Teenage tearaways earn less in later life than upright peers, say Simon Burgess and Carol Propper

Teenage boys who are violent, take drugs and indulge in under-age drinking are sometimes lightly dismissed as lads behaving badly. But our analysis of a survey of a group of Americans now in their mid-thirties indicates that the effects of such behaviour may be long-lasting and serious.

Teenagers who behave badly hugely damage their life-chances - with the exception of underage drinkers. Those who used hard drugs heavily as teenagers were, a decade later, paid wages 40 per cent lower than their clean-living peers. Men who were violent as teenagers suffered a similar loss of earnings by their late twenties. But those who drank alcohol under-age earned more than their well-behaved peers in later life, and cannabis use by teenagers had no effect on their later earnings.

Our research was based on the National Longitudinal Survey of Youth, a large, representative survey of Americans born between 1957 and 1965. Participants were interviewed first in 1979 and asked about violence and use of drugs and alcohol a year later. They have been interviewed every year since. We examined their earnings at the age of 28, approximately ten years after they first answered questions about their anti-social behaviour as teenagers.

É«ºÐÖ±²¥

We found that teenage drinking was common, with three out of four white men and one in two black men admitting to it. Half of the sample said they had smoked hash, and half had threatened or actually hit someone during the year. On the other hand, few took hard drugs as teenagers (slightly more whites than blacks did). Black men regularly taking hard drugs faced a bigger earnings loss than whites, but violent behaviour's impact on earnings was greater for white men.

Two factors make it likely that the full brunt of teenagers behaving badly is even greater than this research suggests. First, some of the behaviours occur together. Of the young men regularly taking hard drugs, 10 per cent also admitted to frequent violent behaviour, compared with just 1 per cent in the population as a whole. Such people have drastically lower earnings in later life. Under-age drinking is associated with cannabis use and, to a lesser extent, with hard drugs. Second, such activities may have indirect effects on earnings. If drugs use harms an individual's studies at school or college, this is another blow to future earning power.

É«ºÐÖ±²¥

One question is whether these results arise because of other factors. For example, it could be argued that young men in poor neighbourhoods may be more likely to take drugs or engage in violence and also to face worse earnings opportunities. To answer this, we repeated our analysis while controlling for a wide range of background factors such as parents' education and jobs and where people lived. The results were largely unchanged.

There are clearly a number of paths that individuals might take between a violent adolescence and an impoverished later life. They may continue their behaviour and spend spells in prison, or they may change their ways but lag behind peers on the career ladder, or they may engage in sporadic violence and bump along in dead-end jobs. We do not address this question (our dataset does not track continued violent behaviour or drugs use); we simply establish the fact of a substantial earnings loss persisting ten years after teenagers behave badly.

Simon Burgess and Carol Propper both work at the University of Bristol and at the Centre for Analysis of Social Exclusion, London School of Economics.