This briefing discusses the impact of immigration on crime, both in the UK and in an international context.
- The labour market opportunities of different migrant groups are likely to play an important role in determining the impact of migration on crime rates.
- A continuous reduction in overall property crimes in England and Wales since 2002 corresponded with a rising foreign-born population, but there is no evidence to suggest that rising migration caused this decline in crime rates.
- The foreign-born share of the population is unrelated to violent crime according to the most recent research findings.
- In recent economic analysis, a 1% increase in the asylum seeker share of the local population is associated with a 1.1% rise in property crime.
- In recent economic analysis, a 1% increase in the share of the population that was born in the A8 countries leads to 0.4% fall in property crime.
- Foreign-nationals are less likely to report being a victim of crime than British nationals.
- The labour market opportunities of different migrant groups are likely to play an important role in determining the impact of migration on crime rates.
Understanding the Evidence
First, it is important to note that for the purposes of this briefing, migrants are usually defined as anyone born outside the UK. Other definitions are possible and in use in various official statistics and policy debates (see the Migration Observatory briefing “Who Counts as a Migrant? Definitions and their Consequences”). In the section on victims of crimes, however, it is foreign nationals who are identified as migrants, since data on crime victimisation tracks nationality rather than birthplace.
Just as important, this briefing addresses the issue of crime in general, including property crime and violent crime. It does not address immigration offences in particular, which are violations of immigration law. This topic is covered elsewhere (see the Migration Observatory briefing on “Immigration Offences: Trends in Legislation and Criminal and Civil Enforcement”).
The research in this briefing draws on data from the Home Office (Crime Statistics, Asylum Statistics, Prison Statistics, UKBA A8 Accession Reports) and from individual police force data on arrests. However, one potential problem in exploring the link between immigration and crime is that the data do not capture all crimes. Data is collected on individuals who are arrested and/or convicted, but will not include all crimes committed. Therefore, just comparing foreign-born and UK-born populations can be misleading. A series of factors might lead to foreign-born individuals having higher or lower arrest/conviction rates than the UK-born even if they commit crimes at the same rates. For instance, migrants may commit crimes that are easier (or harder) to detect or police could allocate more (or less) resources to catching migrant offenders or courts could be more (or less) likely to convict migrants.
Because of this limitation in the data, more complicated approaches are needed to try to determine if migration causes either higher or lower crime rates. One approach is to use reported crime data (which does not identify the perpetrator) and examine the differences in immigration and crime in local areas within a country. To see how this works, consider two local areas: Hackney and Gloucester. The foreign-born share of the population is higher in Hackney than in Gloucester. If crime were also higher in Hackney than in Gloucester this could be an indication that migrants affected crime rates. However, there could also be other reasons why Hackney has a higher crime rate than Gloucester. For example, high-density areas tend to have higher crime rates, as do areas with more social deprivation. A better approach is to examine whether the change in crime in an area is related to the change in the foreign-born population share. This approach controls for any time-invariant characteristics of the area that are potentially correlated with crime rates.
One remaining problem is that such an estimate might only capture a correlation rather than a true causal effect. For example, suppose that crime is increasing in an area and the UK-born are leaving that area. This opens up housing for incoming migrants. Thus the foreign-born share of the population rises at the same time at which crime is also rising, but migration is not actually causing crime to rise in this example.
One possible solution to this problem is to identify exogenous changes in the foreign-born population in an area—changes that are unrelated to other underlying characteristics of the area—and see if these are correlated with changes in crime rates. In other words, this approach looks for rises in the foreign-born share of the population in an area that could not be related to current conditions in the area, and then tests whether crime rises or falls. Such estimates do suggest a causal relationship rather than a mere correlation. These are, therefore, the central focus of this briefing.
