PDF Archive

cities, where the risk ofpeople committing crimes is far larger than in other parts
of the country.
That these hypotheses have no relevance is demonstrated empirically by this
study. Immigrant overrepresentation in crime is eaused neither by unfavourable
division regarding sex, age nor by residential area. N or is it eaused by any generally lower social economic status (calculated as per SEl code) in Sweden.
Continuing from this we can pick out two main hypotheses. One says there
are factors in the immigrant situation itselfwhich can generate crime; difficulties
in adjustment, financial problems, etc. The other says (the selection hypothesis)
that immigration is to a certain extent selective, i.e. that a proportionally larger
section of those who emigrate are criminals or are in the high risk zone for becoming criminals.
Both these hyperheses are supported by this study, although primarily concerning immigrants from the Nordie countries. Although the crime level in Finland, Norway and Denmark is probably generally lower than in Sweden - certainly not higher. The proportion of immigrant Norwegians and Danes registered for crimes is almost twice that of the equivalent proportion of Swedes. The
proportion of Fins registered for crimes is more than twice as high. The
overrepresentation of Scandinavians registered for crimes could confirm the hypothesis that the immigrant situation generates crime, even though in this case
this is not as probable. Adjustment problems for Scandinavian immigrants should
not be as large.
The weightiest argument against the hypothesis that the situation of immigrants generates crime is that there are large differences in crime participation
between immigrants from different countries oforigin. This primarily concerns
immigrants from South East Asia who, compared with most other immigrants
from far off countries, display three times less crime participation. It is improbable that the immigrant situation of people from South East Asia is sufficiently
different from the situation of people from other distant countries.
If there is agreater selection of the criminally inclined amongst persops ernigrating to other countries, then this should also apply to South East Asians. One
explanation, if this selection hypothesis was to ,be found valid, is that there is
much to indicate that criminality in South East Asia is comparatively low, and
because there are originally fewer criminally inclined in these countries, this characteristic would be reflected by emigrants. However, actual proofthat criminality
in South East Asia is low is not particularly strong.
A comparative study has been made arnongst 12 European countries and USA,
Canada, Australia, New Zealand and Japan based on interviews concerning exposure to crime. The study found that Japan seems to have the lowest crime rate per
capita of these countries (van Dijk, 1992).

120

Another fact in favour of the selection hypothesis over the immigrant situation hypothesis is the criminological information regarding criminal careers. Those
who commit crimes as adults have usually begun to commit crimes when very
young, generally between 10-14. It is becoming increasingly rare for people to
make their crime debuts at ages older than this. Debuts over the age of 20 are
very rare indeed. If the hypothesis claiming that the immigrant situation eauses
crime then immigrants would be behaving contrary to this pattern and making
their debuts at much higher ages, despite coming to astrange environment, which
could equallywell have a preventive effect as regards making a debut in crime,
The se1ection hypothesis therefore seems most likely to explain the
overrepresentation of immigrants in crime, although we should not reject the
immigrant situation hypothesis. Criminal behaviour is usually eaused by several
interacting factors. If, for example, it is probable that immigration tends to involve a larger proportion of persons with a higher crime inclination than the
norm, then factors prevalent within the immigrant situation will contribute to
the likelihood of these persons actually committing crimes.
There is insufficient material contained wirhin the framework of this statisrical survey to warrant a more indepth test of different hypotheses on the reasons
for the overrepresentation of immigrants in crime.

Criminality amongst the children of u'nITllgl·alTts.
percent of
During the 1985-1989 period children ofimmigrants ,",V>.lUU.HLL_U
the crime registered in Sweden, i.e. about the same proporttön
According to crime participation figures (the proportion re~;istete<f
children of immigrants were overrepresented in companson
overrepresentation of 1.5). This was partly due to
å.§:e>idi\iisJlört
proportion of young people). When we allow for sex,
overrepresentation drops to 1.4. Thus, compared with
overrepresentation in crime of the children
to
As regards overrepresentation of the children
Swedes and underrepresentation in comparison to lll1l1l1h'<LUL
the~e differences are due to dissimilar risks ofbeing detecteds
pinpoint the differences in risk of detection between
apart from certain special types of crime, so it is ditncult
when comparing the children of immigrants and Swedc~s.
should arise differences in the risk of detection when cotrip:g
the children of immigrants.
Because in the crime participation figures for the
residential area have been allowed for and the gerler;l1

rttrctnt to

121