argentina), 2000–24

Source: Sistema Nacional de Información Criminal for 2000-23 data; Ministerio de Justicia y Seguridad de Santa Fe for 2024 data.

Notes a. This box is based on Schargrodsky and Tobón (2025).

Working

Police Forces Could Make a Big Difference, but Shaping Them Is Difficult

Police forces reduce crime through deterrence (police presence deters crime by increasing the likelihood of apprehension for would-be criminals) and incapacitation (offenders captured by the police cannot commit new crimes). As in the case of prisons, most police forces in LAC fail to fulfill their functions.

Police forces in LAC range in size from 173 per 100,000 people in Honduras to 968 per 100,000 people in Saint Vincent and the Grenadines. The region displays the widest variation in police force sizes across countries and the highest regional median in the world. However, most LAC countries are within the European range in police size per hundred people. (figure 2.11). From these numbers, it is impossible to say that LAC countries are over-policed or under-policed because factors other than size determine police effectiveness.

Police force size may, in fact, be an issue in places where crime is high and police forces are small. But there is no ideal police-to-people ratio and no reason why police personnel should be spread homogeneously across the territory. How police personnel are deployed across a country’s or a city’s geography can, instead, play a critical role. Here is an instance where better data and better processing tools, including artificial intelligence and data analytics, could improve patrolling strategies and the efficient real-time allocation of resources. Geographical strategies include specific interventions aiming to recover state territorial control, like Brazil’s Unidades de Policía Pacificadora (UPPs), which sought to reclaim control of favelas dominated by criminal groups, and El Salvador’s Plan de Control Territorial, which regained state authority over areas once controlled by gangs.78

FigUre 2.11.

the size of police Forces Varies across Countries

Police personnel per 100,000 people, LAC countries, and comparators, 2021 or last year available since 2015

Police personnel per 100,000 population

Source: World Bank staff calculations based on United Nations Office on Drugs and Crime (UNODC) data.

Note: LAC countries are shown in dark blue. Bars for world regions (yellow) show the median country, while light blue dots indicate each region’s minimum and maximum values. Data for Chile are from 2017. Brazil is not included in the figure because no information is available after 2013. LAC = Latin America and the Caribbean.

Resource allocation between patrolling and investigative activities is another dimension that matters for police effectiveness in the fight against organized crime. There is often a bias in favor of patrolling because visible police deployment responds to citizens’ demands for police presence. However, patrolling helps combat common crime but is not enough to address organized crime, which requires stronger police investigative capacities.

78 Monteiro and Weintraub (2025).

BOX 2.3.

mexico City’s High-impact Crime reduction strategya

Mexico City’s homicide rate peaked at 30 per 100,000 people in 2018 after four years of growth. A new crime reduction strategy was implemented in 2019 to reduce high-impact crimes and improve the public’s perception of security.b The strategy is still being implemented and consists of six key pillars:

1. Intelligence-based policing—the Secretariat of Citizen Security (SSC) police force was given new investigative powers, enhancing its operations with intelligence-driven, targeted actions;

2. Inter-institutional coordination under the centralized command of the Chief of Government to promote continuous collaboration between the Prosecutor’s Office, the SSC, and local boroughs, through periodic meetings;

3. Strategic allocation of police in high-crime areas for more rapid and effective response;

4. Use of advanced technological tools to optimize surveillance and incident response;

5. Prioritizing the professionalization and welfare of police personnel, resulting in a cumulative salary increase of 45 percent between 2018 and 2022, and certification of nearly 90 percent of the police personnel; and

6. Social prevention programs to tackle underlying causes.

The strategy resulted in a 49 percent decrease in high-impact crimes between July 2019 and July 2024, with daily averages dropping from 131 to 57 crimes. The perception of insecurity also fell from 92 percent in early 2018 to 52 percent by mid-2024, a 41 percent decrease. Homicides have been dropping each year since the effort began.c

Notes

a. This box is based on text contributed by Thomas Favennec from LAB-CO.

b. Mexico City, Gobierno de la Ciudad de México (2024a); Mexico City, Secretaría de Seguridad Ciudadana de la Ciudad de México (2024a).

c. Mexico City, Gobierno de la Ciudad de México (2024b); Mexico City, Secretaría de Gobierno de la Ciudad de México (2024b).

Police forces in LAC are also usually under-trained to address the challenges they face, particularly those posed by organized crime. Better training in best policing practices for both patrolling and investigative work can make a difference. There is also a need for more institutional support, including better working conditions, wellness practices, and compensation. These are necessary to prevent corruption, reduce personnel rotation, address mental health distress (common to the occupation), and foster loyalty to the state among the police.79

Another issue when considering police effectiveness in LAC is the need for coordination with other state agencies and, sometimes, other security forces. Police forces must coordinate and collaborate with other state agencies to be effective. Mexico City is a good example of a multi-pillar strategy to fight crime that rests strongly on intelligence-based policing and inter-institutional coordination under a centralized command (box 2.3). Coordination is also essential when local and national forces overlap. Moreover, in response to the growing presence of national and transnational criminal organizations, better coordination (through centralized command or other mechanisms) of police interventions inside and across LAC countries, and with countries outside the region, is necessary.

Countries in the region vary in their degree of decentralization of police forces. For instance, while Colombia has only one (national) police force distributed across the territory, Argentina has federal and provincial police forces.80 Police (de)centralization presents significant trade-offs. Local police forces may have the advantage of better local knowledge. But, if organized crime groups respond to police interventions in one area by moving their activities to another, decentralized and isolated police responses are likely insufficient. On the other hand, coordination and competition among security forces may prevent their capture by criminal organizations. However, perhaps more important than the degree of institutional decentralization is the capacity of police forces to coordinate internally across units and with other state actors, as stated.

Police effectiveness also depends on trust from the community. According to the World Value Survey (2017–22), levels of citizen trust in the police in LAC are the lowest globally. On average, 67 percent of respondents in LAC countries report having little or no trust in the police, compared to 36 percent in non-LAC countries. Distrust affects crime reporting and cooperation. Tackling the use of force, abuse, and biased enforcement (for example, the targeting of minorities) to which police forces are prone can contribute to restoring trust. 81 Preventive interventions focusing on at-risk officers seem cost-effective in reducing these behaviors. 82 Another approach shown to reduce bias is increasing diversity within police departments by recruiting underrepresented groups.83 Increased institutional support, as previously discussed, can also help.

Effective monitoring and accountability measures are crucial for improving police-community relations.84 For instance, body-worn cameras have the potential to influence policing outcomes, notably reducing fatal and nonfatal police use of force and civilian complaints against the police. From a financial standpoint, these cameras could pay for themselves by reducing investigation costs and payouts or settlements. 85 Finally, militarized policing, which treats suspects like wartime enemies, is associated with abuse and, hence, undesirable.86 Community-oriented policing efforts emerge as an alternative approach to militarized policing. However, they have yielded mixed results, and their effectiveness is strongly context-specific (box 2.4).

Justice Systems Are the Weakest Link

Most serious crimes in LAC are neither investigated nor prosecuted. Judicial systems in the region lack the capacity to investigate and process high-impact offenses properly, and in some cases, they can be captured by offenders. 87

The available indicators of the effectiveness of criminal investigation and adjudication—from the World Justice Project—suggest that except for Chile, Costa Rica, and Uruguay, Latin American countries rank the lowest relative to countries in other world regions (figure 2.12). The first indicator assesses the quality of criminal justice based on the perception and experience of whether perpetrators of crime are effectively apprehended and correctly charged. It also measures whether the police, investigators, and prosecutors have adequate resources, are free of corruption, and perform their duties competently. The second indicator measures whether perpetrators of crime are effectively prosecuted and punished and whether criminal judges and other judicial officers are competent and produce speedy decisions. It provides a closer assessment of impunity, confirming that this is a significant problem in the region.

80 Argentina has additional federal-level security forces such as the national Gendarmería and Prefectura, in charge of protecting the country’s national borders.

81 Braga, Brunson, and Drakulich (2019).

82 Abril et al. (2024); Parker, Ross, and Ross (2024); Stoddard, Fitzpatrick, and Ludwig (2024).

83 Ba et al. (2021); Cox, Cunningham, and Ortega (2024); Donohue and Levitt (2001); Harvey and Mattia (2024); Miller and Segal (2019).

84 Rivera and Ba (2023).

85 Ariel, Farrar, and Sutherland (2015); Assaraf et al. (2024); Barbosa et al. (2021); Braga et al. (2017); Braga, MacDonald, and McCabe (2022, 2023); Fagundes, Monteiro, and Souza (2023); Lum et al. (2019); Mancha, Monteiro, and Weintraub (no date); Munyo and Rossi (2020); Williams et al. (2021).

86 Magaloni and Rodriguez (2020).

87 This discussion refers to justice systems as all public authorities that (1) manage and solve disputes between citizens, private organizations, and state agencies—including formal and informal dispute resolution mechanisms; and (2) investigate, prosecute, and convict criminal offenses. This broad characterization includes different types of public authorities, such as formal judicial systems, criminal investigation agencies, and prosecution services, as well as informal dispute resolution mechanisms such as conciliation, mediation, and other restorative instruments.

BOX 2.4.

rio de Janeiro’s Community-Oriented policing strategya

In 2008, in Rio de Janeiro, the state launched Pacifying Police Units (known as UPPs for their acronym in Portuguese) to reclaim territory from criminal groups. The initiative aimed to move away from the “militarized” policing of favelas, the low-income urban areas housing about 20 percent of Rio’s population controlled by gangs. Young officers trained in human rights and community-oriented policing were enlisted to foster better relations with favela residents and reduce armed confrontations. Implementation varied by leadership style, including offering soccer and karate classes for kids, conflict resolution, and community meetings to discuss security. UPPs with more than 10,000 police officers were deployed in about 160 favelas.

UPPs reduced fatal police shootings by 45 percent compared to previous militarized policing. However, their impact varied with the military capacity of local criminal groups and the prevailing form of local governance. Where criminal groups had strong military power and offered civilian benefits for support, UPP interventions turned communities into war zones. Where they had lower military capacity and harsher governance, UPPs improved local security conditions. Where they shared resources from illicit markets with state associates while providing local security and assistance to civilians, UPP crackdowns disrupted this order. Where they exploited communities through illicit associations with state agents, engaging in harmful activities like human trafficking and kidnapping under law enforcement protection, by breaking these illicit pacts UPPs improved local security and human rights. Where rival groups engaged in constant conflict over territorial control, UPPs played a crucial role in resolving violent situations, often acting as key actors to bring order.

