Lessons from randomized evaluations on managing and preventing crime, violence, and conflict.
What are the most promising strategies for reducing crime, violence, and conflict? The past decade has seen a dramatic expansion in the experimental literature designed to help answer this question. Moving beyond evaluations of individual programs, these studies seek to advance our understanding of what drives individuals and groups towards violence and conflict and the levers at our disposal for their reduction.
This evidence review, prepared by staff at the Abdul Latif Jameel Poverty Action Lab (J-PAL) and Innovations for Poverty Action (IPA) for the Department for International Development (DFID), offers a broad review of the expansion of this literature and seeks to capture some of the emerging insights from across these studies. The review has been prepared as part of J-PAL and IPA’s Governance, Crime and Conflict Initiative (GCCI), a £12-million investment by DFID launched in 2017 to produce new research on effective policies to promote peace and good governance, reduce crime, and support individuals and communities recovering from conflict.
We reviewed the existing experimental and rigorous quasi-experimental literature for studies that help to answer six questions identified in conversation with DFID staff1:
- What does and does not work in policing, including community policing?
- What does and does not work in terms of justice provision, including criminal justice and corrections/prisons?
- What do RCTs tell us about how to reduce the violent behavior of individuals in high-crime or conflict settings?
- What do RCTs tell us about how violent organizations/groups make strategic choices between violent and non-violent action?
- What do RCTs tell us about what works in peacebuilding, reconciliation and community-based/alternative dispute resolution?
- Does RCT evidence demonstrate that including women in interventions increases stability, conflict resolution, dispute resolution or violence reduction outcomes?
1 The views expressed here are those of the authors and do not necessarily reflect those of the Department for International Development.
Innovations for Poverty Action (IPA) is a research and policy non-profit that discovers and promotes effective solutions to global poverty problems. IPA brings together researchers and decision-makers to design, rigorously evaluate, and refine these solutions and their applications, ensuring that the evidence created is used to improve the lives of the world’s poor. Since our founding in 2002, IPA has worked with over 575 leading academics to conduct over 650 evaluations in 51 countries. Future growth will be concentrated in focus countries, such as Myanmar, where we have local and international staff, established relationships with government, NGOs, and the private sector, and deep knowledge of local issues.
Los grupos armados urbanos—especialmente las bandas criminales—son un obstáculo para el desarrollo económico y la consolidación de entornos pacíficos en muchas ciudades del mundo. Frecuentemente, estos grupos actúan como cuerpos estatales, ejecutando acciones típicamente gubernamentales como la resolución de disputas, la provisión de justicia y seguridad, la prevención del crimen, la regulación de mercados y la recolección de impuestos. En Medellín, Colombia, en alianza con la alcaldía de la ciudad, un grupo de investigadores diseñó una intervención para aumentar la participación del gobierno municipal en la provisión de servicios públicos y evaluar el impacto de ésta sobre la legitimidad del Estado y las bandas criminales.
We evaluate the impact of a large-scale information and mobilization intervention designed to improve health service delivery in rural Uganda by increasing citizens’ ability to monitor and apply bottom-up pressure on underperforming health workers. Modeled closely on the landmark “Power to the People” study (Björkman and Svensson, 2009), the intervention was undertaken in 376 health centers in 16 districts and involved a three-wave panel of more than 14,000 households. We find that while the intervention had a modest positive impact on treatment quality and patient satisfaction, it had no effect on utilization rates or health outcomes (including child mortality). We also find no evidence that the channel through which the intervention affected treatment quality was citizen monitoring. The results hold in a wide set of pre-specified subgroups and also when, via a factorial design, we break down the complex intervention into its two most important components. Our findings cast doubt on the power of information to foster community monitoring or to generate improvements in health outcomes, at least in the short term.
