One of the concerns of the economic theory of crime has been the effect of the justice system on criminal activities. Theorists have proposed that the probability of being caught and sanctioned is a factor affecting the decisions made by criminals [Erlich 1996]. It is assumed that these are variables over which the government has control. Criminologists are less unanimous as to the effectiveness of the justice system, but their theories also assume a certain degree of system autonomy. In each case, the effect that the existence of armed groups (including private armies and protection services, mafias, insurgent groups, and other forms of organized crime) may have on the performance of the judicial system has been, to date, unexplored terrain. This paper will argue, with reference to the Colombian ease, that violence, and in particular that exercised by such groups, can become an obstacle to the adequate administration of criminal justice.
Available evidence is presented on the effect of violence and the threats posed by armed groups on the different stages of criminal proceedings. Using municipal level information gathered in Colombia, this paper attempts to trace the impact of these groups, suggesting that the effect begins with changes in the availability and quality of information on homicides.
Colombian Criminal Justice and the Effects of Violence
In the case of Colombia, the pressure that organized violence has exerted on the judicial system over the last two decades can be corroborated simply by reading the national press. It is also interesting to note the close association between attacks and threats from armed groups and modifications in the Colombian penal code over the last decade [Saiz 1997].
At the national level, statistics show a negative relationship between violence, using as an indicator the homicide rate, the presence of armed groups, and various performance indicators for the criminal justice system. In the last two decades, the Colombian homicide rate has more than quadrupled. In a parallel fashion, the influence of the principle armed organizations - the guerrilla, the drag mafia (or narco-traffickers), and paramilitary groups - has increased [Thoumi 1994]. During the same period, the capacity of the justice system to investigate homicides has been considerably reduced. The proportion of homicide cases that reach the courts, which in the 1960s was above 35 percent, today is less than 6 percent. In 1975, for every 100 homicides, more than 60 suspects were captured; in 1994, this figure had been reduced to 20. Conviction rates, which in the 1960s reached 11 percent of the total number of homicides committed, have dropped to barely 4 percent today [Ruhio 1996a].
These relationships can be explained in two ways. The traditional reading would be that the poor performance of the justice system has provided incentives for violent behavior in Colombia. It is also possible to argue that one of the factors that contributed to the paralysis of the criminal justice system in Colombia was precisely this violence and in particular that exercised by private protection services and extra-judicial prosecution.
A peculiarity of the Colombian system, which has been suggested as an explanation of its present-day incapacity to clear homicide cases, is its tendency to concentrate on infractions that are relatively innocuous and easily resolved to the detriment of those that are more serious and more difficult to investigate or to prosecute. The investigation of criminal incidents in Colombia, in practice, is limited to those with a suspect already identified, that is to those offenses that are all but resolved from the moment the victim first registered the complaint [Rubio 1996a].
On the other side, the data from surveys of victimization show how the attitudes and responses of citizens are contaminated by both the deficiencies of the justice system and an environment of threat and intimidation. Colombian society is characterized not only by high levels of violence, but also by the fact that citizens do not turn to the authorities to resolve criminal incidents. Even for a matter as serious as murder, more than half of the households victimized stated that they had "not done anything," and only 38 percent reported that they had made a formal complaint [Rubio 1996b].
Among the reasons offered for not filing complaints, two are worth particular emphasis. The first, cross-cutting all three surveys of victimization and quite peculiar in itself, is that of "lack of proof." This is a symptom of the form in which the criminal justice system in Colombia has increasingly delegated to private citizens the responsibility for investigating crime. The second is that of "fear of reprisals," which in the last decade has doubled its weight as a motivation for not reporting crime.
An analysis of the municipal data for 1995 shows that the presence of any armed agent in the towns negatively affects the quality of the information on homicides. A basic indicator of the quality of the statistics on violent deaths can be constructed based on the observed differences among different sources. For an important fraction of Colombian municipalities - more than 25 percent in fact - a defect in judicial data is readily apparent: the homicides registered by coroners, or by the National Police, are greater in number than those recorded in the judicial system. The probability of this underreporting increases significantly with the presence of guerrilla, narco-traffickers, and/or paramilitary groups in the vicinity.
In those municipalities where this phenomenon of underreporting has occurred - in general the more violent locations - it can be observed that denunciations per inhabitant in each category of offense are on average lower than the rates found in municipalities where the statistics are consistent. The observed association among underreporting of violent deaths, the presence of private armed groups, and low levels of denunciation can be explained in different ways, all of which reflect deficiencies in the operation of the criminal justice system. These explanations are consistent with a scenario in which armed agents and mafias sell private protection services or private justice services.
The phenomenon of underreported violence not only affects the reported levels of criminality, but also distorts the perception held of the effect of armed groups on criminality. The combination of the two effects described above means that in the typical Colombian municipality any armed presence reduces the number of formal complaints entered by somewhere between 15 percent and 25 percent.
The influence of private justice on judicial statistics is not limited to a negative impact on the formal complaint process. In these data, an effect is also visible on the number of investigations begun and on the priorities of the criminal justice system. This effect can be estimated by examining the distribution of investigations by offense. In the less violent municipalities, or those with no armed presence, the proportion of total offenses represented by attempted murder is higher.
Consistent with a scenario of mafias that impede the investigation of homicides, the data for municipalities show a negative association between violence and the interest taken by the system to clear cases of attempted murder. The presence of more than one such group in a municipality also has a devastating effect on the system's priorities to the detriment of the prosecution of serious offenses. In order to have an idea of the magnitude of this impact, it is enough to note that the presence of two groups in a town has an effect similar to taking a step from a peaceful society to a state of civil war.