A takeaway from this experience is that the state’s ability to regain territorial control also depends on police behavior. Effective territorial control shifts when communities view police as legitimate and not as criminal themselves. However, if groups maintain order better than police and residents fear law enforcement more than criminal bosses, they may prefer criminal organizations.

Notes

a. This box is based on Magaloni, Franco-Vivanco, and Melo (2020).

High levels of impunity for serious crimes are confirmed by most measures of criminal justice performance in those LAC countries experiencing more crime and violence. In this context, the ineffectiveness of the justice system, combined with fragile public institutions and weak, nonstrategic security forces, contributes to a vicious cycle that interconnects and intensifies broad social, economic, and political disadvantages; powerful criminal organizations and other illegal groups; and large, persistent illegal markets.

The region’s weak judicial systems imply that countries lack the capacity to detect, investigate, and dismantle complex criminal organizations with multiple actors and layers or, at the very least, to punish the criminals and activities that generate the worst welfare externalities for the region’s citizens.

The lack of judicial capacity may partly be explained by insufficient performance indicators that would provide the correct incentives to all actors in the law enforcement and criminal justice chain. For example, in cases related to organized crime activities, the focus on short-term operational results, such as capturing and prosecuting the weakest links of an organization, is not the best measure of strategic success—especially if such operations hurt more ambitious investigation efforts that could lead to capturing and prosecuting the upper tiers of the chain of command. Similarly, indiscriminate repressive actions by law enforcement authorities may lead to unanticipated increases in

FigUre 2.12.

most latin american Countries rank poorly on the effectiveness of their Criminal Justice systems

Criminal Justice Indicators, 2023

a. Criminal investigation system is effective [0–1]

Criminal investigation is e ective [0–1]

b. Criminal adjudication system is timely and effective [0–1]

Criminal adjudication system is timely and e ective [0–1]

Source: World Justice Project Rule of Law Index 2023.

Note: The indexes are based on data from general population polls and qualified respondents’ questionnaires, completed by criminal law experts. Scores range from 0 to 1, with 1 indicating a more effective system. LAC countries are shown in dark blue. EAP = East Asia and Pacific; ECA = Europe and Central Asia; EU = European Union; LA = Latin America; LAC = Latin America and the Caribbean; MENA = Middle East and North Africa; NAC = North America; SSA = Sub-Saharan Africa.

violence and prove counterproductive to the safeguarding of civilians in certain regions.88 Indeed, the availability of appropriate criminal justice performance indicators can improve accountability and public trust while providing the correct incentives for strategic (and not only operative) effectiveness (box 2.5).

To improve criminal justice capacity, specifically in fighting organized crime, authorities must turn to prioritization, which entails focusing available resources on investigating and eliminating those crimes that are more harmful to society. Examples of such crimes include homicidal violence, child prostitution, and extortion.89

Judicial authorities are also central to efforts to reduce incarceration. Reforming eligibility conditions for focused law enforcement and imprisonment in national and subnational-level penal codes requires the involvement of judicial authorities, as well as broader discussions on drug policy. Progressing in the conversation about the regulation of drug markets can alleviate LAC prison overcrowding and free resources to target the central structures of criminal organizations more efficiently.90

88

89

90

BOX 2.5.

Justice indicators: definition and purposea

Justice indicators are measurable metrics that assess various aspects of the justice system, helping identify deficiencies, monitor reforms, ensure transparency, and align operational and strategic objectives. a They play crucial roles in measuring performance, increasing accountability, and improving resource allocation, establishing benchmarks that support strategic planning and evidence-based policy making.

These indicators can be classified into three categories: input indicators evaluate resources and reform progress; output indicators assess institutional efficiency; and outcome indicators measure broader objectives such as public trust and crime reduction. For justice indicators to be effective, they must align with justice system goals. They thus require a high level of strategic awareness from law enforcement and criminal justice authorities to measure and prioritize results that maximize long-term outcomes. They also must be easily understood, multidimensional in scope, and actionable. Their development involves collaboration among stakeholders—including policy makers, justice officials, and civil society organizations—and depends on reliable data sources and the standardization of data collection methods.

Challenges to developing and using justice indicators include institutional resistance to performance evaluations— quite common in Latin America and the Caribbean, low data quality and availability, and the potential for institutions to manipulate metrics for appearance rather than genuine improvement. Additionally, benchmarking across countries or (in some cases) across states within a country can be complicated by differing legal frameworks and structures. Addressing these challenges requires political commitment, independent audits, methodological rigor, and creativity.

Examples of justice indicators include case clearance rates, operational performance rates (such as indictment or conviction rates), trial durations, and access to legal aid. To maximize their effectiveness and impact, indicators such as these should be integrated into broader reform strategies by linking them to performance improvement initiatives, ensuring transparency through public access to data, encouraging collaboration among justice institutions, and regularly updating them to reflect the evolving needs of the justice system.

Notes

a. This box is based on Dandurand, Kittayarak, and MacPhail (2015).

Finally, as discussed, in some of the territories under their control, criminal organizations help people resolve disputes, replacing state institutions and taking on the role of imparting justice.91 Therefore, justice institutions that function well are among the first forms of state presence governments must bring to territories where state services have been absent.92

Supporting innovative conciliation and arbitration systems and other restorative and reparation mechanisms can be a first step toward regaining the state monopoly over justice provision in places where non-state actors have control. In the long term, they can work simultaneously with formal justice institutions, managing minor disputes and preventing them from reaching the courts. They can complement and operate in alignment with conventional judicial institutions.

Alternative dispute resolution mechanisms are not new in LAC but have been deployed in limited contexts (box 2.6). The region’s experience with these mechanisms can inform strategies for bringing state presence to the provision of justice in territories lost to the control of criminal groups.

91 Feldmann and Mantilla (2021).

92 Arjona and Saab (2024).

BOX 2.6.

mobile service Units may Be a way to Bring the state where it is absent

Most developing countries cannot provide essential services throughout their entire territory. One innovative approach tested in several contexts is providing services through mobile vehicles. Mobile clinics, for instance, have been widely deployed to reach people who cannot access quality health services.a In the context of the fight against organized crime, mobile justice services can be a first move to regain control over territories under criminal governance.

Mobile courts are increasingly used worldwide to extend justice services to remote and underserved areas.b Rigorous evaluations of these itinerant programs are scarce. In rural Liberia, a nongovernmental organization (NGO) provided mediation and advocacy services through community paralegals trained in formal law. To reach remote communities, the program deployed “mobile paralegals” on motorbikes to 160 villages across 5 of Liberia’s 15 counties. Legal aid led to an increase in clients reporting fair case outcomes, satisfaction, and better relationships following case resolution. c

The Mobile Unit for Attention and Orientation to Victims (MUAOV), launched in 2012 in Colombia, is another program that has been evaluated. This program was designed to address the needs of victims of armed conflict in remote and marginalized areas. The mobile units provided personalized consultations, specialized legal advice, training on rights and transitional justice mechanisms, and information about national and local programs for victims. Between 2012 and 2020, the MUAOV served more than 125,000 people, addressing issues such as human rights violations, administrative requests for compensation, and comprehensive reparations.d It facilitated the administrative processes related to reparations, increased the rate at which victims received compensation, and improved victims’ perceptions of justice and security.e

Notes

a. Vidrine et al. (2012); WHO (2006).

b. Harper (2011); Open Society Foundations (2013); UNDP (2014); World Bank (2006, 2008).

c. Sandefur and Siddiqi (2013).

d. Colombia, Ministerio de Justicia y del Derecho de Colombia (2020).

e. Vargas et al. (no date).

State Action to Fight Organized Crime Requires a Better Understanding of How It Operates

The lack of reliable information about organized crime and the challenges of systematically and comparably measuring it across countries are immense. This means policy making and state action to combat organized crime often fail, for lack of good data and information to guide them. Indeed, part of the institutional weaknesses discussed in the previous section results from a lack of data or capacity to use the available data to inform the course of action. Strategically allocating the police force in the territory, scaling up the sophistication of criminal analysis to target and carry out operations that gravely hurt criminal business instead of only capturing a low tier of an organization, and identifying youth at risk are examples of actions where better data and a better understanding of organized crime are prerequisites.

One of the biggest challenges is that organized crime has both visible elements—such as lethal violence—and hidden aspects that are much harder to quantify. This explains why authorities often respond reactively, particularly when high-profile incidents occur, as seen recently in Ecuador. However, areas where organized crime has the most influence often remain under the radar, and quietly controlled activities are rarely scrutinized. This allows criminal organizations to grow undetected until they reach a level where control becomes challenging.

As mentioned, while homicides are relatively well-documented, other crimes face underreporting, with the degree varying by type of crime, and the socioeconomic status, education, gender, and ethnicity of the victim. The main alternative sources of information are victimization surveys. However, these are not usually representative at a granular level and are not systematically collected, often because they are carried out by nongovernmental organizations (NGOs) with intermittent funding.

Thus, part of the problem is the absence of systematic official surveys. While LAC countries regularly gather information on labor markets and inflation, they neither systematically collect standardized surveys on criminal activity nor include survey modules in official household and business surveys.93 This is despite security being identified as one of the top concerns of public opinion in poll studies such as Latinobarómetro. Surveys capturing criminal activity in representative ways at the neighborhood level—necessary to inform the targeting of policies and programs—in areas facing critical levels of organized crime and violence are even rarer.94

The use of administrative data to inform state action and policy design is also not systematic or extended. Suspicious Activity Reports (SAR), through which financial institutions report transactions potentially associated with money laundering and other financial crimes to state authorities, are an exception. Still, their use to explore patterns and inform prosecution processes is very limited in LAC (box 2.7).

The most helpful use of administrative records to understand the workings of organized crime and inform prevention or security policies shows up in scholarly research, almost always as isolated efforts. Most of the time, these research projects are carried out by academic researchers who invest years of work in limited geographic areas. The experiences of long-term research focused in Medellín (Colombia) and El Salvador show there is much to gain from research specialization and research partnering with state agencies and city governments.

Box 2.8 presents a detailed description of Medellín’s criminal governance as uncovered by the Medellín Impact Lab, an endeavor from Universidad EAFIT in Medellin, the University of Chicago, and Innovations for Poverty Action. The research team adopted a multifaceted approach, gathering and analyzing administrative data, conducting citywide surveys representative at the neighborhood level, carrying out extensive qualitative interviews over the years, and, importantly, working side-by-side with the city government as much as possible to experiment with policy. For example, their work permitted identifying schools with youth at high risk of recruitment to target preventive interventions, measuring their impact on the youth’s changing life projects, making adjustments to the programs, and scaling them up. The Medellín Impact Lab’s innovative and comprehensive approach to studying organized crime offers valuable lessons for policy makers elsewhere: it is a good example of research that informs more effective public policies and demonstrates the importance of long-term, collaborative efforts in tackling complex social issues.

