Educational curricula in many countries and, arguably, most parents are strongly concerned with promoting children’s cognitive skills at young age. The strong and unambiguous correlation between cognitive ability on the one hand, and economic and life outcomes on the other hand, justifies the concern with cognitive skills (Burks et al 2009, Heckman et al 2006, Firkowska-Mankiewics 2011).

However, there is convergent evidence that, independently of cognitive skills, children’s self-control, ie the ability to control their impulses and to delay gratification, is highly important for academic, economic, and health outcomes during adolescence and beyond (Duckworth and Seligman 2005, Sutter et al forthcoming, Moffitt et al 2011). This evidence comes from different fields of the social sciences, including sociology, psychology, and economics. For example, in a study with more than 500 high school children, Sutter et al (forthcoming) find that more patient children are less likely to smoke or consume alcohol, have lower body-mass indexes, save more of their pocket money, and have better grades for conduct at school, compared to their impatient classmates. In this study, patience was measured using monetary incentives in a delay gratification task, as commonly done in the experimental economics literature.

The observed childhood choices and outcomes have a strong impact on adulthood outcomes. Apart from the indirect effects of self-control through childhood outcomes on adulthood outcomes, Moffitt et al (2011) also provide evidence for a direct effect of young age self-control ability on later life outcomes. Notably, these effects are robust if the authors control for family socioeconomic background and cognitive ability. Consistent with these findings, a strong link between self-control and life outcomes has also been found by Mischel et al (1989) for a large set of outcome measures, and by Chabris et al (2008) for health outcomes. Importantly, higher cognitive ability may not always be a good substitute for self-control in the acquisition of important academic and technical skills (Duckworth et al forthcoming). Self-control and self-control strategies seem crucial prerequisites for good life outcomes independently of cognitive ability.

Learning self-control?

The importance of self-control suggests that it can be beneficial to promote self-control skills early in life. The seminal research by Mischel et al (1989) already showed that children are able to develop self-control strategies. More recently, Romer et al (2010) find that confronting youths with self-control situations and letting them experience the potentially negative outcomes of impulsive behaviour in a controlled way, increases their ability to delay gratification. Similarly, the ability to take a future time perspective helps children to delay gratification in the presence of tempting immediate rewards. Riggs et al (2006) and Diamond et al (2007) present results from elementary and preschool programmes that successfully increase young children’s ability to exert self-control. These results are consistent with the social-psychological notion of self-control as a muscle that needs training to develop its full strength (Muraven and Baumeister 2000).

The study by Moffitt et al (2011) showed that children at all levels of self-control strength can benefit from increases in this ability. This suggests that interventions could be applied broadly across the self-control spectrum. However, self-control interventions compete with other important educational goals. It may therefore be necessary to identify children with low self-control, to channel resources to these individuals because they are most at risk of ending up with poor life outcomes.

Measurement: Early detection of at-risk youths

There are two approaches to the identification of low-self-control children in the literature. On the one hand, psychological rating scales can be elicited from parents, teachers, or the children themselves (see for example the detailed discussion in Duckworth and Seligman 2005). On the other hand, time-discounting tasks that offer the participants a choice between an immediate payoff and a larger delayed payoff have been used by both psychologists and economists to measure self-control (Duckworth and Seligman 2005, Chabris et al 2008, Sutter et al forthcoming). These choice-based tasks are often incentivised, using real monetary incentives

While rating scales are cheap and easy to elicit, they are not based on revealed preferences. They may be subject to manipulation, if teachers or parents want to affect the allocation of resources. Moreover, rating scales may in some cases be tautological if they contain items that overlap with the behaviours they are supposed to predict. For example, an item asking about smoking behaviour may enter a questionnaire about self-control in health behaviours, which is then used to predict smoking behaviour.

Discounting experiments with monetary payoffs have been shown to be feasible and meaningful with children as young as age ten, such that a direct elicitation from the children of interest is possible. Discounting tasks are therefore less likely to be manipulated. Because of their abstract framing, discounting tasks are more likely to tap into a general preference trait. Similar to the measurement of academic performance by weighting both class reports and standardised tests, we believe that policymakers should therefore apply a combination of teachers’ ratings and revealed discounting preferences.

