Learn Some Economics: How an Earthquake Kept Teens from Dropping Out of SchoolPosted: April 20, 2011
Across Canada, policies have been put into place that force teenagers to stay in school until a certain age (between 16 and 18, depending on the province). But what if I told you it could be easier than that? Easier in the sense that we may not have to force them. Or not all of them. I’m referring to peer effects. If policymakers can convince a segment of the teenage population to stay in school, in some circumstances that segment can positively affect others to finish high school as well. Let’s look at a case from Italy in the 1980s. Yes, I said 1980. Data take time to be collected, analysed, written about and published. Here we go.
Economists Piero Cipollone and Alfonso Rosolia published a paper in 2007 in the American Economic Review entitled Social Interactions in High School: Lessons from an Earthquake, in which they measured the effect of male high school graduation rates on the probability that females in the same cohort graduate. The authors provide evidence that after a large earthquake in southern Italy, the high school achievement of male teens positively affected the high school achievement of female teens in 1980.
Here’s the background: in November 1980, a 6.89 magnitude earthquake struck southern Italy, killing 3,000 and injuring nearly 10,000. 300,000 people were rendered homeless. As a result, a law was passed in the affected regions allowing males over the age of 14 (born before 1966) to be exempt from compulsory military service in an effort to help rebuild towns. Military service was known to interfere with high school graduation as some teenage boys were drafted in their last year of high school. Having entered the military without a high school diploma, these young men were unlikely to return to school 12 months later in order to obtain them. Effects of the absence of compulsory military service would be apparent in high school graduation rates of males born between 1963 and 1965 (aged 15 to 17 at the time of the earthquake). The authors therefore use the earthquake as a natural experiment.
The authors chose the towns that bordered the earthquake area as their treatment group. Damage to these areas was minimal yet teenage boys were still granted the military exemption, making these towns perfect candidates for measuring the exemption’s effect on graduation rates. The treatment group consisted of 57 towns made up of approximately 300,000 people. The authors used the towns adjacent to the treatment towns but outside of the earthquake zone as the control group. The control group consisted of 60 towns made up of 600,000 people. Treatment towns were less densely populated, had lower birth rates, higher death rates and were smaller on average than control group towns in the sample, although these small dissimilarities were controlled for by the authors in their economic model.
The Italian education system was structured in a way that students who start high school in the same cohort, will move through the grades together up to senior year. It is important to highlight that high school graduation rates were similar in the treatment group towns and control group towns for both males and females before the military exemption was introduced. To identify the effect of the military service exemption, the authors use data from the 1991 census and compare schooling attainment of cohorts in treatment group towns (qualified for exemption) to that of cohorts in control group towns (did not qualify for the exemption). Given these groups’ high school graduation rates were similar before the exemption, any divergence after the exemption is meaningful.
And in fact, there was a divergence. The authors find that the military exemption increased the likelihood that teenage boys in the treatment group graduated from high school by more than 2 percentage points a consequence of which teenage girls in the same cohort and born in the same towns were also more likely to graduate from high school. Specifically, a 1-percent-point increase in boys’ graduation rates increased girls’ rates by approximately 0.7 percentage points. Since females were never required to serve in the military, the authors can safely assume, having controlled for a number of variables, that females’ increased graduation rate is a direct result of peer group effects. Or, more commonly known as the power of a high school crush. I don’t think I’m wrong to assume that this peer group effect worked both ways: as more teen boys stayed in school, this caused teenage girls to stay in school, which in turn resulted in another set of boys staying in school as well (not because of the military exemption but because of the cute girls). The authors estimate that 180 teen boys and 150 teen girls in treatment towns, who otherwise would have dropped out of school, completed high school as a direct result of the military service exemption. The authors also point out that compulsory military service may affect the future economic outcomes of teenage boys and girls given the drop-out rate of those drafted.
I admit, these results are not entirely surprising. But it’s cool that what we would expect was proven. By respected economists. And published in the top economic journal in the world. The moral of this paper (aside from questioning the existence of compulsory military service of which I have virtually no knowledge)? If you want a teenage girl (or boy) to behave a certain way, convince other teenage boys (girls) to behave in that desired way too. Obvi. Policymakers worldwide, take note.
Source: Cipollone and Rosolia, Social Interactions in High School: Lessons form an Earthquake, The American Economic Review, June 2007, Volume 97, Number 3.