The relationship between UPP presence and other violent crime in Rio de Janeiro

While evidence might point to an overall reduction in homicide, the data is not as compelling for an inverse relationship between the number of UPPs in the city installations and incidence of other violent crimes in Rio de Janeiro. Once again, I have accounted for population growth and decline, and all of the possible correlations I point out I did not find to be attributable to natural growth alone. I will not perform bivariate analyses as I did in the last post after taking into consideration the limitations of the small sample size, and instead compiled the following set of visual aides and graphs.

The graph below shows the number of intentional injuries (injuries inflicted by one party on another with the intent to cause bodily harm), plotted by year, in the city of Rio de Janeiro. Remember that the first UPP was installed at the end of 2008. Four units were installed in 2009, eight in 2010, and five thus far in 2011:

Sample Monthly Intentional Injury Incidence in Rio de Janeiro City, 2002-2011

Intentional injury incidence in Rio had been relatively stable, if not on an overall decline, since 2003. There is a definite, unprecedented upswing in the number of reported intentional injuries after 2009, a pattern that is not seen in other municipalities without UPPs (such as the Baixada and Grande Niteroi):

Sample Monthly Intentional Injury Incidence in Grande Niteroi, 2002-2011

Let's examine the case of Tijuca/Vila Isabel/Maracana once more. Allow me to remind you that this is a neighborhood with not one, not two, but five different UPPs, making it an excellent area to analyze. In theory, one would expect to see a significant drop in violent crime, assuming that a) violent crime is intrinsically linked to drug gangs, and b) the UPPs are serving their intended purpose. In practice, what we see is just the opposite:

Sample Monthly Incidence of Intentional Injury in Tijuca/Vila Isabel/Maracana, 2002-2011

Note the undeniable decline in intentional injury incidence from 2003-2008. 2008-2009 marks the first increase in this type of crime since 2002-2003. Although Tijuca did not receive a UPP until 2010, the rise in intentional injuries in 2008-2009 may suggest the migration of crime; because South Zone comunidades were the first to receive UPPs, some traficantes fled to comunidades in areas without a police presence. Tijuca's comunidades are not only close to the South Zone; they also are dominated by the Comando Vermelho (CV), which off Rio's three drug factions has has been the sole target of the UPPs. As you can see in the map below, the CV (red) controls nearly all of the South Zone's major comunidades, with the notable exception of Rocinha (controlled by the Amigos dos Amigos - yellow). I have delineated the South Zone comunidades with a purple line, and those of Tijuca with a blue line. Note the proximity and shared command bf the CV of the two regions, suggesting a logical migration of traficantes explused by the South Zone UPPs. The migration of crime is another important aspect of UPP installation, which I will expand upon in greater detail in an upcoming post.

Rio de Janeiro Comunidades Mapped By Ruling Drug Faction

Another relationship that merits further examination is the relationship between disappeared persons, which I discussed in the last post, and the number of bodies found by PMs that have not yet been ruled homicides. While the number of corpses recovered by polices has plummeted, the number of persons reported missing has gone up. Incidence of corpse recovered had leveled off around 2005, after a long period of decline. Incidence dropped dramatically in 2008 after the installation of the first UPP, and continued on a declining trend, while the number of disappeared persons skyrocketed for the same time period -  implying a possible correlation. While it would be easy to assume that fewer corpses found indicates a safer city, the fact that many disappeared persons are eventually ruled deceased (often by foul play), must be considered. It is quite possible that the relationship between corpses recovered and disappeared persons could indicate a) an underreporting of corpses found, b) adoption of better body disposal tactics on the part of the murderer, who now may have a 24/7 police force patrolling his comunidade, or c) an overall degradation of the quality of criminal investigations. The dual bar-line graph below is telling:

Sample Monthly Disappeared Persons vs. Corpses Recovered Incidence in Rio de Janeiro, 2002-2011

 

On most days of the week, the local media releases reports of what appears to be escalating violent confrontations between drug traffickers and military police (PM) in the extreme North Zone of Rio, where there are no UPPs. In particular, the comunidades of Morro da Serrinha and Morro do Juramento have become veritable battlegrounds over the course of the past several months. Existing conflict has been exacerbated by flight of CV traficantes from areas like Tijuca to comunidades deeper into the North Zone, and taking refuge in comunidades such as Serrinha and Juramento, which belong to their same drug faction and have no UPP presence. Unfortunately, it is extremely difficult to compile data on these areas, as the PM's administrative system underwent extensive revisions this year; regions were subdivided or expanded, and gained or lost police batallions accordingly. As statistics are published by administrative region after compiling data from the corresponding batallions, it is nearly impossible to compare crime data in areas affected by the border re-drawings in years prior to 2011.

