Personality, Schooling and Occupational Choice over the Life-Cycle


It has long been recognized that cognitive skills are important determinants of labor market success (Becker (1964); Griliches (1977)), but there is increasing evidence that noncognitive skills also play a salient role. For example, Heckman et al. (2010) analyze direct measures of both cognitive and noncognitive skills that were gathered as part of the Perry Preschool Program evaluation. They find that the ability to plan and to exert self-control significantly affect lifetime earnings and employment outcomes for males. Devising effective social policies that maximize the potential for human development and welfare requires an understanding of the mechanisms through which cognitive and noncognitive skills influence choices and educational and labor market outcomes over the life-cycle.
To achieve this, in the proposed project we develop and estimate a dynamic model of schooling, work and occupational choices and quantify the importance of personality traits, as measured by the so-called “Big Five,” in explaining the inequality of expected lifetime utility at the age of 16. Our model allows both cognitive and noncognitive traits to affect educational and labor market outcomes through multiple channels, by affecting pecuniary or nonpecuniary returns from schooling or by affecting the utility of choosing white collar or blue collar occupations. Our analysis is inspired in part by Keane and Wolpin (1997), which estimates a similar type of model without personality traits. They find that 90 percent of the total variance in expected lifetime utility is explained by unobserved skill endowments and this paper explores whether personality traits constitute an important part of these endowments. The importance of the initial unobserved heterogeneity has also been found in a number of follow-up studies. For example, in the micro literature, Yamaguchi (2012) finds that the endowment differences prior to labor market entry account for 70% of the log-wage variance in the first year and 35% even after 20 years. Sullivan (2010) finds that 56% of the variance in discounted expected lifetime utility is explained by initial heterogeneity. In the macro literature that fits moments using more aggregated outcomes, researchers find similar results. For example, Huggett et al. (2011) conclude that 61.5 percent of the variation in lifetime earnings and 64.0 percent of the variation in lifetime utility is attributable to initial conditions. The accumulated evidence points to the importance of early life endowment heterogeneity as essential to explaining lifetime schooling and career paths. However, few papers explore the underlying sources of this heterogeneity. Keane and Wolpin (1997) find that the family background characteristics account for less than 10 percent of the total variation in expected lifetime utility. None of the existing studies examine the role of personality traits, probably because the longitudinal datasets used in these studies did not include personality trait measurements.

Funded By: 
Award Dates: 
July 1, 2016 - June 30, 2017