Desarrollo y
Seftiembre de ""
in marginalized urban areas of developing countries. Typically, the
probability of being arrested and incarcerated for a murder is estimated
to be below 10 per cent in many Latin American cities24. Are these stylized
facts consistent with the previous general model? If this is not the ease,
how do they modify its predictions, in particular with respect to the
economic determinants of crime and violence?
If we consider the extreme case where there exists a market of a given size
for illicit activities and where those engaged in them have no big risk of
being arrested and prosecuted, then the issue becomes one of industrial
organization and occupational choice. The main difference with other
economic sectors and occupational choices is that there is likely to be no
market rule in the control or production of illegal activities so that
individuals operating in them rely on their pure capacity to physically
neutralize or eliminate potential competitors. At some stages of the
organizational development of this sector in a given local environment,
‘non-market’ competition is strong and is responsible for a high level of
violence among persons or gangs. At another stage or in a different
environment, the sector may be fully controlled by organized crime with,
paradoxically, some drop in the level of violence. The analysis of crime and
violence linked to illegal activities thus becomes that of the conditions
under which some type of organization of this particular sector of activity
will predominate over others24.
If a high degree of non-market competition in the sector of illegal activities
is responsible for the violence and criminality observed in some parts of
metropolitan areas in developing countries, are the causes and possible
remedies identified above still valid? As a matter of fact, this analysis of
the causes of violence does not deeply modify the nature of the initial
model of crime. As already noted above, it simply makes the premium, x,
of getting into illegal activity exogenous, rather than more or less loosely
related to the mean income in the whole urban population, and modifies
the nature of the risk, q, involved and penalty, K incurred. Risk is not any
23 A rate of H percent is reported for El SaKadorin Londono and ( Iucrrcro ( J99S. p 37). Thisfigurc
was lo¾er than 6 per cent in 1983 in CalL and probably of the same order <∣f maüniludc in other
metropolitan areas in Colombia. Moreover, it most Curtainlyhasworscncd since then, ‰e (∣ucrrc∙
ro ¢1997. ρ. 98).
ɔ4 On organized crime see Fiorentini and Pckzman ( 1995),
79