Today, however, O Globo touched on a different source of violent crime in Juramento and Serrinha: tripartite conflict. No longer is the political landscape of the communidades defined by conflict between the police and the traficantes; the rise of the militia marks the introduction of a third party to Rio's urban war. Since its inception, the militia - which is comprised largely of current and ex-PMs - has further blurred the line between police and traficantes as it attempts to wrest control of the comunidades from drug lords. Militia now enjoy de facto authority in many of the city's comunidades, imposing vigilante rule on residents, extorting funds, inciting violence, and otherwise assuming a role very similar to that of the drug traffickers.

Clearly, the militia perpetrate violent crime in Rio de Janeiro. However, the extent to which the militia has contributed to increased violent crime is difficult to measure, due to its ties to the PM. Residents of areas under militia control are unlikely to confirm militia presence, much less denounce a crime that has been committed against them.

Underreporting of crime by both local police forces and afflicted residents is far more likely to occur in areas with little infrastructure and police presence, such as much of Rio's West and North Zones. Unfortunately, it is precisely these two areas that many fleeing traficantes are now calling home, and precisely these two areas where residents from razed comunidades are being relocated. Looking toward the future, it will be interesting to track violent crime related to the migration of crime and the proliferation of poverty on the city's outskirts. The result is the exact recipe for violence - an easily exploitable population and drug trafficking - that has plagued Rio de Janeiro since the CV exploded on to the scene in the 1960s.

The relationship between UPP presence and homicide in Rio de Janeiro

This post is a continuation of Thursday's post on statistics on violence and crime in Rio de Janeiro.

In this series of posts, I will use data taken from the Public Security Institute (ISP)'s database to address several claims about the nature of violence and crime in Rio de Janeiro following the installation of the UPPs, including:

1. Homicide is down in Rio de Janeiro because of the UPPs

2. The UPPs have not reduced crime, so much as they have displaced it to areas without UPPs

3. Car and petty theft has been reduced since UPP installation

4. Robberies of pedestrians has gone down since UPP installation
5. Neighborhoods in the immediate vicinity of UPPs have seen a reduction in crime, while neighborhoods without UPPs have not
6. Rio de Janeiro, on the whole, is safer with the UPPs

I will address the following categories of crime:

1. Homicide

2. Other violent crime

3. Assault and robberies in collective transport vehicles

3. Robberies and "furtos" (theft where face-to-face contact with the thief was not made)

4. Threats and extortion

5. Disappearances and cadavers recovered

I have also used data on police activity in the following areas:

1. Apprehension of drugs

2. Apprehension of arms

3. Imprisonments

It should be noted that the categories are independent. For example, if a rape ended in homicide, it will be listed exlusively under the category of rape, which includes "rape followed by death." It will not appear in the "homicide" category.

In addition to the limitations to the data mentioned in yesterday's post, , there were several major changes made to the way in which crimes are categorized after 2009 (to be explained in further detail later on). It's unclear why this method0logical change was made, but it's notable that it coincides with the first year of data available in which UPPs are present. If the ulterior motive of the change in data collection periods was to make data analysis on pre-2009 crime unnecessarily time-consuming, convoluted, and less precise, the ISP has succeeded with flying colors. Now, any researcher wanting to examine the relationship between UPPs and crime must account for re-categorization and elimination of certain types of crime.


Population growth and decay has been considered, and will be highlighted when relevant.

Because the sample size is very small (data is only available for the past 10 years), a note of caution must be attached to these findings; an ideal sample size is 15 or greater.

I will begin by presenting my findings on homicide. Unless otherwise indicated, it can be assumed that any significant increases in homicide to which I refer have outpaced population growth. For example, a 4% increase in homicide does not represent an absolute increase if the population of the region in question is growing at the same rate. Population growth and decay are two factors that are usually ignored by the municipal government when it publicly discusses crime rates. For the following data presentations, it is accepted that homicide and drug trafficking in Rio de Janeiro are intrinsically linked. All data points were taken from the same months for each year to ensure consistency and accuracy.


HOMICIDE IN RIO DE JANEIRO


These are my findings on homicide in the city of Rio de Janeiro. Population growth, according to the 2010 Census, is 1.4% per annum.

