The Origin of Violence: Poverty, Inequality, or Weakness of Justice?

As I explored the literature on violence in Colombia, I came across the 1987 report by the Commission for the Study of Violence (CEV), which I wrote about a couple of days ago. The CEV’s approach is broad and vague, fostering a framework of multiple interpretations where essentially anything one says about the phenomenon ends up being true.

If a pair of academics do not seek empirical support for their theses, and one claims that the cause of violence is corruption while the other asserts it is a matter of education, both would be somewhat right under the CEV’s perspective, because it’s always possible to connect corruption and education with violence through a series of argumentative steps. This, in a way, is of little use and somewhat trivial. What is crucial is to establish which of these two variables has more power to explain, for example, the number of homicides. Only when the importance between the two is determined does it make sense to structure a narrative to interpret the results.

The Economists are Here!

In the 1990s, Colombian economists finally began to study the phenomenon of violence. And I say “finally” because, in other countries, they had already started to delve into a field traditionally reserved for sociologists, anthropologists, and psychologists. It was Gary Becker, a Nobel Prize winner in Economics, who in 1968 proposed looking at the issue of violence from a decision analysis perspective. From there, economists discovered an intellectual territory, untouched and fertile, perfect for taming with their econometric models and linear regressions. (The sociologists, anthropologists, and psychologists did not find this intrusion into their field amusing at all. I don’t blame them. It’s hard to love economists.)

The new Colombian “violentologos” favored empirical studies based on hard data, and one of their main challenges was dealing with the scarcity of reliable data. Often the problem was the underreporting of crimes, which can lead to mistaken conclusions. For example, in Pablo Escobar’s Medellín of the 90s, the number of homicides shot up to nearly 6,000 a year, but paradoxically the number of reports of other crimes fell. Some academics interpreted that the locals had discovered a pleasure in killing, now doing it for any reason (perhaps due to the violent culture I discussed in a previous entry?), instead of realizing that in a city at war, people were so intimidated they preferred not to report common crimes. What chance was there for the police to catch a TV thief when they were also dealing with Escobar, who was offering a million pesos for each police officer killed?

Another problem with working with violence data is that it incorporates legal definitions that are arbitrary and subject to change over time. The Penal Code changed the definition of personal injury in 1980 and 1987, making someone who only studied the numbers reported by the Police see an artificial decrease in this crime.

The number of homicides is possibly the indicator with the least noise, while also being a good proxy for violence in general. It’s the indicator I’ve used in my own analyses, where I’ve studied national homicide trends over time and their breakdown between those associated with the conflict and those that are not. I am now working on doing the same but at a local level, focusing on certain key municipalities and departments.

Instead of studying evolution over time, many researchers have conducted what is called “cross-sectional analysis”, looking for the relationship between an interest variable – for example, homicides in each department – and another that may explain it – for example, the degree of coverage of primary education. These studies have all sorts of limitations (they often assume a linear, generally univariate dependency, and the directionality of the association cannot be inferred), however, they are a good starting point for delving into the possible causes of a phenomenon. Out of the dozens of potential explanatory variables for violence in Colombia, three are particularly interesting.

Against Conventional Wisdom

Poverty – measured, for example, by real departmental GDP per capita, GDP growth rate, the increase in the degree of unsatisfied basic needs, or the infant mortality rate – is not the cause of violence in Colombia. This is the conclusion of a long series of statistical studies which, together, have even observed the opposite: the expansion of regional wealth precedes the increase in criminality. Although counterintuitive, this empirical result is consistent with the war dynamics prevalent in Colombia, where the goal of a set of criminal agents is the extraction of wealth and the financing of their armies. Wherever there is cocaine, poppy, emeralds, gold, and oil, that’s where guerrillas, paramilitaries, and Bacrim settle and unleash horror.

Digesting this result is not easy, as it clashes with the sensibility many of us have regarding two tragedies that seem to intertwine. For example, a friend on Facebook wrote the following on his wall: “Peace is not just about not shooting bullets, but also about not having twelve or thirteen million poor peasants.” Who dares to refute it? Isn’t it terrible that there are millions of poor peasants? But the problem with this statement, which seems to come from the CEV script, is that it confuses two terrible phenomena that have been empirically shown to be independent (or whose dependency even goes in the opposite direction of what is expected), collapsing them under the same definition.