As mentioned, another team of researchers has conducted extensive and long-term research on the impact of gangs and public policy in El Salvador.95 Their studies have provided invaluable insights into gang recruitment strategies, the socioeconomic consequences of gang territorial control, and the effectiveness of various public policy interventions in this country. This research has shed light on the complex dynamics of gang control and its far-reaching effects on communities. Box 2.9 expands on the lessons from this experience.

This team’s vibrant research agenda is another example of the role rigorous and long-term research can play in informing state action and policy design at the country or city level. By understanding the complex dynamics of gang activity and its impact on society, policy makers can develop more effective strategies to combat crime and support community development in El Salvador and elsewhere.

93 There are a few exceptions. Bolivia, El Salvador, Guatemala, and Peru systematically include victimization questions in their household surveys. Brazil, Chile, and Colombia have had a one-time module in the household survey, while Costa Rica and the Dominican Republic have occasionally added a module. However, the number and scope of victimization questions in household surveys vary significantly across countries. El Salvador and Mexico conduct annual victimization surveys. Other countries have only collected victimization surveys occasionally.

94 Medellín (Colombia) has an annual official survey representative at the neighborhood level and São Paulo (Brazil) has one representative for nine city zones. Bogotá (Colombia) has an annual survey representative at the locality level collected by the city’s Chamber of Commerce. Fortaleza (Brazil) has a one-wave survey collected by a research team.

95 This research has been led by Micaela Sviatschi (Princeton University) and Carlos Schmidt-Padilla (University of California, Berkeley).

BOX 2.7.

Follow the money: dismantling Criminal Organizations requires disrupting the Financial Flows and networks that sustain these groupsa

Anti-money laundering (AML) measures are designed to prevent and detect the concealment of criminal profits. These measures can play a crucial role in the fight against organized crime in LAC by targeting the financial infrastructure that supports criminal activities.

Conceptually, AML measures contribute to:

• Disrupting financial networks and recovering assets. By identifying, freezing, and recovering assets linked to criminal organizations, AML measures can disrupt the financial flows and networks that sustain these groups and take the profitability out of such crimes.

• Enhancing transparency. Implementing stricter regulations on financial transactions, such as requiring reporting on the source of funds and the ultimate owners of companies, can increase transparency and make it harder for criminals to launder money.

• Targeting corruption and other financial crimes. By introducing AML controls, such as filing suspicious financial transactions, countries can reduce the opportunities for criminals and organized crime groups to profit from crimes such as corruption and tax evasion.

• Supporting the integrity and growth of financial markets. By increasing AML controls, countries promote productivity and investor confidence in local markets, which in turn promote growth.

• Fostering domestic and international cooperation. To effectively implement AML controls, countries must facilitate regular information sharing between relevant domestic authorities, such as supervisors, financial intelligence units, law enforcement, and foreign counterparts.

The primary challenge lies in implementation. Although targeting criminal finances is an effective policy strategy, it requires significant domestic coordination and innovation, as the information and powers needed to tackle illicit financial flows are held by different agencies, and criminals are often adapting their methods to move and conceal funds. As a result, only one in five countries is effectively investigating, prosecuting, and convicting money laundering offenses at a high or substantial level.b Advanced technologies such as artificial intelligence and blockchain can help detect suspicious activities and trace illicit funds more effectively. However, well-trained personnel capable of designing and implementing strategies to follow and investigate patterns of suspicious financial activity is the most crucial missing input.

Notes

a. This box uses text contributed by Yara Esquivel and Ailsa Hart from the World Bank Financial Integrity and Stability team.

b. FATF (2022).

BOX 2.8.

Understanding and Countering Organized Crime in medellína

The city of Medellín (Colombia) is home to an extensive network of organized crime, featuring about 400 street gangs known as combos, which collectively employ around 12,000 individuals. These gangs are deeply embedded in the city’s socioeconomic fabric, controlling approximately two-thirds of its neighborhoods. Gang members’ income varies significantly by rank, with foot soldiers earning about US$10,000 per year (in purchasing power parity terms), which places them near the city’s workers’ median income. At the top, leaders within the gangs and the larger mafia-like structures known as Razones earn salaries that place them well within Colombia’s top 1 percent of income earners. Combos often provide protection and order in their neighborhoods to reduce police presence and foster loyalty among residents while protecting their lucrative drug trafficking operations.

The hierarchical structure of these gangs is meticulously organized. Combos, which consist of 15 to 50 members, engage primarily in retail drug sales and extortion. Their operations extend to other coercive services such as loan sharking and debt collection. Razones oversee dozens of combos. They not only supply wholesale drugs and other resources to their affiliated combos but also coordinate their activities to ensure order and maximize profits by splitting geographical markets, enforcing collusion on drug prices, and facilitating peaceful negotiations to avoid costly conflict when required. Razones also communicate among themselves to coordinate activities at their level and maintain peace. Agreements at this level reflect a sophisticated organizational structure of business. As a result of these arrangements, violence in Medellín has fallen sharply compared to 15 years ago, when markets were under dispute. The remaining homicidal violence is targeted and meant as a punishment for breaches in the prevailing agreements.

The Medellín Impact Lab has been instrumental in uncovering the intricate details of this criminal governance. Administrative data played a crucial role in this research. Working with the city government, the team secured access to various data sets, including arrest records, geolocated crime reports, social security data, and school performance records. This comprehensive administrative data allowed for real-time mapping of crime hotspots and predictive analyses, essential for understanding the dynamics of organized crime in the city and informing the design of targeted security and prevention policies and programs.

Additionally, the research team has accompanied the city government in conducting annual large-scale victimization surveys representative of all of the city’s low- and medium-income neighborhoods. They also directly conducted three large-scale surveys covering approximately 8,000 residents and businesses across Medellín in 2019, 2023, and 2024. These surveys have been designed to capture underreported phenomena such as extortion and gang governance. They have shown, for example, that most extortion goes unreported—only about 15 official reports are filed per week compared to the estimated 150,000 households and businesses paying extortion fees—highlighting the need for better diagnostic tools and more effective interventions.

Collection of qualitative data has been equally crucial. Over several years, the research team conducted hundreds of interviews with residents, business owners, community leaders, police officials, and gang members. This qualitative approach provided deep insights into the gangs’ organizational structures, income sources, and governance strategies. It underscores the importance of understanding the economic and political organization of organized crime groups to design policies that can disrupt their operations, minimizing potential unintended consequences.

Partnerships have been vital for this research, including its experimental stage. Collaborations with local government, police, and nongovernmental organizations (NGOs) have facilitated designing and implementing interventions to reduce crime. One such intervention, a program named Operación Convivencia, intensified the presence of city outreach workers in neighborhoods where gangs were entrenched and state legitimacy was low. This program was piloted in a few neighborhoods before being scaled up to cover 40 neighborhoods, with the results from the impact evaluation guiding new approaches by the city to tackle order and counter gang governance.

Box 2.8. Understanding and Countering Organized Crime in Medellín (continuation)

The Medellín Impact Lab also initiated a pilot project focused on gang recruitment of minors, which aims to provide alternative opportunities and support systems for at-risk youth and reduce the allure of gang involvement. The pilot’s success has led the city to commit funds to implement the intervention with two cohorts of 300 adolescents per year for at least three years. The program is called Parceritos. A first cohort is part of an ongoing evaluation to measure the program’s impact on youth recruitment and overall crime rates in the medium and long term.

Notes

a. This box is based on Blattman et al. (2022); Blattman et al. (2024a); Blattman, Lessing, and Tobón (2025).

BOX 2.9.

specialized research on gangs in el salvadora

Organized crime in El Salvador, predominantly driven by gangs such as MS-13 and Barrio 18, represents a complex challenge characterized by territorial control, violence, and illicit activities, including extortion and drug trafficking. In 2015, it was reported that more than 50 percent of municipalities in El Salvador had gang presence.b

A team led by scholars from Princeton University and the University of California (Berkeley) has conducted extensive and long-term research on the impact of gangs and public policy in El Salvador. Their work has revealed, for example, the profound effect of gangs on children in El Salvador. Criminal organizations employed varied strategies to recruit vulnerable children, including direct coercion and offering increased social status and acceptance. Children exposed to gang leaders during their formative years are more likely to become affiliated with these gangs as adults.c Moreover, prisons in El Salvador often served as breeding grounds for further criminal development and network-building among inmates. For example, inmates exposed to gang leaders were more likely to engage in gang-related activities upon release.d

The territorial control exerted by gangs significantly hindered development by imposing restrictions on residents’ mobility and access to opportunities. In gang-controlled neighborhoods, individuals experienced lower material well-being, reduced income, and limited educational opportunities compared to those living just 50 meters outside gang territory. This restricted mobility also limited labor market options, as individuals in controlled areas found commuting to work in non-gang areas challenging. This research also documents how criminals deeply influenced firms’ behavior and political outcomes in municipalities affected by gang violence. In particular, it shows that in municipalities affected by gangs firms were less likely to open, and voter turnout was reduced.e

Research on public policies addressing these issues has yielded promising results. The Safe Schools (Escuelas Seguras) program, implemented in 2015, involved the preventive use of police patrolling to reduce gang recruitment around educational centers. This initiative led to fewer school dropouts and reduced recruitment and arrests of vulnerable cohorts. As a result, there was a decline in extortion and homicides five years after the gang operations were disrupted.f

The study of truces between gangs shows both positive and negative outcomes. While they can reduce violence in the short term, truces often strengthen criminal groups, leading to increased criminal activities and territorial control. g

Box

2.9.

Specialized Research on Gangs in El Salvador

Government crackdowns on gangs, like the one in 2022, have shown promise in reclaiming neighborhoods and restoring residents’ freedom of movement. Yet these crackdowns can have unintended consequences, such as increased stigma and discrimination against residents of former gang territories in the job market, as shown by the team’s most recent research on El Salvador. While such measures address immediate safety concerns, their long-term success depends on context and careful consideration of the potential to weaken institutional legitimacy.h

Notes

a. This box is based on the work of Micaela Sviatschi (Princeton University) and Carlos Schmidt-Padilla (University of California, Berkeley).

b. Martinez et al. (2016).

c. Sviatschi (2022).

d. Golcalves et al. (no date).

e. Melnikov, Schmidt-Padilla, and Sviatschi (2020).

f. Castro et al. (no date).

g. Brown et al. (2024).

h. Melnikov, Schmidt-Padilla, and Sviatschi (no date).