Sample Monthly Homicide Rate in Rio de Janeiro City , 2002-2011

Very simply, this graph shows that since 2002, homicide has been on an overall decline, well before the installation of the UPPs beginning in 2008. Note that homicide rose from 2008-2009, the first year available for post-UPP installation analysis.


This graph demonstrates initial difficulties in establishing causality between UPPs and homicide reduction; it would be difficult to prove that homicide incidence has been dropping due to increased UPP presence, and not because of some other independent variable.


So what is the relationship between the number of UPPs present in the city of Rio de Janeiro and homicide? I performed a simple linear regression demonstrating the correlation between the dependent variable, "homicide", and the independent variable "UPP units". While the data does indicate at statistically significant strong negative relationship (R=-.837) between the number UPP units and homicide (p<.01), the sample size, N, is too small to conclude that the decline in homicide may be related to UPP installation:

N(10)

β =-5.265, t(8) = -4.320, p<.01

A major source of concern for is the sharp increase in "disappeared persons" immediately following the installation of the UPPs in 2008. Take a look at this graph:

Sample Monthly Disappeared Persons Reports in Rio de Janeiro, 2002-2011

In general, disappeared person reports had been declining since 2002, until 2008 when the incidence spiked sharply. This pattern begs the question, if most homicides in Rio are related to drug trafficking and police intervention, have homicides really gone down? Are UPP-related occupations related to increased "disappearances"? For anyone that has seen Tropa de Elite 2, you know what I'm talking about.


Compare the pattern of post-2008 disappearances in Rio de Janeiro to those in Baixada Fluminense, a municipality without UPPs, but with many violent comunidades:


Sample Monthly Disappeared Persons Reports in the Baixada Fluminense, 2002-2011

While Rio de Janeiro's population has grown at just 1% per year, The Baixada's grew at 5%, so some increase in incidence of disappeared persons should be expected. Note that in the post-UPP period (2008 onward), there is no evidence of an overall increase in disappeared persons. Not so in the case of Rio, where these disappearences cannot be attributed to natural increase.

From here on out, I'll skip the equations, and simply indicate whether the correlation was statistically significant or not, how strong the relationship was, and the direction of the relationship (positive or negative).


HOMICIDE IN GRANDE NITEROI AND THE BAIXADA FLUMINENSE


Many have postulated that, with the installation of UPPs exclusively in the city of Rio de Janeiro, traficantes have migrated east and north to the Grande Niteroi and the Baixada Fluminense, driving up homicide rates. Grande Niteroi has had a negative population growth rate, while the Baixada has grown at roughly 2% per annum.


Sample Monthly Homicide Rate in the Baixada Fluminense, 2002-20011

Here, we see a similar pattern to Rio de Janeiro, albeit less obvious. Homicide in the Baixada presented a general decreasing trend from 2002-2008 prior to the UPPs. Unlike Rio de Janeiro, however, homicide incidence rose after the installation of the first UPP, and continued to climb with each subsequent year, indicating a possible migration of crime.


In this case, linear regression indicates a weak negative relationship between UPP installation and homicide incidence, that is not statistically significant (p <.01).


Data supports the claim that, in fact, UPPs may have contributed to increased homicide incidence in the Baixada Fluminense; however, it is too early to draw a conclusion. The same is not true in Grande Niteroi, where homicide has been decreasing steadily since 2002, and does not appear to have been negatively or positively affected by UPP presence. Here it must be noted that Grande Niteroi, in comparison to the Baixada, has far fewer comunidades and therefore is a less likely destination for migrating traficantes.


Data does not support the claim that homicides in the Lakes Region - a municipality in the extreme easternmost corner of Rio de Janeiro state - have risen as a result of in-migration of traficantes from Rio de Janeiro city; in fact, homicides have fallen since 2008.


HOMICIDE BY NEIGHBORHOOD


Another point of interest in homicide by region and neighborhood. Because of the micro-level analysis and relatively small, sensitive data points, I will use only graphical analysis and not include linear regressions. Unfortunately, some reporting precincts conveniently underwent a geographic reconfiguration between 2009 and 2010. Territories were assigned either more or less police departments based on their new size. These changes obfuscates research and precludes the analysis of crime in neighborhoods in which violence has reportedly gone up since the UPPs have been installed, making reports of either reduced or increased violence extremely difficult to substantiate. I have tried my best to only include neighborhoods which have not undergone these changes.