Economic analysis of crime postulates a cost-benefit calculation underlying decisions to commit crimes
The traditional economic model of crime developed by Becker (1968) has individuals rationally choosing between criminal and legal activity by comparing the expected net benefit from each activity. In addition to the probability of being caught and the consequent punishment, the key driver in these models is the difference between potential legal earnings in the labour market and the returns to criminal activity. All else being equal, individuals with poor labour market opportunities are more likely to be involved in criminal activity. For instance, young males who left school with no qualifications are likely to have different crime propensities than experienced professionals with graduate degrees. There is a substantial body of evidence supporting this key contention (e.g. Machin and Meghir 2004).
In order to apply this theory to immigration and crime it is important to recognize that different migrant groups are likely to have differences in all of the components of the cost-benefit analysis that they undertake. To the extent that the UK-born and foreign-born differ in these parameters, the model would predict different criminal participation rates. For example, suppose that foreign-born individuals thought that the probability of being caught committing a crime was much higher than the UK-born thought, or that the consequences—possibly including deportation for foreign nationals—would be worse. All else being equal, this would deter criminal behaviour among the foreign-born relative to the UK-born. Or suppose migrants had much worse legal labour market opportunities than the UK-born – whether because their unemployment rates were higher or their wages were lower. This would increase the likelihood of criminal behaviour of migrants relative to the UK-born.
Property crime declined as the foreign-born population rose in England and Wales, but evidence does not suggest migration caused this decline
Figure 1 shows the number of reported property crimes—defined as burglary, robbery, theft and handling, fraud and forgery and drug crimes—in England and Wales since 1983 together with the foreign-born share of the population. There has been a continuous reduction in property crimes in England and Wales since 2002. This period has also witnessed a rise in the foreign-born share of the population. Indeed, there is a negative correlation between the level of property crimes and the foreign-born share of the population for the whole period presented in Figure 1. However, correlation is not the same as causation, and more detailed studies across a range of countries do not find evidence that the rise in migration caused the crime rate to drop.
The share of asylum seekers in the local population is related to a 1.1% rise in property crime but no change in violent crime. A rise in A8 migrants as a share of the population is associated with a 0.4% fall in property crime and has no relationship to violent crime
Bell et al. (2013) examine local crime patterns in England and Wales from 2002 to 2009 in order to determine whether there is any causal effect of an increase in the foreign-born population on crime. They focus on two large groups of migrants that arrived in the UK over this period. First, asylum seekers arising initially from the dislocations in former Yugoslavia and subsequently from war-torn societies such as Iraq and Afghanistan. Second, the large migrant flows coming from the A8 countries, particularly Poland, since 2004. The research showed that it is possible to derive causal estimates for both migrant groups and found that the share of asylum seekers in the local population was related to a rise in property crime, while a rise in A8 migrants was associated with a fall in property crime. Neither group was associated with statistically significant changes in violent crime. Estimates suggest that a one percentage point increase in the asylum seeker share of the local population is associated with a 1.1% rise in property crime. Since asylum seekers accounted for only around 0.1% of the population, the macro effects were small. A one percentage point increase in the share of the population that was born in the A8 countries leads to 0.4% fall in property crime.
Bell et al. (2013) suggest that the estimated effects for asylum seekers and A8 migrants may be the result of differences in the labour market opportunities of the two groups. The A8 migrants who arrived in the UK came almost entirely for work reasons and have higher employment rates than the UK-born. The motivation of asylum seekers was different, and they are not allowed to work in the UK upon arrival and also have reduced access to welfare benefits. Given the lengthy process involved in deciding asylum applications, this restriction is likely to have increased the relative returns to crime.
The international evidence highlights labour market opportunities of migrant groups as a determinant of their impact on crime rates
International evidence from studies using similar methodologies to Bell et al. (2013) include Bianchi at al. (2012) using Italian provincial data and Spenkuch (2011) across US counties. The Italian evidence finds zero overall impact of migrants on either violent or property crime. The US study finds no evidence of a causal link between migrants and violent crime, but does suggest a significant positive link with property crime. The estimates imply that a 10% increase in the share of migrants in an area would raise the property crime rate by 1.2%. Spenkuch (2011) breaks the migrant stock into Mexican and non-Mexican migrants. He argues that Mexican migrants in the US have much poorer labour market outcomes than other groups. His results suggest that the effect of migrants on property crime in the USA is driven by Mexican migrants. This highlights the importance of focusing on the labour market opportunities of different migrant groups.