Organized Crime Is a First-Order Problem for LAC Because It is a Growing Barrier to Development

LAC is beleaguered by organized criminal groups that, in many places, evolved from drug trafficking organizations into transnational corporations that control drug trafficking corridors, smuggling and migration routes, and scores of illegal and legal activities. This chapter has discussed how organized crime hinders the region’s development. It cannot be emphasized enough how much of a first-order problem organized crime is in LAC. Urgent policy action and full political support are necessary from governments across the region to contain its expansion and social damage.

Organized Crime Breeds in the Absence of Opportunities

This chapter has summarized the increasing reach of organized crime in LAC: its defining features, and its relationship to the region’s epidemic of violence. It has underlined how this is ultimately a development problem that finds a breeding ground in the lack of opportunities and political participation of the region’s most vulnerable people. This implies that in addition to the urgent actions already discussed in relation to the fundamental pillars of security policy (prisons, police forces, and the justice system), fighting organized crime also requires addressing a range of multiplying factors of criminal activity.

Conceptually, these factors can be divided into those that can help reduce criminal organizations’ labor supply and access to capital (thus increasing their operating costs) and those that curtail the demand for the types of goods and services they provide (thus decreasing their income).

Regarding labor supply, the lack of opportunities—especially for young men, who are the primary victims and main perpetrators of criminal violence—facilitates recruitment into criminal organizations. The research agendas from Medellín and El Salvador discussed in this chapter unambiguously support this insight. In many cases, recruits

are not forced into crime, but criminal groups offer significantly better income opportunities than those available in legal markets. They also offer career prospects and elements of “emotional mobility” (respect, appreciation, and recognition)—aspects that the precarious non-criminal jobs the region offers to most of its youth often fail to provide.96

Moreover, many of these men have been brought up in violent contexts, whether in their communities or households—64 percent of children under 15 in LAC have experienced violent discipline at home in the form of severe physical punishment or emotional aggression.97 Only 10 LAC countries have legislation prohibiting these practices.98 Constant exposure to violence increases the likelihood that young people internalize aggression as a survival and empowerment mechanism, making gang affiliation seem like a form of protection and empowerment.99 The influence of friends and acquaintances is also a determining factor. Research on youth motivations for joining groups suggests that those who join for instrumental reasons, such as economic benefits or protection, are more likely to engage in criminal activities. Peer pressure also plays a crucial role, as social dynamics within gangs reinforce participation in criminal activities.100

The lack of education and economic opportunities interacts with, and is exacerbated by, the region’s political inequality. Large segments of the population in LAC have no voice or agency. Political inequality is a source of de facto disenchantment that increases the demand for alternative forms of governance, feeding into the appeal that the types of services criminal organizations provide.

This analysis suggests that there are policies that could deal simultaneously with supply and demand mechanisms. In many parts of the region, insufficient opportunities and political agency exist alongside with a weak state. Such weakness is evident in several dimensions, such as the lack of state monopoly over justice and security services, the unequal state presence throughout the territory, its inability to provide essential services to all citizens, excessive regulation coupled with poor enforcement, and its frequent resort to militarized solutions, which can backfire in the control of organized criminal violence and in terms of citizens’ trust. A weak or absent state is why there is space for organized crime groups to offer services, including protection and conflict resolution—two dimensions that are at the core of a functioning state.

Working-age men and women need better alternatives to crime in the form of quality jobs that offer prospects for improvements along their life cycle. The reforms required to provide this type of jobs have been delayed for a long time, and the region has paid a high price for this, not only in providing an abundant supply of labor for criminal activities but also in terms of foregone growth, prosperity, and overall development. While these reforms are complex, they are an essential part of public security. Better prisons, better police, and better justice systems are indispensable, but so are better opportunities. Without the latter, it is doubtful that the former will deliver the security and tranquility that citizens justly expect from their governments.

LAC Can Build Strategic Capacity to Manage Organized Crime

This report does not argue that state building is a precondition for fighting organized crime. That would entail delaying urgent policy action on organized crime per se—and strengthening government institutions should guide state action in LAC regardless of the prevalence of organized crime. Instead, governments should put the fight against organized crime at the forefront of the policy agenda now. To that end, they must identify the behaviors and outcomes that cannot be tolerated and draw red lines to guide context-specific prioritized state action.

96 Amaya and Martínez d’Aubuisson (2021); Blattman, Rodríguez-Uribe, and Tobón (no date); Bouchard and Splinder (2010).

97 UNICEF (2018a).

98 UNICEF (2018b).

99 Hoeve et al. (2009); López Stewart et al. (2000); Rojas-Flores et al. (2013).

100 Lachman, Roman, and Cahill (2013); US, District of Columbia Metropolitan Police Department (2011).

There are no acceptable crimes, but given the complexity and multiplicity of forms that organized criminal activity can take—as well as the limited capacity of the state to confront them—policy makers must prioritize and start by identifying the kind of crimes that have the most adverse impacts and focus all state efforts, in a coordinated way, to counteract them. This type of focus can generate new challenges, but given the breadth of the problem, the region cannot aspire to attack the entire phenomenon of organized crime holistically. Neither the United States nor Europe have managed to eradicate organized criminal activity completely. Instead, they have learned how to manage the problem by controlling the most harmful activities and behaviors for society.

Rethinking prisons, restructuring the police, prioritizing judicial action, promoting institutional coordination, raising the costs of criminal business, and thwarting the demand for criminal services cannot be achieved without a better comprehension of the nature, complexity, dynamics, and multiple expressions of organized crime. Producing better information is an urgent priority in the fight against organized crime in the region. As suggested by the experiences ongoing in Medellín and El Salvador, deep knowledge of the context and original evidence—both quantitative and qualitative— are key to a more comprehensive understanding of organized crime in specific territories. Harmonizing administrative records of crime and violence across countries, producing subnational information by including victimization modules in household and business surveys, and engaging in data collection at the submunicipal (neighborhood or even street) level are essential preconditions for nuanced—and more policy-relevant—analyses.101

Security Policy and Prevention Policies Are Complements in the Fight Against Organized Crime

Policies that contribute to fixing the workings of LAC’s precarious labor markets and generate better job opportunities for all—including education and skill training systems and better regulatory frameworks that level the playing field for business— are fundamental; these will increase the opportunity costs of engaging in criminal activities, and thus reduce the labor supply of organized crime, as is likely the case in the more developed parts of the world. Simultaneously, policies to prevent recruitment targeted at high-risk youth must be prioritized because they can help break criminal organizations’ access to vulnerable populations. Actions can include targeted educational interventions, community-based prevention and protection, and resocialization of former criminals. Also, interventions addressing harmful masculinities and gender inequality can help replace norms that underlie the acceptability of different forms of violence with those that construct peaceful societies.102 Mentorship and emotional support programs, where young people can find role models outside the criminal context, have also proven effective.103

Prevention cannot substitute for security policies but is a necessary complement. Similarly, even a good security policy without long-term prevention measures will be insufficient. A combination of security and prevention interventions will likely yield the best results (box 2.10).

International Coordination and Collaboration Are Critical to Face Organized Crime

As emphasized throughout this discussion, LAC states are responsible for building the strategic capacity they need to manage organized crime, protect their citizens from its most harmful expressions, and give them the opportunities to have lives they can value. However, they are unlikely to succeed on their own because organized crime is a problem that surpasses national borders and requires coordinated solutions. Achieving effectiveness in the fight against organized crime requires high levels of international coordination and collaboration.

101 Blattman et al. (2022).

102 Zermeño (2024).

103 National Gang Center (no date).

BOX 2.10.

Cases in which security and prevention policies strengthen One anothera

Tackling organized crime requires prevention and security responses that effectively complement one another. Examples of such complementarity include:

1. Combatting illegal economies. Security interventions aiming to reduce the supply of drugs, such as the interdiction of trafficking areas, reduce criminal revenues. Complementary prevention interventions may increase the effectiveness of these supply-reduction efforts, including prevention and psychosocial interventions aimed at decreasing the consumption of drugs by youth that may lower the demand for drugs or delay the age of first consumption and reduce dependency.b Similar interventions that combine prevention and security can be considered for other illegal economic activities, such as migrant smuggling: for instance, combining increased policing of irregular routes with efforts to promote legal migration pathways and socioeconomic inclusion at places of origin.

2. Territorial development approaches. Approximately 50 percent of crimes occur in 2.5 percent of the street sections in LAC cities, and homicides occur in only 1.4 percent of them.c While hot-spot policing may reduce localized crime under some conditions, it can lead to the relocation of criminal activity into new areas if implemented in isolation from prevention strategies. Instead, complementary prevention efforts in nearby neighborhoods may reduce the capacity of criminals to relocate criminal activity.

Notes

a. This box uses text contributed by Paula Rossiasco from the World Bank’s South America Forced Displacement and Migration Social Inclusion Team.

b. Preventive programs with active and/or participatory methodologies have been effective in preventing drug use and juvenile delinquency. See Aguilera-Torrado and Payares-Ortiz (2021); Strang et al. (2012).

c. Chainey et al. (2019).

Improving cross-country platforms for exchanging information is critical—across customs and tax authorities, financial intelligence units, and criminal investigation units. The currently limited information exchanges allow criminals to move from one country to another, bringing their criminal business across borders.

Facilitating exchanges of criminal evidence from one jurisdiction to another is also key. It requires stepping up another level of collaboration to standardize crime definitions, probatory evidence, and protocols for investigation in penal codes across countries, as in the European Union. The standardization effort required is even more significant in countries with federal governments and multiple penal codes. The United States is a good example of the latter.

Collecting periodic and comparable information to measure organized crime and violence in its multiple expressions is also a step toward a more significant coordinated international effort. Not only is this important in its own right, but it signals that it is a policy priority and enables progress to be benchmarked.

Another paradigmatic example of the need for improved coordination among public and private agents within and across countries is the handling of Suspicious Activity Reports (SAR) from banks and other financial institutions, which still fail to inform efforts to capture and imprison organized crime leaders or help dismantle (at least part of) their business (box 2.7). Strengthening units in the ministries of finance charged with detecting money laundering through banks and financial institutions is thus a critical priority. However, this is probably not enough. The rise of cryptocurrencies poses an additional challenge, as some financial transactions occur through parallel channels

that avoid the formal financial system. Thus, finance ministries, perhaps in conjunction with central banks, must strengthen their understanding of these parallel channels and devise measures to minimize their illegitimate use.

