Desarrollo y sg∣gB∣MD.
Septiembre oe ∣ws S
crime. This should give some idea of the scope of policies aimed at
controlling and reducing the extent of inequality and poverty' in urban
areas. Although we can rely only on very rough estimates, we shall see
that it is surprisingly important.
Table 2 puts together some crude estimates of the various components of
the cost of crime in the L1S and in Latin America as a proportion of ÜDP.
They draw respectively on Freeman (1996) and Londono and Guerrero
(1998) -hereafter L(}-. As a first approximation, one may consider that
estimates for other countries or regions may be obtained by sealing these
estimates up or down depending on observed crime rates.
The first line of the table corresponds to the straight estimate of property
crime, that is, the total amount of robberies, thefts, burglaries, frauds,
embezzlements, etc. For the US, the figure is obtained from the National
Crime Victimization Survey's. For Latin America it is based on special
surveys taken in six countries18 as part of a research project organized by
IDB. Assuming that the average amount involved in property crimes is
proportional to income per capita, the figures appearing in this row of
Table 2 suggest that the rate of property erime in Latin America is three
times that in the US, which seems a reasonable order of magnitude -as a
matter of fact the same as that observed for homicides. Notice that this
row is entitled ‘transfers’ and is not included in the social cost of crime.
Such a view corresponds to the theoretical model seen above -see equation
(5)- where property crime indeed appears as a simple exchange of property
status, and therefore as a transfer of wealth or income from the victim to
the criminal. However, part of this wealth maybe destroyed in the transfer.
This part appears in the second line of Table 2 as a ‘monetary’ cost of
crime. But this item also includes the actual costs incurred by victims in
addition to what they lost, that is. medical expenses in case of violent
robbery, repair of property in case of a burglary, time spent dealing with
the police or justice personnel, etc. There is no direct estimation of that
cost in the Latin American ease14. The figure appearing in Table 2 is
obtained by assuming the same proportionality factor as in the US. The
-ls Brazil, Colombia, El Salvador, Mexico, ltru. Venezuela.
v> LG only report ¢).2 per ocni for medical expanses and a r∏uvh ħig)ιc∙τ uι∣!<ιuιιι f«»i ‘productivity
losses' which we do not take into account in Table 2.
91