Work in progress

Working papers

• Violence and coercive policing: dynamics and consequences of the overpolicing-underpolicing paradox in Brazil’s largest city

  Author: Thiago R. Oliveira

  Asked to revise and resubmit, American Journal of Sociology.

  See abstract

  Abstract: Policing approaches that rely on repeated intrusion in the lives of citizens can be accompanied by public cynicism about the ability of legal institutions to ensure public safety. This study provides a quantitative assessment of the dynamics of phenomenon of overpolicing and underpolicing over time. It does so in the context of one of the largest cities in the Global South, with a focus on shifts in support for personal use of violence via diminished perceptions of legitimacy. Drawing upon a three-wave longitudinal survey representative of eight neighborhoods in São Paulo, Brazil, I demonstrate that perceptions of overpolicing and underpolicing (a) mutually reproduce each other over time, (b) vary significantly by neighborhood, (c) increase after aggressive police stops, (d) undermine police legitimacy, and (e) contribute to more favorable attitudes towards the acceptability of violence. This study provides further evidence on the costs of coercive policing, with significant implications for people’s recognition of the ruling power of legal authority.

  Manuscript available on SocArXiv.

  Replication materials here.

• It’s a matter of time: assessing the impact of aggressive police stops on (dis)trust in legal authority

  Author: Thiago R. Oliveira.

  See abstract

  Abstract: Does the experience of being stopped by the police (including being stopped by the police at gunpoint) have a negative effect on trust in legal authority over time? Previous research suggests a link between negatively-experienced police stops and distrust of legal institutions. Yet, we lack clear evidence on the existence and dynamics of any putative causal effect. To address this gap, I draw on a three-wave longitudinal survey of adults who reside in São Paulo, Brazil, and examine the impact of police stops and police stops at gunpoint on three aspects of trust in legal authority (attitudinal change in perceptions of police fairness, perceptions of overpolicing, and cynicism about police protection). In the context of a multi-period difference-in-differences design, I rely on (i) a matching framework for panel data that assumes a dynamic treatment adoption to estimate short-term effects of a recent police stop, and (ii) a doubly robust estimator that assumes a staggered treatment adoption to assess the long-term effects of a first experience of being stopped by the police (at gunpoint). I conclude, overall, that it seems unlikely that general police stops damage trust in legal authority in São Paulo over time. However, a recent aggressive police stop with officers pointing a gun at the citizen does seem to have a negative short-term effect on perceived police fairness, while the first experience of a police stop at gunpoint among people with no previous contact seems to have a positive long-term effect on perceived overpolicing. This study contributes to a growing international literature that describes the social costs of aggressive policing strategies.

  Manuscript available on SocArXiv.

  Replication materials here.

• Violência como fator socializador na construção da legitimidade da polícia: um estudo com adolescentes de São Paulo, Brasil

  (Violence as a socialising factor shaping police legitimacy: a study with adolescents in São Paulo, Brazil)

  Authors: Renan Theodoro, Debora Piccirillo, Aline Gomes, & Thiago R. Oliveira.

  Under first review at Análise Social.

  See abstract

  Abstract. This paper investigates how adolescents are socialised to accept or reject police violence and abuse of power, and how these dispositions influence police legitimacy. Data came from a survey with 724 participants born in 2005, residents of the city of São Paulo, Brazil. Direct and indirect effects of experiences of violence and police contact over adolescents evaluations of police were estimated using structural equation modelling. Results indicate that aggressive and illegal policing, as well as exposure to violence in the neighborhood erode confidence in policing.

Manuscripts in preparation

• Socialization through violence: exposure to neighborhood and police violence and the development of legal legitimacy among adolescents in São Paulo

  Authors: Thiago R. Oliveira, Jon Jackson, Renan Theodoro, Debora Piccirillo, & Rick Trinkner.

  See abstract

  Abstract: We examine the influence of exposure to neighbourhood and police violence on the legal socialisation of adolescents aged 11 to 14 years living in the city of São Paulo, Brazil. In a unique context of idiosyncratic and violence policing where the state's ability to control crime is low, we discuss the extent to which being frequently exposed to neighbourhood crime and violence (e.g., listening to gunshots, witnessing or hearing about citizens carrying guns, being robbed, or selling drugs), relatively forcible police behaviour (e.g., police stops or arrests), or aggressive and improper forcible police behaviour (e.g., officers assaulting a member of the public) undermine the development of judgements about the legitimacy of the law. Drawing on data from a cohort-based, four-wave longitudinal survey of 2005-born adolescents living in São Paulo from 2016 to 2019, we use growth curve models to estimate developmental trajectories of legitimacy beliefs. Results indicate that exposure to police brutality can damage the process of legal socialisation, and that adolescents who study at schools where most other students are frequently exposed to neighbourhood violence tend to develop more antagonist views about the authority of the law.

  Replication materials here.

• Economic inequality and the spatial distribution of stops and searches: evidence from London

  Authors: Joel Suss & Thiago R. Oliveira.

  See abstract

  Abstract: We study the spatial distribution of stops and searches in London in 2019, and assess the extent to which salient, spatially-granular economic inequality is associated with a higher concentration of searches net of previous crime rates and spatial effects. We interpret stop and search as a tool of social control and demonstrate the degree to which economic inequality contributes to filter out locations where members of the public might stopped and searched by police officers. We use data at the level of Lower Super Output Areas (LSOAs) and draw on a novel measure of local inequality based on real and estimated housing values. Our analytic strategy involves estimating Spatial Durbin Models to account for spatial dependency and negative binomial regression models to estimate the number of stops and searches by LSOA. Results suggest that local economic inequality is strongly associated with the number of stops and searches, even taking into account spatial effects and previous crime rates. We also find that the effect of inequality depends on the level of affluence: the better off a location is, the stronger the effect of inequality on police stops and searches. We conclude by extending the cultural meaning of stop and search as a tool of social control. Police decisions regarding where to stop and search members of the public could be an additional mechanism that contributes to high levels of ethnic and social disproportionality in stop and search.

• Self-legitimacy and commitment to democratic policing in Brazil: assessing the support for police militarization among São Paulo military police officers

  Authors: Viviane Cubas, Frederico Castelo Branco, Thiago R. Oliveira, & Justice Tankebe.

  See abstract

  Abstract: To what extent do police officers who are confident of their own authority support (or oppose) police militarization? Previous work has established possible links between police officers’ self-legitimacy and their commitment to democratic policing, which is based on principles of citizen participation, equity and responsiveness. Understood as the process through which policing activities take more and more characteristics of the military culture and behavior, police militarization premises on the idea that police officers should be trained as if they were going to war. Based on this nearly diametrical distinction and given the positive association found between self-legitimacy and commitment to democratic policing, we ask whether self-legitimacy is negatively associated with support for police militarization. Using data from a survey of officers from the São Paulo Military Police in 2016, we assess the extent to which measures of self-legitimacy are correlated with such support controlling for agents’ identification with military values – we also assess the extent to which self-legitimacy mediates the effects or other variables on support for police militarization, such as perceived public support and perceived distributive justice within the organization. Results of structural equation models indicate that while identification with military values mediate some statistical effects of perception of work environment on support for police militarization, self-belief in authority vested in them does not.

• Pastas e prontuários do “Complexo do Tatuapé” (São Paulo/SP – 1990-2006): alterações nos padrões da justiça juvenil após o ECA

  (Records from the Tatuapé Complex (São Paulo/SP - 1990-2006): juvenile justice’s changing patterns after ECA)

  Authors: Thiago R. Oliveira, Bruna Gisi, & Marcos César Alvarez

  See abstract

  Abstract: