Over the last days several people have notified us about emails from somebody posing as Statistics Norway. Usually the emails include the recipient winning a mobile phone. These emails are not sent from Statistics Norway and should be deleted.


Reports 2017/36

Crime among immigrants and children of immigrants in Norway

This publication is in Norwegian only.

Open and read the publication in PDF (2.1 MB)

This report compares the prevalence of registered offenders among both immigrants and Norwegian-born persons with two immigrant parents as compared to the remaining population. We explore subsample variations based on country of birth, the reason for immigration, crime type, and change over time. The analysis is presented in two chapters.

In chapter 3 we use data on all immigrants, Norwegian-born persons with two immigrant parents and people in the remaining population who were 15 years or older and permanent residents as of January 1, 2010, and explore the proportion in each group that was charged with at least one offense committed between 2010 and 2013. The results show that both immigrants and Norwegian-born persons with two immigrant parents are overrepresented as registered offenders, with the rate of overrepresentation being highest in the latter group. Among immigrants, the overrepresentation is most substantial among family immigrants and refugees, as well as for individuals from African countries. For Asian immigrants the picture is more complex. Overall, Asian immigrants are overrepresented. However, while immigrants from certain Asian countries are similarly overrepresented, other Asian countries are underrepresented. Individuals from Western Europe and North America, as well as education immigrants, are underrepresented as well. The pattern is, with some minor exceptions, relatively similar for Norwegian-born persons with two immigrant parents. The patterns of over- and underrepresentation also apply to most types of offenses, except for drug offenses where most immigrant groups are underrepresented. Overall the overrepresentation is substan¬tially reduced when we account for differences in age and gender, especially in the groups with the highest rates of overrepresentation – including Norwegian-born persons with immigrant parents. Place of residence and employment have limited explanatory power once the demographic differences are accounted for. For most immigrant groups a certain level overrepresentation persists also after socio-demographic characteristics are taken into account. It is, however, important to stress that the vast majority of individuals in all population groups were not registered as offenders during the period we consider.

In chapter 4 we use annual data on the population of Norway between 1992 and 2015, to explore the proportion of immigrants, Norwegian-born persons with two immigrant parents and people in the remaining population who were charged for at least one offence a given year. We again limit the analysis to permanent residents aged 15 years or older as of January 1 each year. These analyses show that immigrants and Norwegian-born with immigrant parents were overrepresented among charged offenders throughout the period we consider here, but that the level of overrepresentation has decreased over the last 10-15 years. The rates of overrepresentation in various groups reflect those described in Chapter 3. The decline is most substantial for the groups that were the most overrepresented in the beginning of the century – including immigrants from African countries, refugees and Norwegian-born with immigrant parents. The time trends vary somewhat depending on crime type.

There are two important limitations to these analyses. Firstly, they are based on registered offenses committed by formal residents of Norway. This means that offenses that are not reported and cleared-up by the police, as well as offenses committed by individuals not part of the formal population, are not included. Secondly, they are purely descriptive, in the sense that they describe associations and correlation in the data. This means that the analyses are well suited to describe patterns of over/underrepresentation among different groups, but not suited to say something about why or how these patterns emerge. The described associations may therefore never support a causal interpretation.

Read more about the publication