ABSTRACT
Over the past two decades, most Latin American countries have struggled to create or reform their police forces while simultaneously confronting sharp increases in violent crime. Reformers have gravitated towards community-oriented policing practices, which aim to rely on preventive tactics and build closer ties between police officers and the public. The implementation of such policing practices can be daunting, particularly for countries with authoritarian histories of policing and/or military repression. When confronting rising crime rates, some new democracies are tempted to revert to authoritarian policing practices and circumvent constitutional safeguards on human rights and civil liberties. Still, some countries have resisted such temptations, and seriously invested in community-oriented policing practices. However, do these community-oriented policing practices lead citizens to trust the police more? Or does public trust depend more heavily on results – particularly in countries undergoing crises in public security? To answer these questions, we analyze public trust in the police in 17 Latin American countries through a series of regression analyses of the Latin American Public Opinion Project’s (LAPOP) 2016–2017 survey data. We find that community-oriented policing practices tend to garner more public trust, but that perceptions of police effectiveness are equally important.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Notes
1 The authors would like to thank LAPOP and its major supporters (the United States Agency for International Development, the Inter-American Development Bank, and Vanderbilt University) for making the data available. For over three decades, LAPOP has been the leading organization collecting public opinion data in Latin America, and its most recent surveys include several questions related to trust in police, experiences with and perceptions of police work, as well as perceptions and experiences with crime.
2 We identified these countries based on their records as of 2016, the year of our survey data, thus we included Nicaragua and Chile in this group. However, the police’s excessive use of force against demonstrators in 2018 and 2019 (respectively) clearly jeopardizes their inclusion in this category in the future.
3 This aligns with the procedural justice literature, which argues that if people perceive the process and the treatment they receive as fair, they have a greater tendency to accept the outcome, even if they disagree with it (Tyler Citation2006). The main argument is that compliance with the law does not rely upon the threat of force or the prior achievement of specific goals, but rather the belief that the police enforce the law equitably (Tankebe Citation2013, Tyler et al. Citation2013, Tyler et al. Citation2014).
4 In the 2016–2017 AmericasBarometer, people who identified as indigenous were significantly more likely to indicate that they had experienced discriminatory behavior because of the color of their skin at one point in their lives.
5 We report homicide rates as of 2016 as that is the year of our survey data.
6 The full 2016–2017 LAPOP questionnaire is available here: https://www.vanderbilt.edu/lapop/ab2016/AB2017-v18.0-Eng-170523_W.pdf.
7 Earlier LAPOP surveys had additional items to measure police interactions with citizens, particularly the 2008 wave. See Cruz (Citation2015) for an analysis of the impact of police abuse on public trust in police in prior LAPOP surveys.
8 The variable measuring the size of respondents’ town or city was coded as: (1) rural area; (2) small city; (3) medium city; (4) large city; (5) capital city.
9 The income scale was calculated based upon answers to the following survey items: Do you or any member of your household have any of the following possessions? TV; car; refrigerator; telephone; cell phone; computer; microwave oven; washing machine; drinking water; sewage system. Responses were coded as (1) yes and (0) no. We created an index of personal income using a means formula that included a case if there were valid responses to at least eight of the ten items.
10 We report our OLS regression results for ease of interpretation. We replicated these results using ordinal logistic regression, and the variables in these analyses had the same signs and levels of significance. Results of ordinal logistic regression are available upon request to the authors.
11 Nicaragua reports a GDP less than a quarter of that of Costa Rica, yet over the past decade its homicide rate was roughly the same. Nicaragua’s rate of violent crime is 80% lower than the other post-conflict countries in the region, Guatemala and El Salvador, despite registering lower levels of GDP per capita than these other countries.
12 Indeed, in a high-profile trial, officials from the Nicaraguan National Police and Supreme Electoral Council were charged with complicity in money laundering and drug trafficking operations (Meléndez and Orozco Citation2013).