The lack of coordination among state institutions is a characterizing element of underdevelopment, and lies at the root of most failed policies in LAC. However, improving coordination and collaboration among the state agencies in charge of security policy both domestically and externally with their peers in other countries must be a priority because of the costly ramifications of the absence of coordination and the myriad ways it weakens the capacity of individual agencies to fulfill their mission.

Finally, putting a halt to the illicit trafficking of legally sold US firearms that fuel violence in LAC critically depends on international acknowledgment of the problem and collaboration to solve it. Also, the region cannot abstain from the global conversation about drug policy. The failed war on drugs, especially concerning cannabis and cocaine, has allowed organized crime to control drug markets, leading to higher rents and territorial dominance.104 Effective regulation can reduce these illegal revenues and support participants in formerly illicit markets, as seen in New York’s cannabis market. To reduce crime in LAC, the global discussion on regulation must progress.

References

Abi-Habib, M. 2024. “How a Tourist Paradise Became a Drug-Trafficking Magnet.” New York Times, September 17, 2024.

Abi-Habib, M., A. Correal, and J. Nicas. 2024. “In Latin America, Guards Don’t Control Prisons, Gangs Do.” New York Times, February 21, 2024.

Abril, V., E. Norza, S. Perez-Vincent, S. Tobon, and M. Weintraub. 2024. “Building Trust in the Police: Evidence from a Multi-site Experiment in Colombia.” Center for Open Science, SocArXiv mrh5q_v1.

Abt, T. 2019. Bleeding Out: The Devastating Consequences of Urban Violence—and a Bold New Plan for Peace in the Streets. Basic Books. Acevedo, C. 2008. “Los costos económicos de la violencia en Centroamérica.” San Salvador: Consejo Nacional de Seguridad. Agan, A., J. L. Doleac, and A. Harvey. 2023. “Misdemeanor Prosecution.” Quarterly Journal of Economics 138 (3): 1453–1505.

Agan, A. Y., A. Garin, D. K. Koustas, A. Mas, and C. Yang. 2023. “Labor Market Impacts of Reducing Felony Convictions.” NBER Working Paper 31773, National Bureau of Economic Research, Cambridge, MA.

Agência Brasil. 2024. “Crimes ambientais na Amazônia Legal aumentam 88%, aponta PRF.” October 29, 2024.

Aguilera-Torrado, A., and A. Payares-Ortiz. 2021. “El club juvenil como estrategia para la prevención del consumo de drogas y la delincuencia juvenil. Caso Barrancabermeja.” Revista Criminalidad 63 (2): 155–74.

Aguirre, K., and R. Muggah. 2020. “Arming the Americas.” In The Oxford Handbook of the Sociology of Latin America, edited by X. Bada and L. Rivera-Sánchez, 717–37. Oxford Academic.

Aizer, A., and J. J. Doyle, Jr. 2015. “Juvenile Incarceration, Human Capital, and Future Crime: Evidence from Randomly Assigned Judges.” Quarterly Journal of Economics 130 (2): 759–803.

Akee, R., A. K. Basu, A. Bedi, and N. H. Chau. 2014. “Transnational Trafficking, Law Enforcement, and Victim Protection: A Middleman Trafficker’s Perspective.” Journal of Law and Economics 57 (2): 349–86.

Albarracín, J. 2018. “Criminalized Electoral Politics in Brazilian Urban Peripheries.” Crime, Law and Social Change 69 (4): 553–75. Alsan, M., A. M. Barnett, P. Hull, and C. Yang. 2024. ““’Something Works’ in US Jails: Misconduct and Recidivism Effects of the IGNITE Program.” NBER Working Paper 32282, National Bureau of Economic Research, Cambridge, MA.

Amazon Watch, InfoAmazonia, Amazon Underworld, and GI-TOC (Global Initiative against Transnational Organized Crime). 2023. “Amazon Underworld Economias Criminosas Na Maior Floresta Tropical do Mundo.”

Amaya, L. E., and J. J. Martínez d’Aubuisson. 2021. Criminal Governance in El Salvador. Global Initiative against Transnational Organized Crime (GI-TOC).

Amnesty International. 2016. “Brazil: Police Killings, Impunity, and Attacks on Defenders.” Amnesty International Submission for the UN Universal Periodic Review–27th Session of the UPR Working Group, May 2017.

Amnesty International. 2022. “El Salvador: President Bukele Engulfs the Country in a Human Rights Crisis after Three Years in Government.” June 2, 2016.

Arcia, G. 2012. “The Evolution of Violence: Economic Development and Intergroup Conflict in Guatemala, El Salvador, and Costa Rica.” In Economic Development Strategies and the Evolution of Violence in Latin America, edited by W. Ascher and N. Mirovitskaya, 71–93. Palgrave MacMillan.

Arias, E. D., and N. Barnes. 2017. “Crime and Plural Orders in Rio de Janeiro, Brazil.” Current Sociology 65 (3): 448–65. Ariel, B., W. A. Farrar, and A. Sutherland. 2015. “The Effect of Police Body-Worn Cameras on Use of Force and Citizens’ Complaints against the Police: A Randomized Controlled Trial.” Journal of Quantitative Criminology 31 (2015): 509–35.

Arjona, A. 2016. Rebelocracy: Social Order in the Colombian Civil War. Cambridge University Press.

Arjona, A. 2021. “The Effects of Violence on Inequality in Latin America and the Caribbean: A Research Agenda.” UNDP LAC Working Paper Series 12, United Nations Development Programme.

Arjona, A., M. Chacón, and L. García-Montoya. 2025. “The Impact of Political Assassinations on Turnout: Evidence from Colombia.” Proceedings of the National Academy of Sciences 122, no. 11.

Arjona, A. M., and S. N. Kalyvas. 2012. “Recruitment into Armed Groups in Colombia: A Survey of Demobilized Fighters.” In Understanding Collective Political Violence, 143–71. London: Palgrave Macmillan UK.

Arjona, A., N. Kasfir, and Z. Mampilly, eds. 2015. Rebel Governance in Civil War. Cambridge University Press.

Arjona, A., and A. Saab. 2024. “State Capacity and Rebel and Criminal Governance: The Centrality of Dispute Resolution Institutions.” Working paper.

Assaraf, N., A. Mouro, D. M. Papy, N. Castillo, and B. Ariel. 2024. “Behind the Yellow Sticker: Paradoxical Effects of a Visual Warning of Body-Worn Cameras on the Use of Police Force.” Journal of Experimental Criminology, 1–23. https://doi.org/10.1007/ s11292-024-09627-8.

Austin, A. 2023. “Why Have Ecuador’s Drug Trafficking Gangs Turned to Extortion?” InSight Crime, December 20, 2023.

Aziani, A., G. A. Bertoni, M. Jofre, and M. Riccardi. 2023. “COVID-19 and Organized Crime: Strategies Employed by Criminal Groups to Increase their Profits and Power in the First Months of the Pandemic.” Trends in Organized Crime 26 (2): 114–35.

Ba, B. A., D. Knox, J. Mummolo, and R. Rivera. 2021. “The Role of Officer Race and Gender in Police-Civilian Interactions in Chicago.” Science 371 (6530): 696–702.

Barbosa, D., T. Fetzer, P. C. Souza, and C. Vieira. 2021. “De-escalation Technology: The Impact of Body-Worn Cameras on Citizen-Police Interactions.” CEPR Discussion Paper No. DP16578, Centre for Economic and Policy Research, London.

Becker, Gary S. 1968. “Crime and Punishment: An Economic Approach.” Journal of Political Economy 76 (2): 169–217.

Bernal, C., M. Prem, J. F. Vargas, and M. Ortiz. 2024. “Peaceful Entry: Entrepreneurship Dynamics during Colombia’s Peace Agreement.” Journal of Development Economics 166 (January): 103119.

Bhuller, M., G. B. Dahl, K. V. Løken, and M. Mogstad. 2020. “Incarceration, Recidivism, and Employment.” Journal of Political Economy 128 (4): 1269–1324.

Biondi, K. 2016. Sharing This Walk: An Ethnography of Prison Life and the PCC in Brazil, edited and translated by J. F. Collins. University of North Carolina Press.

Bisca, P. M., V. Chau, P. Dudine, R. A. Espinoza, J.-M. Fournier, P. Guérin, N.-J. H. Hansen, and J. Salas. 2024. Violent Crime and Insecurity in Latin America and the Caribbean: A Macroeconomic Perspective. Washington, DC: International Monetary Fund.

Blanco, L., and I. Ruiz. 2013. “The Impact of Crime and Insecurity on Trust in Democracy and Institutions.” American Economic Review 103 (3): 284–88.

Blattman, C. 2023. Why We Fight: The Roots of War and the Paths to Peace. Penguin.

Blattman, C. 2024. “Bad Medicine: Why Different Systems of Organized Crime Demand Different Solutions.” Center for Open Science, SocArXiv ghcpj_v1.

Blattman, C., G. Duncan, B. Lessing, and S. Tobón. 2024a. “Gang Rule: Understanding and Countering Criminal Governance.” Review of Economic Studies, rdae079.

Blattman, C., G. Duncan, B. Lessing, and S. Tobon. 2024b. “Statebuilding in the City: An Experiment in Civilian Alternatives to Policing.” Working paper, Pearson Institute for the Study and Resolution of Global Conflicts.

Blattman, C., G. Duncan, B. Lessing, and S. Tobón. 2024c. “Gangs of Medellin: How Organized Crime is Organized.” Working Paper, BFI Friedman Forum, University of Chicago Law School, January 10, 2024.

Blattman, C., B. Lessing, J. P. Mesa-Mejía, and S. Tobón. 2022. “Measuring Organised Crime: Challenges and Solutions for Collecting Data on Armed Illicit Groups.” Briefing Note 3, SOC ACE (Serious Organized Crime & Anti-Corruption Evidence).

Blattman, C., B. Lessing, and S. Tobon. 2025. “How to Map and Combat Urban Organized Crime: Lessons from the Medellín Impact Lab.” Briefing Note 33, SOC ACE (Serious Organized Crime & Anti-Corruption Evidence), January 2025.

Blattman, C., A. Rodríguez-Uribe, and S. Tobón. No date. “Who Expects to Join Criminal Gangs and Why? Occupational Choice among 15,000 Teenage Boys in Medellín.” Working paper 2024-03, Universidad EAFIT.

Bouchard, M., and A. Spindler. 2010. “Groups, Gangs, and Delinquency: Does Organization Matter?” Journal of Criminal Justice 38 (5): 921–33.