Policy implications

We have discussed the strong evidence that the ability to exert self-control and develop self-control strategies at a young age is important for a child’s development. In particular, self-control at a young age affects academic, economic, and health outcomes long after the child has grown up. Because self-control skills can be trained, early life interventions are possible, with potentially large welfare benefits. Although individuals at all levels of the self-control scale seem to benefit from increases in self-control ability, scarce resources may be warranted to identify at-risk children and youth and to direct resources for interventions to them. Because children from disadvantaged backgrounds often score low on self-control ability (Evans and Rosenbaum 2008, Moffit et al 2011), and may also be those for which parents may least be able to promote these skills, interventions as part of the formal education may therefore serve distributional purposes. Instead of redistributing (economic) outcomes later in life, such early interventions allow children from disadvantaged backgrounds to develop better self-control skills. In this way, it is possible to redistribute along the whole spectrum of life outcomes, including outcome categories that are otherwise hard to redistribute, like health, occupational skills, and education.

Burks, SV, JP Carpenter, L Goette, and A Rustichini (2009), “Cognitive skills affect economic preferences, strategic behaviour, and job attachment”, Proceedings of the National Academy of Sciences, 106:7745–50.
Chabris, CF, D Laibson, CL Morris, JP Schuldt, and D Taubinsky (2008), “Individual laboratory-measured discount rates predict field behaviour”, Journal of Risk and Uncertainty, 37:237–69.
Diamond A, WS Barnett, J Thomas, and S Munro (2007), “Preschool program improves cognitive control”, Science, 318:1387–88.
Duckworth, A, P Quinn, and E Tsukayama (forthcoming), “What No Child Left Behind leaves behind: The roles of IQ and self-control in predicting standardized achievement test scores and report card grades”, Journal of Educational Psychology.
Duckworth, AL, and MEP Seligman (2005), “Self-discipline outdoes IQ in predicting academic performance of adolescents”, Psychological Science, 16:939–44.
Evans, GW, and J Rosenbaum (2008), “Self-regulation and the income-achievement gap”, Early Childhood Research Quarterly, 23:504–14.
Heckman, JJ, J Stixrud, and S Urzua (2006), “The effects of cognitive and non-cognitive abilities on labor market outcomes and social behaviour”, Journal of Labor Economics, 24: 411–82.
Firkowska-Mankiewics, A (2011), “Adult careers: Does childhood IQ predict later life outcomes?”, Journal of Policy and Practice in Intellectual Disabilities, 8:1–9.
Mischel, W, S Yuichi, and ML Rodriguez (1989), “Delay of gratification in children”, Science, 244:933–38.
Moffit, TE, L Arseneault, D Belsky, N Dickson, RJ Hancox, H Harrington, R Houts, R Poulton, BW Roberts, S Ross, MR Sears, W M Thomson, and A Caspi (2011), “A gradient of childhood self-control predicts health, wealth, and public safety”, Proceedings of the National Academy of Sciences, 108:2693–98.
Muraven, M and RF Baumeister (2000), “Self-regulation and depletion of limited resources: Does self-control resemble a muscle?”, Psychological Bulletin, 126:247–59.
Riggs, NR, MT Greenberg, CA Kusche, and MA Pentz (2006), “The meditational role of neurocognition in the behavioral outcomes of a social-emotional prevention program in elementary school students: Effects of the PATHS curriculum”, Prevention Science, 7:91–102.
Romer, D, AL Duckworth, S Sznitman, and S Park (2010), “Can adolescents learn self-control? Delay of gratification in the development of control over risk taking”, Prevention Science, 11:319–30.
Sutter, M, MG Kocher, D Rützler, and ST Trautmann (forthcoming), “Impatience and uncertainty: Experimental decisions predict adolescents’ field behavior”, American Economic Review.

420 Reads