Copacabana and Leme are two geographically small neighborhoods in the wealthy South Zone. The tiny area has two UPPs, both installed in relatively small comunidades in 2009. Probably more than any other area of the city, Copacabana/Leme is praised for the successes of its UPPs. Let's take a look at how effective the UPPs have been at reducing homicide in these two neighborhoods:


Sample Monthly Homicide Rate in Copacabana/Leme 2002-2011

The UPPs in Copacabana/Leme were installed in 2009. Because the area never had a high homicide rate, it's extremely difficult to conclude that the UPPs played a role in murder reduction.

Again, the sharp increase in disappeared persons in Copacabana/Leme following the installation of the UPPs is disturbing (and is not the result of natural increase; the area has experienced -4.5% growth):


Sample Monthly Disappeared Persons Reports in Copacabana/Leme, 2002-2011

Now let's examine a neighborhood that has not one, not two, but six UPPs in its vicinity: Tijuca/Maracana/Vila Isabel. With so many UPPs (in nearly all of the area's comunidades), one would expect the homicide rate to plummet if the UPPs are having their intended affect on violent crime.

Sample Monthly Homicide Rate in Tijuca/Maracana/Vila Isabel, 2002-2011

In this neighborhood, there is no indication of an overall reduction in homicide between 2009 and 2011. The 6 UPPs were all installed in 2010, so one would expect to see a decrease in both 2010 and 2011. Instead, homicide rates for 2010 and 2011 match the overall average for 2006-2011.

Now, let's examine a neighborhood in which there are no nearby UPPs: Manguinhos/Mare/Bonsucceso:


Sample Monthly Homicide Rate in Manguinhos/Mare/Bonsucceso, 2002-2011

Note the overall decreasing trend and leveling-off, regardless of the absence of a UPP. The pattern of homicide incidence for the 2006-2011 period is very similar to that seen in Tijuca/Maracana/Vila Isabel, above.


CONCLUSION


Overall, there is no compelling evidence to indicate that the UPPs have caused a decline in homicides in the city of Rio de Janeiro. Overall homicide rates had been declining steadily in most areas since 2002, well prior the installation of the first UPP in late 2008. More data is needed in order to demonstrate any possible causal relationship between the number of UPPs in the city and a reduction in homicides.


While initial evidence from the Baixada Fluminense - a municipality which many hypothesize has become a destination for traficantes fleeing comunidades in Rio - indicates a rise in homicides, it is also too early to tell if the increase can be attributed to the flight of crime to non-UPP areas.


The increase in disappeared persons since the installation of the UPPs is disturbing. If, for example, many of the disappeared reports represent homicides, then there could be a downward bias in the homicide rates for the post-UPP period. Hopefully, the recent spike in disappearances is not attributable to increased police presence in the comunidades (see my post on how military police cover up their involvement in homicides here).


In the next post, I will discuss incidence of other violent crimes in Rio de Janeiro in the wake of the UPPs. Following that will be a set of posts on armed robbery and other types of theft. Lastly, I will demonstrate the relationship between UPP installment and the apprehension of arms and drugs, and discuss evidence of the displacement of violence to other areas of the state.

Statistics do not yet show a "safer Rio" post-UPP installation

Lately, it seems as though there has been a surge in violent crime and armed robberies in neighborhoods with UPPs. As the media in Rio is dominated by pro-government management, one can only assume that the crimes which newspapers - such as O Globo - choose to report represent a mere fraction of the crimes which actually transpire.

Regardless, Beltrame and the public security forces continue to insist that the UPPs have made major strides in reducing crime in Rio de Janeiro. In particular, the Secretary of Public Security likes to pepper their proof-positive discourse by citing reductions in car theft, homicide, petty crime, and armed assault, often without producing any corroborative statistics. Emphasis is usually placed on the benefits conferred by the UPPs on the those living on the asfalto - that is, the wealthier "legal" neighborhoods surrounding the comunidades.

The general public - especially those comprising Rio's middle class and above - tends to accept that there is a causal relationship between UPPs and crime reduction. Even within the comunidades, the UPPs enjoy some support; residents praising the units are quoted on almost a daily basis by media outlets.

Morro da Coroa borders the touristy Santa Teresa neighborhood.
Coroa's UPP was installed earlier this year. Photo credit: UPPRJ

Others are less optimistic, and postulate that the UPPs have simply "displaced" crime to other municipalities - and even other states. See this article for an overview.