We embed a field experiment in a nationwide recruitment drive for nurses in Zambia to test whether career benefits attract talent at the expense of prosocial motivation. We randomize the offer of career benefits at the recruitment stage. In line with common wisdom, treatment attracts less prosocial applicants. However, the trade-off only exists at low levels of talent; the marginal applicants in treatment are more talented and equally pro-social. These are hired, and they perform better at every step of the chain: they deliver more services, promote institutional childbirth, and reduce child malnutrition by 25% in the communities they serve.
Voter mobilization campaigns face trade-offs in young democracies. In a large-scale experiment implemented in 2013 with the Kenyan Electoral Commission (IEBC), text messages intended to mobilize voters boosted participation but also decreased trust in electoral institutions after the election, a decrease that was stronger in areas that experienced election-related violence, and for individuals on the losing side of the election. The mobilization backfired because the IEBC promised an electronic voting system that failed, resulting in manual voting and tallying delays. Using a simple model, we show signaling high institutional capacity via a mobilization campaign can negatively affect beliefs about the fairness of the election.
In 2016 the city of Bogotá doubled police patrols and intensified city services on high-crime streets. They did so based on a policy and criminological consensus that such place-based programs not only decrease crime, but also have positive spillovers to nearby streets. To test this, we worked with Bogotá to experiment on an unprecedented scale. They randomly assigned 1,919 streets to either 8 months of doubled police patrols, greater municipal services, both, or neither. Such scale brings econometric challenges. Spatial spillovers in dense networks introduce bias and complicate variance estimation through “fuzzy clustering.” But a design-based approach and randomization inference produce valid hypothesis tests in such settings. In contrast to the consensus, we find intensifying state presence in Bogotá had modest but imprecise direct effects and that such crime displaced nearby, especially property crimes. Confidence intervals suggest we can rule out total reductions in crime of more than 2–3% from the two policies. More promising, however, is suggestive evidence that more state presence led to an 5% fall in homicides and rape citywide. One interpretation is that state presence may more easily deter crimes of passion than calculation, and place-based interventions could be targeted against these incredibly costly and violent crimes.
In Mexico, we have continued our global tradition of rigorous, applicable research by building foundational research capacity and conducting evaluations in areas of pressing national concern. Examples of our work detailed in this brief promising insights into everyday issues that affect the lives of people in Mexico.
Designers and funders of payments for ecosystem services (PES) programs have long worried that payments flow to landholders who would have conserved forests even without the program, undermining the environmental benefits (“additionality”) and cost-effectiveness of PES. If landholders self-select into PES programs based on how much conservation they were going to undertake anyway, then those who were planning to conserve should always enroll. This paper discusses the less-appreciated fact that enrollment is often based on other factors too. The hassle of signing up or financial costs of enrollment (e.g., purchasing seedlings) can affect who participates in a PES program. These enrollment costs reduce overall take-up, and, importantly, they can also influence the composition of landholders who select into the program—and thereby the program’s environmental benefits per enrollee. Enrollment costs can increase a program’s benefits per enrollee if they are systematically higher for (and thus deter enrollment by) landholders who would have conserved anyway. Alternatively, enrollment costs can dampen per-enrollee benefits if their correlation with status-quo conservation is in the opposite direction. We illustrate these points with evidence from two studies of randomized trials of PES programs aimed at increasing forest cover in Uganda and Malawi. We also discuss how in other sectors, such as social welfare, policy designers have purposefully adjusted the costs of program enrollment to influence the composition of participants and improve cost-effectiveness. We propose that these ideas for targeting could be incorporated into the design of PES programs.
In 2008, Uganda granted hundreds of small groups $400/person to help members start individual skilled trades. Four years on, an experimental evaluation found grants raised earnings by 38% (Blattman, Fiala, Martinez 2014). We return after 9 years to find these start-up grants acted more as a kick-start than a lift out of poverty. Grantees' investment leveled off; controls eventually increased their incomes through business and casual labor; and so both groups converged in employment, earnings, and consumption. Grants had lasting impacts on assets, skilled work, and possibly child health, but had little effect on mortality, fertility, health or education.
Governments are tasked with delivering basic services such as education, security, and infrastructure, but access to and quality of these services is often undermined by poor oversight, corruption, and lack of community participation.