Braga, A. A., R. K. Brunson, and K. M. Drakulich. 2019. “Race, Place, and Effective Policing.” Annual Review of Sociology 45: 535–55.

Braga, A., J. R. Coldren Jr, W. Sousa, D. Rodriguez, and O. Alper. 2017. The Benefits of Body-Worn Cameras: New Findings from a Randomized Controlled Trial at the Las Vegas Metropolitan Police. Final report to the National Institute of Justice, 2013-IJ-CX0016. Arlington, VA: CNA.

Braga, A. A., J. M. MacDonald, and J. McCabe. 2022. “Body-Worn Cameras, Lawful Police Stops, and NYPD Officer Compliance: A Cluster Randomized Controlled Trial.” Criminology 60 (1): 124–58.

Braga, A. A., J. M. MacDonald, and J. E. McCabe. 2023. “Police Reform in Public Housing Contexts: Body-Worn Cameras, Surveillance, and Harm Reduction in New York City Housing Authority Developments.” Justice Quarterly 40 (4): 1267–94.

Braga, A. A., D. Weisburd, and B. Turchan. 2018. “Focused Deterrence Strategies and Crime Control: An Updated Systematic Review and Meta-Analysis of the Empirical Evidence.” Criminology & Public Policy 17 (1): 205–50.

Brown, Z. Y., E. Montero, C. Schmidt-Padilla, and M. M. Sviatschi. 2024. “Market Structure and Extortion: Evidence from 50,000 Extortion Payments.” Review of Economic Studies, rdae057.

Calderón, G., G. Robles, and B. Magaloni. 2015. “The Beheading of Criminal Organizations and the Dynamics of Violence in Mexico.” Journal of Conflict Resolution 59 (8): 1455–85.

Carreras, M. 2013. “The Impact of Criminal Violence on Regime Legitimacy in Latin America.” Latin American Research Review 48 (3): 85–107.

Castillo, J. C., D. Mejía, and P. Restrepo. 2020. “Scarcity without Leviathan: The Violent Effects of Cocaine Supply Shortages in the Mexican Drug War.” Review of Economics and Statistics 102 (2): 269–86.

Castro, E., F. Coy, C. Schmidt-Padilla, and M. M. Sviatschi. 2024. “Breaking the Gang: A Preventive Approach to Reduce Recruitment in Schools.” June 28, 2024, Seminario de Economía, Universidad de la Plata.

Ceobanu, A. M., C. H. Wood, and L. Ribeiro. 2011. “Crime Victimization and Public Support for Democracy: Evidence from Latin America.” International Journal of Public Opinion Research 23 (1): 56–78.

Cerqueira, D. R. D. C., S. Bueno, R. S. D. Lima, C. Neme, H. R. S. A. Ferreira, P. P. Alves..and K. C. Armstrong. 2019. Atlas da violência 2019. Instituto de Pesquisa Econômica Aplicada (IPEA) & Fórum Brasileiro de Segurança Pública.

Chainey, S. P., G. Pezzuchi, N. O. Guerrero Rojas, J. L. Hernandez Ramirez, J. Monteiro, and E. Rosas Valdez. 2019. “Crime Concentration at Micro-Places in Latin America.” Crime Science 8 (Article 5).

Chassang, S., and G. P. i Miquel. 2009. “Economic Shocks and Civil War.” Quarterly Journal of Political Science 4 (3): 211–28.

Chatterjee, J. 2005. “The Changing Structure of Organized Crime Groups.” Research and Evaluation Branch, Community, Contract and Aboriginal Policing Services Directorate, Royal Canadian Mounted Police.

Chen, M. K., and J. M. Shapiro, 2007. “Do Harsher Prison Conditions Reduce Recidivism? A Discontinuity-based Approach.” American Law and Economics Review 9 (1): 1–29.

Colombia, Ministerio de Justicia y del Derecho de Colombia. 2020. Unidad Movil Atencion Orientacion Victimas. https://www. minjusticia.gov.co/programas-co/justicia-transicional/Paginas/Unidad-Movil-Atencion-Orientacion-Victimas.aspx

Cox, R. J., J. P. Cunningham, and A. Ortega. 2024. “The Impact of Affirmative Action Litigation on Police Killings of Civilians.” NBER Working Paper 32502, National Bureau of Economic Research, Cambridge, MA. Cristosal Human Rights. 2023. “El Salvador: Widespread Abuses under State of Emergency.” April 1, 2023. https://cristosal.org/ EN/2023/04/01/el-salvador-widespread-abuses-under-state-of-emergency/ Dandurand, Y., K. Kittayarak, and A. MacPhail. 2015. “Justice Indicators and Criminal Justice Reform: A Reference Tool.” International Centre for Criminal Law Reform and Criminal Justice Policy.

Davenport, C. 2007. “State Repression and Political Order.” Annual Review of Political Science 10 (1): 1–23.

Davis, J. M. V., T. Meares, and E. Arnesen. 2025. “Improving Programming in Juvenile Detention: The Impact of Project Safe Neighborhoods Youth Outreach Forums.” Journal of Quantitative Criminology 41 (March): 23–50.

Dell, M. 2015. “Trafficking Networks and the Mexican Drug War.” American Economic Review 105 (5): 1738. Detotto, C., and E. Otranto. 2010. “Does Crime Affect Economic Growth?” Kyklos 63 (3): 330–45.

Di Nicola, A. 2022. “Towards Digital Organized Crime and Digital Sociology of Organized Crime.” Trends in Organized Crime, 1–20. https://doi.org/10.1007/s12117-022-09457-y

Di Tella, R., and E. Schargrodsky. 2013. “Criminal Recidivism after Prison and Electronic Monitoring.” Journal of Political Economy 121 (1): 28–73.

Donohue III, J. J., and S. D. Levitt. 2001. “The Impact of Race on Policing and Arrests.” Journal of Law and Economics 44 (2): 367–94. Donohue, J. J., and J. Wolfers. 2006. “Uses and Abuses of Empirical Evidence in the Death Penalty Debate.” NBER Working Paper 11982, National Bureau of Economic Research, Cambridge, MA.

Doran, M.-C. 2017. “The Hidden Face of Violence in Latin America: Assessing the Criminalization of Protest in Comparative Perspective.” Latin American Perspectives 44 (5): 183–206.

Draca, M., J. Garred, L. Stickland, and N. Warrinnier. 2023. “On Target? Sanctions and the Economic Interests of Elite Policymakers in Iran.” Economic Journal 133 (649): 159–200.

Drago, F., R. Galbiati, and P. Vertova. 2009. “The Deterrent Effects of Prison: Evidence from a Natural Experiment.” Journal of Political Economy 117 (2): 257–80.

Dube, A., O. Dube, and O. García-Ponce. 2013. “Cross-Border Spillover: U.S. Gun Laws and Violence in Mexico.” American Political Science Review 107 (3): 397–417.

Dube, O., S. J. MacArthur, and A. K. Shah. 2023. “A Cognitive View of Policing.” NBER Working Paper 31651, National Bureau of Economic Research, Cambridge, MA. Dudley, S. 2020. 3 Dirty Secrets Revealed by the El Salvador Gang ‘Negotiations.’ Technical report, InSight Crime. ECLAC (United Nations Economic Commission for Latin America and the Caribbean). 2018. Mujeres afrodescendientes en América Latina y el Caribe: deudas de igualdad. ECLAC.

Economist, The. 2016. “The Gangs that Cost 16% of GDP.” The Americas. May 21, 2016.

Economist, The. 2023a. “Nayib Bukele Shows How to Dismantle a Democracy and Stay Popular.” The Americas. July 20, 2023.

Economist, The. 2023b. “Illegal Gold Is Booming in South America.” The Americas. November 9, 2023.

Economist, The. 2023c. “Latin America’s Most Powerful New Gang Built a Human-Trafficking Empire.” The Americas. November 16, 2023.

Economist, The. 2024a. “Gangsters in El Salvador Are Terrified of Strongman Nayib Bukele.” Leaders. February 2, 2024. Economist, The. 2024b. “The World’s Most Violent Region Needs a New Approach to Crime.” The Americas. May 9, 2024. El Universo. 2021. “ Dueño de farmacia se negó a pagar $5.000 por ‘vacuna’ y delincuentes tiraron explosivo para destruir su negocio en Pascuales .” August 31, 2021. https://www.eluniverso.com/noticias/seguridad/ explosion-en-farmacia-de-pascuales-por-no-pagar-vacuna-nota/

Eren, O., and N. Mocan. 2021. “Juvenile Punishment, High School Graduation, and Adult Crime: Evidence from Idiosyncratic Judge Harshness.” Review of Economics and Statistics 103 (1): 34–47.

Estelle, S. M., and D. C. Phillips. 2018. “Smart Sentencing Guidelines: The Effect of Marginal Policy Changes on Recidivism.” Journal of Public Economics 164 (August): 270–93.

Fagundes, E., J. Monteiro, and P. Souza. 2023. “Monitoring Street-level Bureaucrats: The Impact of Body-Worn Cameras on Police Performance.” Working paper, Unpublished manuscript.

Feldman, A., and J. Mantilla. 2021. “Criminal Governance in Latin America.” doi:10.1093/ACREFORE/9780190264079.013.697.

Financial Times. 2024. “Mexico’s Drug Cartels Are Thriving.” May 20, 2024. https://www.ft.com/content/ fe04c6ed-73f8-4e17-852b-ce16fd6c3515

Freeman, W. 2024. “Brazil’s Largest Mafia Is Entering Politics. The Government Must Act.” New York Times, October 4, 2024.

Gallego, J. 2018. “Civil Conflict and Voting Behavior: Evidence from Colombia.” Conflict Management and Peace Science 35 (6): 601–21. Gaviria, A. 2002. “Assessing the Effects of Corruption and Crime on Firm Performance: Evidence from Latin America.” Emerging Markets Review 3 (3): 245–68.

GI-TOC (Global Initiative against Transnational Organized Crime). No date. “Busted: Albanian Cocaine Network Run from Prison in Ecuador.” Risk Bulletin of Illicit Economies in South Eastern Europe.

GMH (Grupo de Memoria Histórica de la Comisión Nacional de Reparación y Reconciliación). 2013. ¡Basta Ya! Colombia: Memorias de guerra y dignidad. Bogotá, Colombia: Centro Nacional de Memoria Histórica.

Golcalves, F., P. Matos, C. Schmidt-Padilla, and M. M. Sviatschi. No date. “Mass Incarceration and the Expansion of Gangs: Evidence from El Salvador.” Unpublished manuscript.