Peace is the absence of violence, but by violence, we must understand a very specific group of actions that we want to eliminate: homicides, displacements, disappearances, threats, personal injuries. Poverty is a tragedy in itself, one that we must find solutions for, but we gain nothing by lumping it together with violence if we know that both obey different dynamics.

Another variable is the one Mr. Petro invoked in the 2018 presidential campaign as a fundamental cause of violence: inequality. On this point, I have found material that points in all directions and will need a little more time before having total clarity on the matter. On one hand, I have found several studies conducted in Colombia that reported a null association between violence and inequality – measured, for example, by the Gini coefficient at the departmental level. But I have also read studies that have established that there is indeed statistical evidence between inequality and homicides in Colombia, although these have been pointed out for methodological errors that call their results into question. Internationally, studies have been conducted between certain forms of violence and the concentration of wealth indicating a positive and significant relationship between the two. However, when these models are applied to Colombia, the association fades and barely manages to explain a fraction of the violence.

Colombia is not one of the poorest countries in the world, but it is definitely one of the most unequal. This combination of high violence and high inequality is also shared by South Africa, Brazil, and to a lesser extent Mexico, three countries that, along with Colombia, are characterized by their extremely high homicide rates. (The fifth country with unusual violence, Venezuela, showed until 2011 data on inequality much lower than the others, but I need to find more recent figures.) That these countries are unequal and violent makes one think that there may be some relationship between these variables, but the association may not be as transparent as some want to believe. For example, Chile has inequality rates almost as high as Colombia’s, but a homicide rate that is even lower than the global average. In any case, there doesn’t seem to be a consensus to assert so emphatically that “violence is the product of social inequality and therefore social equity is the basis of peace.”

Inequality is one of the greatest social tensions we recognize in our times, and one that Colombia owes a debt to resolve. A discourse based on income redistribution, to become a slightly more equitable society, does not need to leverage other problems. But our politicians, like Mr. Petro, in their eagerness to articulate grand narratives that explain the country, forcefully include phenomena that have nothing to do with each other. Violence, inequality, and poverty are three different problems, not one.

The Weakness of Justice

The most interesting finding I have come across regarding the causes of violence was one established in the 90s by Armando Montenegro and Carlos Esteban Posada, and which has been verified and deepened by various researchers since then: the weakness of the judicial apparatus does explain violence in Colombia. The original study used the rate of people called to trial for homicide as a proxy for the degree of justice efficiency, but later analyses included other relevant characteristics of the justice system beyond impunity rates.

The structural issues plaguing Colombia’s justice system extend across its entire spectrum, touching every facet and institution involved. This encompasses not only civil, criminal, administrative, and labor courts, but also sweeps through the Comptroller’s Office, the Attorney General’s Office, and the Prosecutor’s Office. It affects Legal Medicine, INPEC, the Police, and the Armed Forces, as well as the notary and registration system, the licensing authorities, and even those tasked with controlling smuggling. Urban and transit authorities are not spared either. This widespread sclerosis has thwarted the administration and dispensation of justice, hindering the enforcement of rights for individuals and communities alike. From the moment a victim faces hurdles at the police station to report a crime, to the overt failures of the penitentiary service that sees prisoners escaping during a dental visit, and up to the systemic inefficiencies that let major criminals slip through the cracks, the entire judicial apparatus is mired in ineffectiveness.

The country’s economic growth did not come with a similar evolution of judicial structures that would allow for a sustained and orderly expansion, and what it generated was a situation in which the strongest had the prerogative to plunder wealth or administer justice themselves. And in all this, drug trafficking entered the country without finding a dike to contain it. Almost forty years later, Colombian justice still has not recovered from the challenge posed by the cartels, and with its ponderous pace, it remains ten steps behind the sophisticated criminal enterprises, still waiting for an opponent that can match their level.

It doesn’t take much science to recognize that justice in Colombia has serious problems, however, it is relevant to discover the role it plays in the high violence we suffer. The conclusions of these studies suggest where we should look for advances to contain the scourge of violence, and this is no small thing, especially on a matter where passions and sophisms confuse everyone.

We cannot wait for Colombia to emerge from poverty, or solve the problem of inequality to expect violence to decrease. Nor can we sit idly by, lamenting a supposed culture that tolerates homicides. The task of reforming justice to address violence is a Herculean task, but in the search for peace that is so often talked about, this is one that at least promises to make decisive advances in the right direction.

Leave a comment