However, by any statistical measure, it is far too soon to demonstrate a correlation between UPP installations and crime reduction/displacement; there simply is not enough available data (the first UPPs were installed less than three years ago). Furthermore, outside of the laboratory, in which one can control for undesired independent variables, it is nearly impossible to prove causality. Any statistician who wishes to show a causal relationship between increased UPP presence and crime reduction would be hard-pressed to substantiate his hypothesis.

In Rio de Janeiro, where accurate, consistent, and impartial data sources can be hard to come by, the challenge of demonstrating causality becomes even more daunting, and claims of UPP success appear all the more dubious. The two main sources of data on crime are the Brazilian Institute of Geography and Statistics (IBGE), and the Public Security Institute (ISP). Both institutes are run by the state government, are notoriously difficult to to navigate, and generally do not publicize public security data prior to the year 2000.

However, even given these obstacles, it is still possible to produce an initial picture of the relationship between UPPs and crime. Using the IPS database, I performed a simple bivariate analysis using data from the regional, city, zonal, and neighborhood levels, using the number of UPPs as the independent variable and categories of crime as the dependent variables. As always, there are many limitations to the data; besides the obstacles described above there were also issues with changes in data collection methodology, categories, and presentation.

At the regional level, I took data from the City of Rio de Janeiro, the Baixada Fluminense, Grande Niteroi, and the Interior. At the zonal level, I examined the Zona Sul, Centro, Zona Norte, and Zona Oeste. At the neighborhood level, I used the examples of Copacabana/Leme, Realengo/Bangu/Anchieta, and Tijuca/Vila Isabel/Maracana (in the South, West, and North zones, respectively). The basis for the choices of neighborhoods was that the sample would feature a wealthy neighborhood with an older UPP* (Copacabana/Leme), a lower-class neighborhood with an older UPP (Realengo/Bangu/Anchieta), and a middle-class neighborhood with a more recent UPP* (Tijuca/Vila Isabel/Maracana).

Population growth rates were taken into account, and all analysis is based on the most recent census data (2010). Crime data was taken from the years 2005-2011, as available. Adjustments in IPS's data collection methodology preclude the examination of several categories of crime prior to 2008, which will be addressed later.

If someone held me to gunpoint right now (add one incidence on to the "assault" category, please), I would have to say that there is very little evidence that suggests that the city of Rio de Janeiro has become safer.

Stay tuned for the cold, hard data tomorrow. As a teaser, I'll leave you with this scatterplot of homicides in Rio de Janeiro. Note that homicides began to plummet well before the installation of the city's first UPP in late 2008, and actually increased in 2009 before leveling off to pre-UPP levels in 2010.

*Copacabana/Leme and Tijuca/Vila Isabel/Marcana neighborhoods now have 2 and 5 UPPs, respectively

30 tourists assaulted in area with two UPPs

Early this morning, 30 tourists in the five-star Hotel Santa Teresa (located in the eponymous neighborhood) were assaulted and robbed by five armed bandits. This is the second reported recent incident involving tourist assaults in the area , which boasts not one, but two UPPs (Morro da Coroa and Morro dos Prazeres).

Earlier this week, a woman was found dead in Cidade de Deus, shot in the face by an unknown assailant. Cidade de Deus has had a UPP since early 2009.

Lastly and sadly, Andre Ferreira, 24, was shot execution-style by military police (PMs) staffing the UPP in Pavao-Pavaozinho-Cantagalo. PMs claimed that the officer shot in an act of "self defense", despite witness reports that Andre was unarmed and numerous civilian accounts of Andre's lack of involvement in trafficking. PMs went on to claim that Andre was a traficante, although it seems that as he was shot in the back, a positive facial ID seems unlikely. As federal deputy Marcelo Freixo (PSOL), explains, "There is no doubt that [Andre's death] was an assassination. Freixo also pointed out that the fact that Andre was only shot once indicates an execution.

Last week I had the opportunity to speak with the Captain of the Cantagalo UPP, Leonardo Nogueira. After listening to him paint an unrealistically optimistic portrait of the relationship between the UPP PMs and the residents of the comunidade (Nogueira claims former traficantes turn themselves in to Nogueira and thank him for his services), I asked him specifically about Andre's death. "That was an act of self-defense," Nogueira claimed, in-line with the official PM reports. "He was a traficante."

And these are just three isolated incidents of violence in comunidades with UPPs. Most will go unreported, biasing accounts of the success of the UPPs in combating violent crime. Obviously, the UPPs are a step in the right direction, but as of now, their effectiveness at reducing the incidence of homicide, assault, and corruption seems questionable.