Gómez Gallego, J. A., J. R. Herrera Vergara, and N. Pinilla Pinilla. 2010. Informe final: comisión de la verdad sobre los hechos del Palacio de Justicia. Universidad del Rosario.

González, Y. 2019. “The Social Origins of Institutional Weakness and Change: Preferences, Power, and Police Reform in Latin America.” World Politics 71 (1): 44–87.

Grabosky, P. 2007. “The Internet, Technology, and Organized Crime.” Asian Journal of Criminology 2 (2): 145–61.

Grenet, J., H. Grönqvist, and S. Niknami. 2024. “The Effects of Electronic Monitoring on Offenders and their Families.” Journal of Public Economics 230 (C): 105051.

Guerrero, E. 2011. La raíz de la violencia. Nexos.

Harper, E. 2011. “Customary Justice: From Program Design to Impact Evaluation.” International Development Law Organization, Rome.

Harvey, A., and T. Mattia. 2024. “Reducing Racial Disparities in Crime Victimization: Evidence from Employment Discrimination Litigation.” Journal of Urban Economics 141 (May): 103459.

Hoeve, M., J. S. Dubas, V. I. Eichelsheim, P. H. van der Laan, W. Smeenk, and J. R. M. Gerris. 2009. “The Relationship between Parenting and Delinquency: A Meta-Analysis.” Journal of Abnormal Child Psychology 37 (6): 749–75.

Holz, J. E., R. G. Rivera, and B. A. Ba. 2023. “Peer Effects in Police Use of Force.” American Economic Journal: Economic Policy 15 (2): 256–91.

Human Rights Watch. 2024. “El Salvador.” Country chapter in World Report 2024. Human Rights Watch.

IACHR (Inter-American Commission on Human Rights). 2015. “Violence against Lesbian, Gay, Bisexual, Trans and Intersex Persons in the Americas.” General Secretariat, Organization of American States.

Idrobo, N., D. Mejía, and A. M. Tribin. 2014. “Illegal Gold Mining and Violence in Colombia.” Peace Economics, Peace Science and Public Policy 20 (1): 83–111.

Insight Crime. 2023. Mexico’s Rising Femicides Linked to Organized Crime. Technical report. Insight Crime.

Insight Crime. 2024. 2023 Homicide Round-Up. Technical report. Insight Crime. International Crisis Group. 2021. Electoral Violence and Illicit Influence in Mexico’s Hot Land. Report No. 89. Brussels: International Crisis Group.

International Institute for Strategic Studies. 2024. The Expansion and Diversification of Mexican Cartels: Dynamic New Actors and Markets. https://www.iiss.org/publications/armed-conflict-survey/2024/ the-expansion-and-diversification-of-mexican-cartels-dynamic-new-actors-and-markets/ IOM (International Organization for Migration). 2021. “Migrant Smuggling.” December 23, 2021. https://lac.iom.int/en/ migrant-smuggling.

Justino, P. 2016. “Supply and Demand Restrictions to Education in Conflict-affected Countries: New Research and Future Agendas.” International Journal of Educational Development 47 (March): 76–85.

Kalsi, P. 2018. “The Impact of US Deportation of Criminals on Gang Development and Education in El Salvador.” Journal of Development Economics 135 (November): 433–48.

Kennedy, D. M. 2011. Don’t Shoot: One Man, a Street Fellowship, and the End of Violence in Inner-City America. Bloomsbury Publishing USA.

Kleinberg, J., H. Lakkaraju, J. Leskovec, J. Ludwig, and S. Mullainathan. 2018. “Human Decisions and Machine Predictions.” Quarterly Journal of Economics 133 (1): 237–93.

Krause, K. 2014. “Supporting the Iron Fist: Crime News, Public Opinion, and Authoritarian Crime Control in Guatemala.” Latin American Politics and Society 56 (1): 98–119.

Kuziemko, I. 2013. “How Should Inmates Be Released from Prison? An Assessment of Parole versus Fixed-Sentence Regimes.” Quarterly Journal of Economics 128 (1): 371–424.

Kuziemko, I., and S. D. Levitt. 2004. “An Empirical Analysis of Imprisoning Drug Offenders.” Journal of Public Economics 88 (9-10): 2043–66.

Lachman, P., C. G. Roman, and M. Cahill. 2013. “Assessing Youth Motivations for Joining a Peer Group as Risk Factors for Delinquent and Gang Behavior.” Youth Violence and Juvenile Justice 11 (3): 212–29.

Lessing, B. 2010. “The Danger of Dungeons: Prison Gangs and Incarcerated Militant Groups.” Small Arms Survey.

Lessing, B. 2013. “A Hole at the Center of the State: Prison Gangs and the Limits to Punitive Power.” CDDRL Working Paper 143, Center on Democracy, Development, and the Rule of Law (CDDRL), Stanford University.

Lessing, B. 2017. Making Peace in Drug Wars: Crackdowns and Cartels in Latin America. Cambridge University Press.

Lessing, B. 2021. “Conceptualizing Criminal Governance.” Perspectives on Politics 19 (3): 854–73.

Lessing, B., and G. D. Willis. 2019. “Legitimacy in Criminal Governance: Managing a Drug Empire from Behind Bars.” American Political Science Review 113 (2): 584–606.

López Stewart, C., M. G. Lara, L. D. Herrera Amighetti, L. S. Wissow, M. I. Gutierrez, I. Levav, and M. Maddaleno. 2000. “Parenting and Physical Punishment: Primary Care Interventions in Latin America.” Revista Panamericana de Salud Pública 8 (4): 257–67.

Lum, C., M. Stoltz, C. S. Koper, and J. A. Scherer. 2019. “Research on Body-Worn Cameras: What We Know, What We Need to Know.” Criminology & Public Policy 18 (1): 93–118.

Magaloni, B., E. Franco-Vivanco, and V. Melo. 2020. “Killing in the Slums: Social Order, Criminal Governance, and Police Violence in Rio de Janeiro.” American Political Science Review 114 (2): 552–72.

Magaloni, B., and L. Rodriguez. 2020. “Institutionalized Police Brutality: Torture, the Militarization of Security, and the Reform of Inquisitorial Criminal Justice in Mexico.” American Political Science Review 114 (4): 1013–34.

Malone, M. F. 2010. “Does Crime Undermine Public Support for Democracy? Evidence from Central America and Mexico.” September 4, 2010. https://ssrn.com/abstract=1668789 or http://dx.doi.org/10.2139/ssrn.1668789

Mampilly, Z. C. 2012. Rebel Rulers: Insurgent Governance and Civilian Life during War. Cornell University Press.

Mancha, A., J. Monteiro, and M. Weintraub. No date. “A New Path to Police Reform? Effects of a New Police Squad in Ceará, Brazil.” Unpublished manuscript.

Martinez, O., E. Lemus, C. Martinez, and D. Sontag. 2016. “Killers on a Shoestring: Inside the Gangs of El Salvador.” New York Times, November 20, 2016.

McDermott, J. 2020. “GameChangers 2019: Latin America’s Top 10 Criminal Groups.” Insight Crime, January 22, 2020. McDougal, T., D. A. Shirk, R. Muggah, and J. H. Patterson. 2013. “The Way of the Gun: Estimating Firearms Traffic across the U.S.Mexico Border.” Igarapé Institute and University of San Diego.

Meléndez-Sánchez, M., and B. Winter. 2024. “The Bukele Model: Why It’s Hard to Replicate.” America’s Quarterly, Podcast, July 11, 2024.

Melnikov, N., C. Schmidt-Padilla, and M. M. Sviatschi. 2020. “Gangs, Labor Mobility and Development.” NBER Working Paper 27832, National Bureau of Economic Research, Cambridge, MA.

Melnikov, N., C. Schmidt-Padilla, and M. M. Sviatschi. No date. “Gangs Crackdowns and Development.” Unpublished manuscript.

Mexico, Mexico City, Gobierno de la Ciudad de México. 2024a. Coordinación interinstitucional y ciudadana. Informe de Gobierno CDMX. https://informedegobierno.cdmx.gob.mx/acciones/coordinacion-interinstitucional-y-ciudadana

Mexico, Mexico City, Gobierno de la Ciudad de México. 2024b. Destacamos reducción en delitos de alto impacto. https://gobierno. cdmx.gob.mx/noticias/destacamos-reduccion-en-delitos-de-alto-impacto

Mexico, Mexico City, Secretaría de Seguridad Ciudadana de la Ciudad de México. 2024a. Subsecretaría de Inteligencia e Investigación Policial.

Mexico, Mexico City, Secretaría de Gobierno de la Ciudad de México. 2024b. Boletín SECGOB 2024. Miller, A. R., and C. Segal. 2019. “Do Female Officers Improve Law Enforcement Quality? Effects on Crime Reporting and Domestic Violence.” Review of Economic Studies 86 (5): 2220–47.

Monteiro, J., and M. Weintraub. 2025. In World Bank Policy Document on Police and Organized Crime. World Bank, Washington, DC. Unpublished.

Munyo, I., and M. A. Rossi. 2020. “Police-Monitored Cameras and Crime.” Scandinavian Journal of Economics 122 (3): 1027–44.

National Gang Center. No date. Mentoring Youth to Prevent Gang Violence. U.S. Department of Justice, Office of Juvenile Justice and Delinquency Prevention.

Nicas, J., P. Bonnefoy, and J. Bartlett. 2025. “Maduro Government Accused of Dark New Tactic: Assassinations.” New York Times, February 10, 2025.

O Globo. 2024. “Um terço da Amazônia Legal está tomado por facções criminosas, aponta levantamento.” December 11, 2024. Open Society Foundations. 2013. “Justice in DRC: Mobile Courts Combat Rape and Impunity in Eastern Congo.” Open Society Foundations.

Owens, E., D. Weisburd, K. L. Amendola, and G. P. Alpert. 2018. “Can You Build a Better Cop? Experimental Evidence on Supervision, Training, and Policing in the Community.” Criminology & Public Policy 17 (1): 41–87.

Papadovassilakis, A. 2023. El Salvador’s (Perpetual) State of Emergency: How Bukele’s Government Overpowered Gangs. Technical report, Insight Crime.

Parker, S. T., M. B. Ross, and S. Ross. 2024. “Driving Change: Evaluating Connecticut’s Collaborative Approach to Reducing Racial Disparities in Policing.” NBER Working Paper 32692, National Bureau of Economic Research, Cambridge, MA.

Pérez, O. J. 2003. “Democratic Legitimacy and Public Insecurity: Crime and Democracy in El Salvador and Guatemala.” Political Science Quarterly 118 (4): 627–44.