While countless programs aim to address these issues, their effectiveness is often not clear. IPA’s governance research investigates ways to increase the performance of the government institutions that serve as the foundation for development. Our research teams have shed light on pressing questions including how to reduce vote-selling, how to increase the demand for government accountability, how technology can be used to reduce corruption, and how information can improve voting behavior, but many unanswered questions remain.
Un equipo de investigadores de Innovations for Poverty Action, en colaboración con el Gobierno Colombiano, desarrolló una auditoría aleatoria de dos de los programas sociales más grandes del país—Sistema de Identificación de Potenciales Beneficiarios de Programas Sociales (Sisbén) y Más Familias en Acción (MFA)—para medir cómo el estatus social de los ciudadanos y los factores políticos locales afectan la eficiencia de los servidores municipales en el procesamiento y atención de las solicitudes ciudadanas.
- La tasa de respuesta de las llamadas a las alcaldías municipales del territorio nacional y las alcaldías locales en Bogotá fue baja: cerca del 65% de los solicitantes recibieron una respuesta en hasta seis intentos.
- Un número importante de alcaldías (148/618 o 23%) son inaccesibles por teléfono durante horas de servicio al ciudadano.
- Era menos probable que las llamadas fueran respondidas en horas de la tarde (después de almuerzo) que, en horas de la mañana, proporcionando una leve evidencia de ausentismo. » Dentro de las llamadas respondidas, menos del 50% de los solicitantes recibieron información correcta sobre Más Familias en Acción o Sisbén.
- El equipo investigador encontró evidencia moderada de discriminación según el acento regional, la clase socioeconómica y el estado migratorio cuando los solicitantes preguntaban cómo acceder a beneficios de MFA y Sisbén.
Do social networks matter for the adoption of new political communication technologies? We collect complete social network data for sixteen Ugandan villages where an innovative reporting mobile platform was recently introduced, and show robust evidence of peer effects on technology adoption. However, peer effects were not observed in all networks. We develop a formal model showing that while peer effects facilitate adoption of technologies with minimal externalities (like agricultural practices), it can be more difficult for innovations with significant positive externalities to spread through a network. Early adopters might exaggerate benefits, leading others to discount information about the technology’s value. Thus, peer effects are likely to emerge only where informal institutions support truthful communication. We show that the observable implications of our model are borne out in the data. These impediments to social diffusion might help explain the slow and varied uptake of new political communication technologies around the world.
Gender gaps in participation and representation are common in new democracies, both at the elite level and at the grassroots. We investigate efforts to close the grassroots gender gap in rural Ghana, a patronage-based democracy in which a dense network of political party branches provides the main avenue for local participation. We report results from a randomized field experiment to address norms against women's participation and encourage women's participation ahead of Ghana's December 2016 elections. The treatment is a large community meeting presided over by the traditional chief, known locally as a durbar. We find null results. The treatment was hampered in part by its incomplete implementation, including by local political party leaders who may have feared an electorally-risky association with a controversial social message. The study emphasizes the importance of social norms in explaining gender gaps in grassroots politics in new democracies and contributes new evidence on the limitations of common civic education interventions used in the developing world.
Can democratic elections reduce rent extraction by public decision makers? Existing research suggests that reelection incentives can reduce the embezzlement of public funds. This paper examines three additional mechanisms through which democratic elections could have an impact on embezzlement, even in the absence of reelection incentives: (1) electoral selection effects, (2) social norms and norm enforcement, and (3) citizens’ trust in decision makers. Evidence from an experiment with 472 groups of citizens in rural Burkina Faso suggests that electoral selection favors benevolent candidates. Furthermore, elections increase citizens’ willingness to punish corrupt decision makers, even if their ability to do so remains unchanged. However, these beneficial effects of elections are offset by an unexpected adverse effect: elections cause citizens to trust decision makers more than they should be trusted. These findings have important implications for the role of information in electoral democracy.