Society tends to perceive an increase in the immigrant population with an increase in crime. But, according to a new study, it is not possible to infer this cause-effect relationship in the case of Spain.
"Crime in Spain is low compared to the rest of Europe. Crime rates have increased slightly in recent years, unlike the immigrant population which has grown at a much greater pace. This suggests a positive yet low correlation between immigration and crime," as explained to SINC by César Alonso-Borrego, lecturer in economics at the Carlos III University in Madrid, Spain.
The researcher is a coauthor of the study published in the American Law and Economics Review Journal. It evaluates whether "this correlation consists of a causal relationship between immigration and crime." The conclusion is that it does not.
The researchers constructed an empirical model to measure the probability of committing crime depending on environmental and individual characteristics, such as educational attainment. They used data from the Spanish Home Office on the crimes committed each year for each 10,000 inhabitants in every Spanish province from 1999 to 2009. Information on the immigrant population was extracted from the Electoral Register and the Labour Force Survey. Furthermore, environmental characteristics were taken into account using measures like the GDP per capita and the unemployment rate in each province.
As the expert points out, "we studied the number of crimes per inhabitant for each place and each year. Amongst the relevant variables was the proportion of immigrants according to their origin and characteristics (age, gender, education and language)."
The use of information for each province causes a potential endogenous problem. In other words, this means the unseen differences between provinces that affect crime levels (like access to opportunities) and the proportion of immigrants.
The researcher stresses that "there tends to be more crime in places that offer more economic opportunities. These places are precisely where there are higher levels of immigrants. This suggests a positive correlation between immigration and crime, which could wrongly lead to attributing this to a causal link between both phenomena." However, thanks to the availability of longitudinal data (values for each variable over a period of various years), the researchers have been able to consistently calculate the veracity of this causal effect.
Estimates confirm the importance of language and education. "In particular, there is less crime amongst the Spanish-speaking immigrant population and, to a lesser extent, those immigrants from the European Union. Furthermore, the immigrant's educational attainment, which is relatively high in relation to native Spaniards, explains how the effect of immigration on delinquency is moderate," according to the experts.
Young men commit more crime
Likewise, as is the case in other countries, the immigration proportion of young men is associated with a higher crime rate since this population group is responsible for most of crimes committed. The expert emphasises that "the proportion of young men is higher amongst the immigrant population than the native population."
Unlike in the USA, this massive influx of immigrants is relatively recent throughout the European Union. This is particularly the case in Spain where the weight of the immigrant population has increased since the year 2000.
"Our results fall in line with well-known 'Latino Paradox' in the USA, where the immigrant Mexican population saw a decrease in crime. This is because the immigrants from Mexico consisted of a 'virtuous selection' of individuals whose tendency to commit crime was lower than that of the native population. Immigration is not a homogenous phenomenon but is rather a range of very different groups that each require different policies depending on the problems associated with their specific characteristics," conclude the researchers.
Explore further: Why people believe undocumented immigrants cause more crime
American Law and Economics Review 14 (1): 165-191, primavera 2012. doi: 10.1093/aler/ahr019