Pérez Esparza, D., and E. Weigend. 2015. “The Illegal Flow of Firearms from the United States to Mexico: A State-Level Trafficking Propensity Analysis.” Journal of Trafficking, Organized Crime and Security 1 (2): 115–25.

Pérez Ricart, C. A. 2021. “Mexico and Central America: Flexibility and Frameworks.” Chapter 6 in The International Drug Control Regime in the Twenty-First Century, edited by A. Idler and J. C. Garzón, 157–83. Hull Publishers.

Pérez Ricart, C. A. 2024. “The Nexus of Firearms and Organized Crime in Latin America.” Chapter 4 in Illicit Firearms Markets and Organized Crime: Global, Regional, and Local Perspectives, edited by D. Bright. Oxford University Press.

Perez-Vincent, S. M., D. Puebla, N. Alvarado, L. F. Mejía, X. Cadena, S. Higuera, and J. D. Niño. 2024. The Costs of Crime and Violence: Expansion and Update of Estimates for Latin America and the Caribbean. Washington, DC: Inter-American Development Bank. Polinsky, A. M., and P. N. Riskind. 2019. “Deterrence and the Optimal Use of Prison, Parole, and Probation.” Journal of Law and Economics 62 (2): 347–71.

Ponce, A. F. 2019. “Violence and Electoral Competition: Criminal Organizations and Municipal Candidates in Mexico.” Trends in Organized Crime 22 (2019): 231–54.

Prem, M., A. F. Rivera, D. A. Romero, and J. F. Vargas. 2021. “Selective Civilian Targeting: The Unintended Consequences of Partial Peace.”  https://ssrn.com/abstract=3203065 or http://dx.doi.org/10.2139/ssrn.3203065

Prem, M., J. F. Vargas, and O. Namen. 2023. “The Human Capital Peace Dividend.” Journal of Human Resources 58 (3): 962–1002. Rivera, R. G., and B. A. Ba. 2023. “The Effect of Police Oversight on Crime and Misconduct Allegations: Evidence from Chicago.” Review of Economics and Statistics 1–45. https://doi.org/10.1162/rest_a_01377

Rodríguez Mega, E. 2024. “Why Are So Many Mexican Election Candidates Getting Killed?” New York Times, May 25, 2024.

Rojas-Flores, L., S. Herrera, J. M. Currier, E. Y. Lin, R. Kulzer, and D. W. Foy. 2013. “We Are Raising Our Children in Fear: War, Community Violence, and Parenting Practices in El Salvador.” International Perspectives in Psychology: Research, Practice, Consultation 2 (4): 269–85.

Rose, E. K. 2021. “Who Gets a Second Chance? Effectiveness and Equity in Supervision of Criminal Offenders.” Quarterly Journal of Economics 136 (4): 1199–1253.

Rose, E. K., and Y. Shem-Tov. 2021. “How Does Incarceration Affect Reoffending? Estimating the Dose-Response Function.” Journal of Political Economy 129 (12): 3302–56.

Saiz, M. 2024. “The Rise of Global Crime Networks: European Mafias in the Americas.” Insight Crime, November 18, 2024.

Sampó, C., and V. Troncoso. 2024. “El tren de Aragua: La transnacionalización del crimen organizado a través del tráfico de migrantes.” Análisis Político 37 (108): 147–76.

Sandefur, J., and B. Siddiqi. 2013. “Delivering Justice to the Poor: Theory and Experimental Evidence from Liberia.” In World Bank Workshop on African Political Economy, Volume 20, 1–61. Washington, DC: World Bank.

Schargrodsky, R., and S. Tobon. 2025. In World Bank Policy Document on Prisons and Organized Crime. World Bank, Washington, DC. Unpublished.

Shem-Tov, Y., S. Raphael, and A. Skog. 2024. “Can Restorative Justice Conferencing Reduce Recidivism? Evidence from the Make-itRight Program.” Econometrica 92 (1): 61–78.

Soares, R. R. 2004. “Crime Reporting as a Measure of Institutional Development.” Economic Development and Cultural Change 52 (4): 851–71.

Soares, R. R. 2006. “The Welfare Cost of Violence across Countries.” Journal of Health Economics 25 (5): 821–46.

Solmirano, C. 2023. “Behind a Rise in Latin America’s Violent Crime, a Deadly Flow of Illegal Guns.” Americas Quarterly, May 15, 2023. Stoddard, G., D. J. Fitzpatrick, and J. Ludwig. 2024. “Predicting Police Misconduct.” NBER Working Paper 32432, National Bureau of Economic Research, Cambridge, MA.

Strang, J., T. Babor, J. Caulkins, B. Fischer, D. Foxcroft, and K. Humphreys. 2012. “Drug Policy and the Public Good: Evidence for Effective Interventions.” The Lancet 379 (9810): 71–83.

Sviatschi, M. M. 2022. “Spreading Gangs: Exporting US Criminal Capital to El Salvador.” American Economic Review 112 (6): 1985–2024. Tamayo Gomez, C. 2020. “Organised Crime Governance in Times of Pandemic: The Impact of COVID-19 on Gangs and Drug Cartels in Colombia and Mexico.” Bulletin of Latin American Research 39 (S1): 12–15.

Thomson, F., P. Meehan, and J. Goodhand. 2024. The political economy of illicit drug crops. The Journal of Peasant Studies, 51(4), p. 763-800.

Tobón, S. 2022. “Do Better Prisons Reduce Recidivism? Evidence from a Prison Construction Program.” Review of Economics and Statistics 104 (6): 1256–72.

UNDP (United Nations Development Programme). 2013. Regional Human Development Report 2013–2014: Citizen Security with a Human Face: Evidence and Proposals for Latin America. New York: UNDP (November).

UNDP (United Nations Development Programme). 2014. Evaluation of UNDP’s Support to Mobile Courts in Sierra Leone, Democratic Republic of the Congo, and Somalia. UNDP.

UNICEF (United Nations Children’s Fund). 2018a. Disciplina violenta en América Latina y el Caribe. UNICEF.

UNICEF (United Nations Children’s Fund). 2018b. “Comprehensive Laws and Social Changes Are Key to Eradicate Physical Punishment Suffered by 1 Out of Every 2 Children in Latin America and the Caribbean.” Press release, April 25, 2018.

United States, Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF). 2022a. “Firearms Trace Data: Central America 2020.” ATF. United States, Bureau of Alcohol, Tobacco, Firearms and Explosives(ATF). 2022b. National Firearms Commerce and Trafficking Assessment: Firearms in Commerce. ATF.

United States, Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF). 2023. 2021 Annual Firearms Manufacturing and Export Report. ATF.

United States, District of Columbia (Washington, DC). Metropolitan Police Department Organizational Chart. 2011. https://mpdc.dc.gov/sites/default/files/dc/sites/mpdc/publication/attachments/Say%20No%20to%20Gangs_SPA.pdf

UNODC (United Nations Office on Drugs and Crime). 2023a. World Drug Report. United Nations.

UNODC (United Nations Office on Drugs and Crime). 2023b. Global Report on Cocaine 2023–Local Dynamics, Global Challenges. UNODC. UNODC (United Nations Office on Drugs and Crime). 2024. World Drug Report. United Nations.

Uprimny, R., S. Chaparro, and L. F. Cruz. 2017. Delitos de drogas y sobredosis carcelaria en Colombia. Documento 37, Dejusticia & Colectivo de Estudios Drogas y Derecho – CEDD

Vargas, J. F., M. Rounseville, J. L. Silva, D. Ortiz, and A. Hathaway. No date. “The Twists and Turns in the Road to Justice in Colombia.” Unpublished.

Vélez, J. 2025. “El Parque Nacional La Paya, otra víctima del poder de los Comandos de la Frontera en Putumayo.” El País, February 22, 2025.

Venezuela Investigative Unit. 2023. “In Lara, Venezuela, Criminal ‘Colectivos’ Control Public Services.” InSight Crime, April 25, 2023. https://insightcrime.org/news/in-lara-venezuela-criminal-colectivos-control-public-services/

Vidrine, D. J., F. E. Fletcher, H. E. Danysh, S. Marani, J. I. Vidrine, S. B. Cantor, and A. V. Prokhorov. 2012. “A Randomized Controlled Trial to Assess the Efficacy of an Interactive Mobile Messaging Intervention for Underserved Smokers: Project Action.” BMC Public Health 12 (1): 696.

Visconti, G. 2020. “Policy Preferences after Crime Victimization: Panel and Survey Evidence from Latin America.” British Journal of Political Science 50 (4): 1481–95.

Wallace, A. 2020. “Muerte de George Floyd: Cuál es la situación de la población negra en América Latina (y el parecido a la de EE.UU.).” BBC News Mundo, June 9, 2020. https://www.bbc.com/mundo/noticias-america-latina-52969557

Weigend Vargas, E., Z. Hans, D. J. Wiebe, and J. E. Goldstick. 2023. “Firearm Manufacturing and Imports in the U.S. and Their Association to Firearm Homicides in Central America and the Caribbean, 1991–2019.” Injury Prevention 30 (5): 381–86. Williams, J., and D. Weatherburn. 2022. “Can Electronic Monitoring Reduce Reoffending?” Review of Economics and Statistics 104 (2): 232–45.

Williams, Jr., M. C., N. Weil, E. A. Rasich, J. Ludwig, H. Chang, and S. Egrari. 2021. “Body-Worn Cameras in Policing: Benefits and Costs.” Working Paper No. 2021-38, Becker Friedman Institute for Economics, University of Chicago. WHO (World Health Organization). 2006. “Mobile Clinics Provide Health Care to People in Some of the World’s Worst Crises.” http:// www.who.int/hac/donorinfo/2016/mobile_clinics/en/.

World Bank. 2006. Honduras–Judicial Branch Modernization. Washington, DC: World Bank. World Bank. 2008. Guatemala–Judicial Reform Project. http://documents.worldbank.org/ curated/en/305631468252615058/pdf/32128a. pdf.

World Bank. 2010. Crime and Violence in Central America: Volume II (Report No. 56781-LAC). Washington, DC: World Bank, Central America Unit, Poverty Reduction and Economic Management Unit, Latin America and the Caribbean Region.

Yashar, D. J. 2018. Homicidal Ecologies: Illicit Economies and Complicit States in Latin America. New York: Cambridge University Press. Zermeño, F. 2024. ¿Por qué y cómo incorporar el enfoque de género y feminista en las políticas de seguridad ciudadana? January 25, 2024, Segunda Semana de la Seguridad Ciudadana, PNUD/USAID.