Spain's INE crime stats by origin spark controversy

Spain's INE breaks down crimes by origin, sparking controversy

A chart showing the 2023 crime rate per 100,000 people, differentiated between Spaniards, Americans, and Africans, triggered a cascade of reactions extending far beyond the data itself. The statistics, released from records of Spain's National Statistics Institute (INE), show a pronounced gap between groups. This is where the problem begins: what exactly does this number measure, who is included, and what is left out?

The data circulated with the label of "scandal" and an explicit request to share it with certain political parties. Within hours, the conversation split into two camps: those who read it as definitive proof of a security issue linked to immigration, and those who argue the figure explains nothing without breaking down age, income, and employment status.

What the crime rate by nationality really measures

The first hurdle is methodological. The statistics cross arrests or offenses with nationality, but not with age or income level. Analysts immediately pointed this out: elderly Spaniards in care homes rarely commit crimes, and this large demographic dilutes the native group's rate. Without controlling for age, any comparison between groups with different demographic pyramids is biased.

Some add a second filter: the category "Spaniards" includes naturalized citizens and those born in Spain to foreign parents, mixing cultures and origins under one label. The circulating proposal is to break it down further—North Africans, Sub-Saharan Africans, Asians, Pakistanis—and separate by type of crime. They argue that with such detail, differences would widen in specific categories.

The third front is international comparison. Studies on immigration and crime in the United States, cited with links to research bodies and work covering 150 years of data, point in the opposite direction: immigrants commit fewer crimes than the native population and help lower the overall rate. This contrast complicates rather than resolves the issue.

Precariousness as a variable and the German counterexample

The second line of friction is socioeconomic. It is argued that Spain's labor market precariousness—low wages against disproportionate housing costs—drives exclusion and, consequently, crime. The comparison with Spanish emigration to Germany in the 1960s and 70s is recurrent: migrants arrived with contracts, strong unions, and rent absorbing a small fraction of their salary.

The counterargument is direct: if precariousness were the main cause, all migrant groups facing similar conditions would offend at the same rate. This is not the case. Some groups arrive with initial capital, family networks, and economic plans, and their conflict with the law is of a different nature. The misery of post-war Spain, with almost no crime, is used as a counterexample that poverty and crime do not always go hand in hand.

The battle for the narrative: banning the data

In the background appears a political discussion: if the statistic is inconvenient, the solution might be to stop producing it. The French case is mentioned, where legislation limits counts by ethnic origin, and there is speculation that this could be the path to eliminating statistical evidence of the phenomenon. This hypothesis gained the most traction in the final part of the conversation.

Meanwhile, the base data remains the same: 77% of registered incivil in Spain are born in Spain, according to the first result when searching for crime statistics by nationality. The paradox is evident: the same source used to blame some serves to exonerate others, depending on how it is read.

The conditional probability problem

The statistical knot lies in distinguishing absolute frequency from conditional probability. That the majority of crimes are committed by the native population is compatible with a higher per-capita rate in a minority group. These are two readings of the same record, and neither is false individually. However, the misunderstanding fuels the debate.

To this is added an exposure bias: crimes committed by migrants are more likely to end in arrest and generate statistical records, while part of native incivil is resolved through other channels. The figure is not a perfect mirror of reality; it is an administrative byproduct.

How much of the gap is culture, how much is age, how much is income, and how much is recording bias? No one in the conversation managed to separate these four variables. And without that separation, the chart will continue to serve its purpose: confirming each person's pre-existing beliefs.