There is growing interest in how schools can shape children’s non-cognitive skills (sometimes known as behavioral traits, soft skills, or personality). One reason for this interest is that non-cognitive skills predict a wide range of economic outcomes, such as employment and lifetime earnings, but are also relatively malleable—at least until adulthood. This opens up a role for policy interventions aimed at school children. For example, schools could focus resources on cultivating positive non-cognitive skills (or eliminating negative ones) similar to the way that they develop math and reading skills. My research shows that policies aimed at reducing so-called “negative” non-cognitive or behavioral traits, however, could harm children in the long run.
Together with two co-authors, I have studied how classroom misbehavior relates to both educational attainment and labor market performance. Surprisingly, we find evidence that some non-cognitive skills that manifest as childhood misbehavior in the classroom (and are predictive of lower schooling attainment) are also predictive of higher earnings later in life. This finding challenges prevailing research, which has generally argued that misbehavior in the classroom reflects underlying skills that are bad for schooling and bad overall.
Our analysis uses data from a study following individuals born in 1958 in Great Britain. When these individuals were 11-years-old, the study asked their teachers to fill out a series of inventories describing each child’s behavior in the classroom. The study followed these children into adulthood, collecting information on their educational attainment and employment. The length of the survey enables us to relate each individual’s economic success to his or her childhood misbehavior in the classroom.
It has been long established in the psychology literature that the survey information collected from teachers on classroom behavior is statistically well summarized by two underlying factors, each reflecting a different non-cognitive skill. One of the factors captures anxious, aggressive, or restless outwardly expressed (thus: externalizing) behaviors. The second embodies withdrawn, inhibited (thus: internalizing) behaviors. Recognizing the distinction between these two underlying traits is the key. If we simply summarize all misbehavior, as some previous research has done, we find that misbehavior lowers schooling attainment and also lowers earnings. This is the basis for the widespread view that childhood misbehavior has a detrimental impact on all economically relevant outcomes.
However, when we recognize that misbehavior in the classroom can be reflective of two very different non-cognitive skills—externalizing and internalizing behaviors—a much more nuanced story emerges. Both of these characteristics are associated with lower schooling attainment. However, whereas internalizing behaviors, like being unforthcoming, depressive or withdrawn, predict lower earnings, externalizing behaviors, such as aggression, predict higher earnings. In other words, the externalizing factor lowers schooling attainment, but appears to have value in the labor market. This finding calls into question the role of schooling in identifying and cultivating skills that are productive.
We find that externalizing behaviors raise earnings even after we account for its negative impact on schooling. We show similar patterns for men and women, however we find large differences in how externalizing affects individuals depending on their socioeconomic status as children. Children from low-income families who exhibit externalizing behaviors see no benefit in the job market. This finding confirms previous results in the literature showing that policies that reduce externalizing behaviors among children from disadvantaged families can boost their lifetime earning potential.[i] However, our work also shows that for children who do not grow up in poverty, externalizing behaviors can be quite lucrative.
Our ongoing research asks whether patterns similar to what we found in Great Britain hold in the U.S. Preliminary evidence shows that for the same levels of externalizing behaviors, African American children face higher earnings penalties than white children. We suspect that part of this is due to higher rates of interaction with the criminal justice system for high-externalizing African Americans.
This research suggests several avenues for future caution in how we design policies targeting non-cognitive skills. This stands in contrast with health or cognition, which are generally perceived as “goods” that improve people’s lives. What we show is that non-cognitive skills, personality or behavioral traits can have very mixed effects. In our case, we show that a factor that predicts lower educational attainment also predicts higher earnings. Finally, our results on externalizing suggest that schools do not always foster the sorts of skills that are valuable in the labor market.
[i] Previous research in economics has shown that externalizing behaviors can hurt labor market prospects for some groups. In particular, a well-known study assesses an early education intervention known as the Perry Preschool Project. The authors show that the intervention lowered externalizing behaviors among a group of African American boys from low-income households, which led to higher earnings. When studying a nationally representative sample of individuals from Great Britain, we find that externalizing behaviors lead to higher earnings for children who do not grow up in poverty.
Assistant Professor of Economics, Johns Hopkins University
The Brown Center Chalkboard launched in January 2013 as a weekly series of new analyses of policy, research, and practice relevant to U.S. education.
In July 2015, the Chalkboard was re-launched as a Brookings blog in order to offer more frequent, timely, and diverse content. Contributors to both the original paper series and current blog are committed to bringing evidence to bear on the debates around education policy in America.
Read papers in the original Brown Center Chalkboard series »