Violence and the Limits of Urban Resilience Capacity: A Spatial and Spatio-temporal Analysis of Homicides in the Metropolitan Region of Campinas, Brazil (2017-2022)
This study employs spatial econometric and Bayesian hierarchical models to analyze homicides in Brazil's Metropolitan Region of Campinas (2017–2022), revealing that socioeconomic factors like income and dependency ratios, alongside strong spatial clustering, significantly influence violence patterns and necessitate territory-sensitive public security policies.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Cities are complex living systems that must constantly adapt to survive. When we talk about a city's ability to withstand shocks, from natural disasters to social unrest, we call this "urban resilience." It is not just about rebuilding after a crisis, but about the daily capacity of a community to stay together, function well, and protect its people. One of the biggest threats to this resilience is violence. When a neighborhood is unsafe, it fractures the social fabric, making it harder for residents to trust one another or plan for the future. To understand how to build safer cities, researchers need to look beyond broad city-wide statistics and examine the specific neighborhoods where violence happens, asking why it clusters in some places and not others.
A team of researchers from the Pontifícia Universidade Católica de Campinas set out to map these patterns in the Metropolitan Region of Campinas, Brazil, a sprawling area of twenty cities centered around a major industrial and technological hub. They focused on the years 2017 through 2022, analyzing 1,150 recorded homicides. Instead of treating the region as a single block of data, they broke it down into its smallest administrative pieces, known as census tracts, which are small neighborhoods containing just a few thousand people. By looking at the data at this fine level, they could see the specific local conditions that might be driving violence, such as income levels, family structures, and the age of the population.
The researchers faced a significant challenge right from the start: violence is a rare event in many neighborhoods. In some census tracts, there were so few homicides that standard statistical tools could not calculate a reliable risk rate. It is like trying to predict the weather in a town where it has only rained once in ten years; the data is too thin to draw a clear picture. To solve this, the team used two different approaches. First, they tried to smooth out the data to fill in the gaps, but this method failed for nearly half of the neighborhoods because the numbers were simply too low. They then switched to a more robust method that looked directly at the raw counts of deaths, allowing them to include almost twice as many neighborhoods in their final analysis. This shift was crucial, as it let them see the full picture of the region rather than just the parts where violence was frequent enough to be easily measured.
What they found was a clear and consistent pattern. Violence was not scattered randomly across the region; it was tightly clustered. The analysis revealed that homicides tended to happen in specific areas where they were surrounded by other areas with high rates of violence, creating "hot spots" of risk. These clusters were most prominent in the southern part of the city of Campinas, near the Viracopos airport, and in adjacent areas of the neighboring city of Hortolândia. Conversely, the northwestern and eastern edges of the region showed much lower risks. The study confirmed that these dangerous areas were not just random occurrences but were deeply connected to their surroundings, suggesting that the risk of violence in one neighborhood is influenced by the conditions in the ones right next to it.
The researchers also looked at what social factors were linked to these patterns. They found that neighborhoods with lower average household incomes consistently faced higher risks of homicide. Similarly, areas with a higher ratio of men to women, and places where a larger portion of the population was dependent on working-age adults for support, showed increased violence. These findings held true even when the researchers accounted for the fact that violence spreads between neighbors. The data suggested that economic inequality and specific demographic pressures create an environment where violence is more likely to take root and persist.
Perhaps the most striking discovery was that these patterns did not change much over time. The study compared the first three years of the period with the last three years and found that the same neighborhoods remained the most dangerous. The risk in a specific tract in 2020 to 2022 was strongly predictable based on its risk in the years before. This indicates that violence in this region is not a temporary spike caused by a single event, but a structural feature of the landscape. The places that are unsafe today are likely to remain unsafe tomorrow unless the underlying conditions change.
The study concludes that building a resilient city in this region requires more than just reacting to crime after it happens. Because the risk is so deeply tied to specific neighborhoods and persists over time, public safety policies need to be tailored to those exact locations. The researchers argue that improving urban resilience means addressing the structural inequalities that make certain areas vulnerable in the first place. By targeting resources and interventions to the specific census tracts identified as high-risk, and by understanding that these risks are linked to poverty and demographic pressures, city planners can begin to break the cycle of violence. The work demonstrates that to make a city safer, one must look closely at the ground beneath the streets, understanding that the safety of the whole depends on the stability of its most fragile parts.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.