Self-Control and Financial Success: What the Dunedin Study Found About Kids and Money
Childhood self-control is linked to adult money outcomes, modestly. In a 2011 study by Moffitt and colleagues, 1,037 children from Dunedin, New Zealand were followed from birth to age 32. Children rated lower in self-control between ages 3 and 11 were, as adults, less likely to be saving and more likely to report money-management difficulties and credit problems, even after accounting for family social class and IQ.
That is a pattern across a group, and it is no verdict on any one child. Self-control looks partly stable in these data, and none of the studies tested whether a program, a class or a parenting method changes it, so they say nothing either way on that question.
A portfolio that runs for weeks
The Dunedin findings link money outcomes to planning ahead and managing impulses, and a course can give students repeated practice at both. Students work with a simulated $10,000 portfolio and live Nasdaq pricing, in formats that run up to 28 weeks. A longer stretch gives a student more chances to face the choice between reacting to a rough week and holding to a plan, with no real money at stake. The course's project-based modules sit alongside the simulator.
No study has tested whether a portfolio project changes anyone's self-control, and this post does not suggest ours does. The design bet is narrower: practice at the decision itself.
What did the Dunedin study find about self-control and money?
The Dunedin Multidisciplinary Health and Development Study followed a complete birth cohort, 1,037 children born in one New Zealand city in a single year, with 96% retention to age 32. Observers, teachers, parents and the children themselves rated self-control at ages 3, 5, 7, 9 and 11, and the ratings were combined into one measure.
Moffitt's team found that at 32, children with lower self-control were less likely to save and had fewer of the financial building blocks of adulthood, such as a home, investments or a retirement plan. They also reported more money-management difficulties and had more credit problems, and informants who knew them well rated them as poorer money managers. The authors describe the study as "observational and correlational".
The link was small. Our translation of the published coefficient: a child one standard deviation below average in self-control was predicted, on average, to land near the 56th percentile on a financial-struggles scale instead of the 50th. Self-control predicted financial struggles better than family social class or IQ did, but all three were modest predictors.
The same cohort was followed to age 45 by Richmond-Rakerd and colleagues. Children with better self-control had more practical financial knowledge, more financial planfulness and better credit ratings in midlife. On a combined measure of financial preparedness, a child one standard deviation above average in self-control was predicted to land near the 63rd percentile instead of the 50th (again our translation), and about the 57th once family social class and IQ were counted. The link shrinks with family background in the picture, and it does not disappear.
Do other studies back it up?
Partly. A US and UK replication reproduced the broad pattern but found the finance links weakened with extra controls. A large meta-analysis links early self-regulation to adult outcomes, though the outcomes its abstract names are not financial. And one small pre-registered follow-up found the preschool waiting measure alone did not predict mid-life money outcomes. The table sets out all five studies.
| Study | Design | People | Self-control and money finding |
|---|---|---|---|
| Moffitt et al. (2011), PNAS | Birth cohort followed to age 32; self-control rated ages 3 to 11. A separate sample of 509 same-gender, non-identical British twin pairs followed only to age 12 | 1,037 children in Dunedin, NZ | Lower childhood self-control predicted less saving, more money-management difficulty and more credit problems at 32. In the twin sample, the twin with poorer self-control at 5 was more likely at 12 to smoke, do poorly in school and show antisocial behavior (no money outcomes) |
| Richmond-Rakerd et al. (2021), PNAS | Same Dunedin cohort followed to age 45 | 938 of 997 living participants assessed at 45 | Better childhood self-control went with more practical financial knowledge, more planfulness and better credit ratings |
| Robson, Allen & Howard (2020), Psychological Bulletin | Meta-analysis of 150 studies | 215,212 participants in total | Self-regulation in early school years was related to adult outcomes including unemployment, crime and mental health up to about age 38; the abstract lists no financial outcomes |
| Koepp et al. (2023), Developmental Psychology | Conceptual replication in two cohort studies | 1,168 people in the US and 16,506 in the UK | Largely reproduced Moffitt's pattern in cohorts born in different decades; ties to finances weakened once home environment and achievement were added |
| Benjamin et al. (2020), Journal of Economic Behavior & Organization | Pre-registered follow-up of a preschool wait task | 113 people from the 1967 to 1973 Bing preschool studies, surveyed in their late 40s | The preschool wait measure alone did not predict 11 mid-life money outcomes (average correlation 0.02) |
What can the self-control studies not tell parents?
It does not show that childhood self-control causes financial outcomes. The designs are observational. The authors adjusted for social class and IQ, and the twin comparison in Moffitt's paper helps rule out shared family background, but it looked at age-12 outcomes and not money, so it does not settle the question for adult finances. In the replication, the links to education, work and finances shrank when the researchers also accounted for children's home environment and achievement.
It does not tell us whether self-control can be changed. In Moffitt's data, childhood self-control correlated 0.30 with a self-control measure taken in young adulthood, which the authors call moderate: stable enough to be measured at age 3 and still say something at 32, yet some children changed rank. Moffitt's team cautioned that this natural change is not necessarily equivalent to change produced by an intervention. None of the studies above tested a program.
The marshmallow test fared poorly in one small follow-up. In one small pre-registered follow-up of 113 people, the preschool waiting measure on its own did not predict 11 mid-life money outcomes. A composite that combined the preschool task with self-rating surveys at 17, 27 and 37 did predict 10 of the 11, at an average correlation of 0.19, and adding the preschool task made no difference to its predictive power.
"Self-control" is also a broad label. In the replication it meant a composite of attention and behavior problems, including hyperactivity, inattention and impulsive aggression. Dunedin is one cohort, born in 1972 and 1973 in one city, and the Richmond-Rakerd follow-up uses the same people, so it is no independent confirmation.
Money habits that don't depend on a daily decision
These are our suggestions, each with the research reason beside it; the studies did not test them.
- Don't read this as a grade on your child. The effects are small to moderate and describe averages across about a thousand people.
- Set up automatic saving. Moffitt's team suggested that default arrangements, such as saving you have to opt out of, lean on less effort and planning. At home, that can be an automatic transfer to savings the day an allowance or paycheck arrives. Our post on allowance for teens is a place to start.
- Give a goal a price and a date. Richmond-Rakerd's team measured financial planfulness, and a price tag and a set date give a child something concrete to plan toward.
- Let a long-horizon account run. A small custodial account or a simulated portfolio, checked once a month instead of every day, gives a teen experience of holding through ups and downs. Our guide to teaching kids about investing lays out ways to start.
- Talk about the wait afterward. Ask what it felt like to hold off and what they would do differently.
Frequently Asked
Questions
In the Dunedin study, children rated lower in self-control had more money-management and credit trouble and less saving at 32, and the pattern held at 45. A US and UK replication found similar patterns, though the links to finances weakened once home environment and achievement were counted. These are modest, group-level associations and not forecasts for individual children.
These studies cannot say. The Dunedin authors observed that some children's self-control rank changed over time, but cautioned that natural change may not be the same as change from an intervention. Whether or not self-control can be changed, parents can choose how much a money habit depends on it, for example through automatic saving. That is our suggestion, not a tested result.
Not by itself, in one pre-registered follow-up. Benjamin and colleagues followed 113 people from the original Stanford preschool studies into their late 40s, and the preschool wait measure did not predict 11 money outcomes (average correlation 0.02). A broader self-regulation index that included surveys from ages 17 to 37 did predict most of them.
If you want your child to practice long-horizon money decisions, see how Rapunzl works for families.
By Maria Rodriguez, Curriculum Designer at Rapunzl.