To summarize, the analysis of data on judicial performance, violent deaths, and an armed presence in Colombian towns presents an interesting case. The initial effect of organized violence on the performance of the criminal justice system in Colombia is shown in the discrepancy, in certain violent towns, in the counting of homicides on the part of prosecutors and judges. The available information is quite revealing as to the source of the problems in counting violent deaths in the country: the criminal justice system itself. These deaths begin to disappear from the statistics in the figures provided by the courts. It is difficult to imagine that, if accurate information on the simple number of deaths is impossible to obtain, there could be any clarity on the circumstances in which these deaths occur or on the perpetrators of these crimes.
This first imbalance between the acts registered by the justice system and the actual occurrence of violence surely affects Colombians' perceptions of justice and their willingness to use the system to report all types of offenses. Their scepticism appears logical in light of the existence of official homicide statistics, which are inferior to its true incidence. The low rate of formal complaint that is observed in the case of towns with an armed presence can, in principle, be consistent with both a reduction or an increase in delinquency. The data offer weak support for the hypothesis of an increase in criminality. The responses of households on the factors that are believed to affect delinquency in their regions tend to support the position that armed groups contribute to insecurity. Available testimony, however, also reveals that in some localities armed groups may be entering the community to bring order, thus reducing the crime rate. The presence of more than one armed force in an area has a devastating effect on justice, which seems to turn into a true "wartime justice" under which the larger the number of violent deaths the less the interest in investigating them, much less solving them. In synthesis, the data show that the influence of an armed presence on the Colombian criminal justice system appears to begin with disinformation about violence. From the moment in which the system, in its statistics and assuredly in its performance, begins to depart from reality, the conditions are set for this vicious circle of disinformation and recourse to private protection services to develop, in which, theories tell us, the mafias grow and flourish [Gambetta 1993].
Lessons of the Colombian Case
Perhaps the main conclusion to be drawn from the Colombian experience of violence is that the increasing severity of the problem poses ever greater obstacles both for its adequate diagnosis and for the design of efficacious policies of control.
The unleashing of violence has various consequences. In the first place, difficulties arise at the most basic level of measurement. In the second place, the available criminological theories, which have been developed and debated in the context of peaceful societies, lose their relevance. Under conditions of severe violence, the design and execution of policies can be affected not only by the uncertainties of diagnosis and the difficulty of evaluating the different alternatives, but also by the simple fact that there is diminishing clarity as to who is actually making the public choices.
This extreme is normally preceded by a marked disinformation surrounding the actors and circumstances of violence and by a reticence to look beyond traditional explanations. These effects are mutually reinforcing: it is precisely the violence that conforms to prevailing theory that is most easily explained in the first place. In this situation, myths abound, and in the area of public policy confusion reigns. The analytic bias toward random violence directs efforts in the wrong direction. The actions against "other violence" become segmented and lose coherence. This is exactly the environment that favors the growth and consolidation of mafias in a society. In the middle of disinformation, marginalized by theories that do not even mention them and aided by ideologies and acting with impunity given the reality of a congested and trivialized justice system, different armed groups appear and gain strength. These groups are leading society along an institutional path that is increasingly permeated by violence and increasingly less capable of controlling them.
It is only in situations where the violent crime rate is low that the dominant criminological theories and the public policies that they have inspired have any relevance. Imagining a continuum of violent deaths that goes from the accidental and casual - death in brawls, due to alcohol and intolerance - passing through armed robberies and ending with private vengeance or mass slaughters ordered by powerful armed forces, there are two conclusions that can be derived from the recent Colombian experience. The first is that as the homicide rate increases, violence departs from the merely fortuitous or random. All the evidence for Colombia corroborates this impression. The little theory available on extremely violent societies also leads in the same direction: there is no such thing as widespread violence perpetrated accidentally that does not lead to the rise of armed groups with an enormous power based on force that later adjust the rules of the game to consolidate their power.
The second major conclusion is that as violence increases and departs from the casual or accidental, it makes it more difficult but more necessary to have a well-performing criminal justice system. The longitudinal and cross-sectional evidence for Colombia shows negative relationships between the performance of the justice system and violence, and it would be inadequate to interpret this as a casual or unidirectional effect. The theories on organized crime predict feedback mechanisms between Mafia influence and inoperability of the justice system. In Colombia, there are numerous and quite various indicators that lead in this direction.
Erlich, Isaac. "Crime, Punishment, and the Market for Offenses." Journal of Economic Perspectives 10, no. 1 (1996).
Fiorentini, Gianluca, and Sam Peltzman, eds. The Economics of Organised Crime. Cambridge: CPER - Cambridge University Press, 1995.
Gambetta, Diego. The Sicilian Mafia: The Business of Private Protection. Cambridge, Mass.: Harvard University Press, 1993.
Rubio, Mauricio. "Crimen sin Sumario - Analisis Economico de la Justicia Penal Colombiana." Documento CEDE no. 96-04. Bogota: Universidad de los Andes, 1996a.
-----. "Inseguridad y Conflicto en las ciudades colombianas." Documento CEDE no. 96-09. Bogota: Universidad de los Andes, 1996b.
Saiz, Ana Maria. "La Presion de los grupos economicos en la legislacion penal colombiana." Trabajo de Grado no publicado. Bogota: Universidad de los Andes, 1997.
Thoumi, Francisco. Economia Politica y Narcotrafico. Bogota: Tercer Mundo, 1994.
Mauricio Rubio is Researcher of Paz Publica Programa de Estudios sobre Seguridad, Justica y Violencia, Universidad de los Andes, Bogota, Colombia. This paper was presented at the annual meeting of the Association for Evolutionary Economics, Chicago, Illinois, January 3-5, 1998.…