"We need to find smart ways to causally assert that education reduces crime," says Professor Olivier Marie

Professor Olivier Marie is a crime-economist. He specialises in the (none) making of a criminal. He was the honorary promotor and presented the honorary degree on behalf of the Erasmus School of Economics to Esther Duflo, professor of Economics at the Massachusetts Institute of Technology. While Duflo uses randomised control trials to investigate how to understand the causes of poverty, Marie focuses mostly on natural occurring experiments, like policy changes, to investigate why people commit crimes.

What is your research about?

“I’m trying to understand why people commit crime. Are they poor? Is it because they have a lot of time on their hands? Is it because negative parental influence? We use econometric tools to try to see if we can find real causal evidence. There is, for example, a strong correlation between low education and crime. People who have less access to education, commit more crime. But it is hard to say that it is actually the education itself, or other factors that influence both education and crime, that makes the difference. We need to find ways of causally assert that education reduces crime.”

Is this hard to prove?

“Well, you have to be clever. For example: I worked on the impact of having access to cannabis in the Netherlands. Does having access to cannabis – or other drugs – make people consume it more often? This is not so obvious. While I was working in Maastricht. In 2010 and 2011 some nationalities were not allowed anymore to enter coffeeshops  as the city was trying to fight drug tourism. At the university we had students from different nationalities. Some could still enter coffeeshops, others not. Instead of asking them in a survey – they might lie – we compared the exam grades of the students who did and those who did not have access. The effect was clear: students who did not have access to coffeeshops, got much higher grades. As in all my research, what is interesting for me is not: what will make the perfect (or terrible) individual from the perfect (or terrible) family commit a crime or not? I’m interested in the people on the margin of deciding to commit a crime or not, and I’m interested in their incentives.”

"My job is mainly to see what causes can put people into criminal behaviour. I work on small examples, and hopefully one day we will have the bigger picture, about behaviour of people in general and what makes them commit crimes”

You got the Veni and the Vidi grant. What is your most recent research about?

“My Veni grant was about income shocks. If people lose their job suddenly, they are more tempt to commit crimes. Is it because they have less money? Or because of the shock or depression following a resignation? Or because they have a lot of time? We use an administrative rule that give some people who lose their jobs in the Netherlands extra money and others not to focus on the impact of income relative to other possible factors that may be driving criminal behaviour. The conclusion seems to be – still working on that one – that it is not completely the income.
The way this can help preventing crimes, is trying to intervene for instance through policies: give some people who got fired some income, offer others psychological support. My job is mainly to see what causes can put people into criminal behaviour. I work on small examples, and hopefully one day we will have the bigger picture, about behaviour of people in general and what makes them commit crimes.”

And the Vidi grant?

“We are currently investigating the effects of the introduction of the birth control pill in 1970 in the Netherlands and how the decrease of unwanted children might have led to a decrease in crimes. After 1970 we see a drop of 45 percent of teenage births in the Netherlands. After abortion was legalised in the US, the next generation committed much less crime because there where less unwanted children. Big academic authors in the US state that fifty percent of the drop of crime in the 1990’s was because of parental selection (both abortion and pill). There is a strong link between fertility, education and crime.
We wanted to look at the Dutch situation, we looked at young women in small Dutch villages and the question whether they had access to the pill or not. We looked at the votes in the villages, how many people voted for the SGP or two other conservative parties. We used the latest ‘Volkstelling’ from 1970 to see whether the GPs and pharmacists in each village were themselves religiously conservative or not. What we found so far, for example, is that there seem to be a strong impact the pill had on shotgun-weddings. Shotgun weddings are weddings arranged to avoid embarrassment due to premarital sex leading to an unintended pregnancy. Until 1970, 18 percent of marriages were shotgun weddings. After the introduction of the pill, this dropped to 6 percent."

Why is this still relevant?

“This is still a very relevant topic: as you know we still have conservative parties right now in this country who put anti-abortion flyers into advertisement brochures. The societal impact is evident, and important, but I also tend to defend my colleagues who do more theoretical work like the econometrists. What they do, might have a longer-term effect but maybe a huge societal impact. At Erasmus School of Economics we say that we share the impact: without the background work of theorists and econometricians - who develop the framework and methods use – no societally relevant research would ever be possible.”