Another way of considering the relationship between migration and crime is to use data on the victims of crimes. The British Crime Survey (BCS) provides the official measure of crime victimisation in England and Wales. This differs from reported crime data since a large number of crimes are not reported to the police. People may choose not to report crime out of fear (of reprisal or because they are undocumented) or merely because the crime is minor and not considered worth reporting. Thus, for example, in 2009 there were 1.2 million property crimes reported to the police in England and Wales, while the BCS estimates that 2.89 million such crimes actually occurred (Marie 2010). However, the long-run trends in the two measures are broadly similar.
The BCS survey asks respondents for their nationality, which enables the examination of crime victimisation rates by nationality. Controlling for an extensive set of personal and locality characteristics, Bell et al. (2013) show that foreign-nationals are actually slightly less likely to report being a victim of crime than British nationals. This might reflect real differences in victimisation rates, or different tendencies to report crimes. The authors suggest one possible explanation for this finding is that migrants have moved into neighbourhood clusters that provide a natural protection against crime, particularly if migrant-on-migrant crime is less socially acceptable (see also Bell and Machin 2013).
Evidence gaps and limitations
There are two key limitations in the existing evidence. First, it is limited in breadth so it will be important to see whether the patterns discussed in this briefing are consistent across a range of countries and migrant groups. There are studies in the criminology literature using local data that suggest that associations between immigration and crime in reality reflect relative deprivation and unemployment rather than direct immigration effects (see for example Bircan and Hooghe 2011). Such evidence is of course consistent with the economic approach discussed above.
Second, a key remaining puzzle is the remarkably high rates of imprisonment of migrants in many developed countries. Figure 2 shows estimates from the OECD of the share of foreign-born individuals in the population and in prison for various countries. If imprisonment rates were identical between migrants and non-migrants, the bars would be the same height within a country. With the notable exception of the United States, countries imprison far more foreign-born individuals than would be expected given their share of the population. Some of this difference can be explained by types of crime that foreign-born individuals commit – in many countries, migrants commit crimes that are associated with higher imprisonment rates. And some can be explained by foreign-born individuals who are arrested while transiting or visiting the country and therefore not counted in the population. But the gaps are too large for these explanations to suffice.
Note: Share of prisoners shows the percentage of prisoners who are foreign-born, while share of population shows the percentage of the general population who are foreign-born.
- Becker, G. “Crime and Punishment: An Economic Approach.” Journal of Political Economy 76 (1968): 175-209.
- Bell, B., F. Fasani and S. Machin. “Crime and Immigration: Evidence from Large Immigrant Waves.” Review of Economics and Statistics, forthcoming 2013.
- Bell, B. and S. Machin, “Immigrant Enclaves and Crime.” Journal of Regional Science 53 (2013): 118-41.
- Bianchi, M., P. Buonanno and P. Pinotti. “Do Immigrants Cause Crime?” Journal of the European Economic Association 10 (2012): 1318-47.
- Bircan, T., and M. Hooghe. “Immigration, Diversity and Crime: An Analysis of Belgian National Crime Statistics, 2001-6.” European Journal of Criminology 8 (2011): 198-212.
- Machin, S. and C. Meghir. “Crime and economic incentives.” Journal of Human Resources 39 (2004): 958-79.
- Marie, O. “Reducing Crime: More Police, More Prisons or More Pay.” Centre for Economic Performance Election Analysis, London School of Economics, 2010.
- Spenkuch, J. “Understanding the impact of immigration on crime.” mimeo, University of Chicago, 2011.
- Migration Observatory briefing – Who Counts as a Migrant? Definitions and their Consequences
- Migration Observatory briefing – Immigration Offences: Trends in Legislation and Criminal and Civil Enforcement