Jérôme Valette: ‘There is no direct causal link between immigration and crime’

Jérôme Valette is an economist at CEPII (Centre d’Etudes Prospectives et d’Informations Internationales), France’s leading economic think-tank, founded in 1978. Valette is also a lecturer in economics at the Université Paris 1 Panthéon-Sorbonne and head of the Dynamics department at the Institut Français de Collaboration sur les Migrations (ICM). 

Together with Arnaud Philippe, an economist specialising in criminal justice, he is co-author of Immigration et délinquance : réalités et perceptions. Their article explores a question that regularly crops up in political debates and statistical analyses: is there a link between immigration and increased crime?

Voxeurop: Is there a link between immigration and crime?

Jérôme Valette: It all depends on the nature of the link. There is no direct causal link between immigration and crime. In other words, given equivalent demographic and socio-economic characteristics, immigrants are no more likely than native-born people to commit a crime.

Immigrants and natives almost always have different demographic and socio-economic characteristics. Men, young people and people in precarious situations are often over-represented among immigrants.

It is these specific factors that increase the odds of committing a crime. The same factors also increase those odds among native-born people.

There are two key points here. First, there is the issue of a causal link: does being an immigrant in itself, all other things being equal, increase the likelihood of committing a crime? The answer is no.

But, secondly, you have characteristics associated with certain waves of migration (such as poverty and economic insecurity) that may increase the likelihood of committing a crime. Yet these are characteristics that, when found in native-born people, also increase the likelihood of committing a crime.

To simplify things, we might say that economic precariousness increases the likelihood of committing a crime, but migration does not, is that it?

If you are an immigrant, on average you are no more likely to commit a crime. This does not hold for poor immigrants specifically, but that is because they are poor, not because they are immigrants.

In other words, to be an immigrant in itself does not increase the likelihood of committing a crime. If an immigrant is more exposed to poverty, it is poverty that is the cause, not the fact of being an immigrant.

Does the question of a link between immigration and crime not itself reveal a prejudice? Very often, as journalists or researchers, we find ourselves trying to do fact-checking on assertions that are essentially just false.

I think that all questions are legitimate from a research point of view, and that there is no reason to favour one question over another. This is especially true given that the link between immigration and insecurity is of great concern to native-born people. It’s legitimate for journalists and researchers to take an interest.

Our role, however, is to approach this issue by going beyond the simple interpretation of crime statistics.

It is essential to understand that when non-immigrants associate immigration with crime, this stems largely from their reading of the statistics. For example, when we see that foreigners make up 8% of the population in France, but 17% of those charged with offences, we are quick to conclude that there is a direct link. This reasoning, although intuitive, is unfortunately not sufficient.

In what way?

There are many reasons why immigrants are systemically over-represented in crime statistics. One example is discrimination.

If immigrants are more likely to face discrimination throughout the criminal justice system – whether in terms of the probability of being arrested or the probability of being found guilty – then they will automatically have a higher probability of being implicated and convicted than non-immigrants. This leads to an apparent over-representation in the data.

To overcome such bias, researchers are looking more closely at a key question: when immigrants settle in a region, is there an increase in crime? Studies show that, on average, no, there is no general increase in crime. The one exception to this is property-related offences, or theft, to put it more simply.

Why this exception? This type of crime is closely linked to economic insecurity and difficulties in accessing the labour market. Both these issues are particularly prevalent among people with irregular status.

You also mention the importance of media coverage. What is the role of the media?

There are two important media practices that can contribute to overestimating the perception of a link between immigration and crime.

The first is the choice made by journalists to report more often on crimes committed by foreigners than on those committed by native-born people. In other words, for the same crime, newspapers tend to focus more on cases involving foreigners. This mechanically creates an over-representation of the crime gap between immigrants and natives in the media.

A study carried out in Switzerland demonstrated this phenomenon at the time of the 2009 referendum on banning the construction of minarets. The researchers showed that municipalities where the press tended to over-report crimes involving foreigners recorded a higher vote against the construction of minarets.

The second mechanism lies in the choice made by journalists of whether or not to reveal the nationality or origin of the people implicated in crimes. An experiment in Germany by the Saxony-based newspaper Sächsische Zeitung highlighted this effect. In July 2016, this newspaper moved overnight to a policy of always mentioning the origin of every individual involved in a crime or misdemeanour, regardless of whether they were foreign or native-born (German, in this case).

This change exposed readers to more neutral media coverage. Articles would state, for example, that “a 25-year-old German has committed a crime” or that “a German is a suspect”. The new approach changed public perceptions in the newspaper’s region of coverage. Attitudes towards immigration became more positive, and concern about migration shifted to a more general concern about crime. Residents began to see crime as a problem involving the whole population, rather than just one linked to immigration.

🤝 This article is published within the Come Together collaborative project