This paper explores the use of a completely self-funded and highly scalable intervention: self-funded commitment  contracts to support individuals in achieving their goals of increased physical activity. We compare the effect of soft (non-incentivised) and hard (incentivised) commitment contracts using a randomised experiment with 1629 members of a large gym in Stockholm, Sweden. We find a significant positive impact of being offered a hard contract on monthly visits to the gym and the probability of meeting one’s contract goal. Hard and soft contracts increase gym visits by 21% and 8%,  respectively, relative to the control group. Trait self-control is negatively associated with the likelihood of accepting a contract and accepting to add stakes. The effect of soft and hard commitment contracts does not, however, differ by trait self-control. Importantly, the effects of both hard and soft contracts were greatest among participants who reported exercising the least at baseline.

It is well established that events in utero and infancy can have important consequences for later life outcomes, but less is known about the impact of events occurring between early childhood and adulthood. This paper investigates long-run health and socioeconomic effects of the introduction of universal health services in Swedish primary schools starting in 1944. Adoption of the reform was not immediate across all school districts, creating variation over space and time. We combine population-based administrative data with novel data on reform adoption at the school district level to identify the causal effect of access to school health services on health and socioeconomic outcomes using a difference in differences design. We also explore school records for a sub-sample of individuals encompassing information on school grades and sickness absences in primary school. Our results suggest few statistically significant, economically meaningful effects of school health services on health and socioeconomic outcomes in either the short- or long-run.

When they have children, women tend to make labour market decisions that result in substantial and persistent losses in earnings. In this pilot study, I experimentally investigate the role of children’s books in the early internalisation of norms regarding gender, family, and careers. I recruited 81 children and 39 of their parents from a preschool in California, and experimentally varied the book to which children were exposed. The treatment book contains a strong positive message about females and careers, whereas the control book contains no message on this topic. As a measure of internalised norms, I measured implicit and explicit biases immediately after the book reading and collected information about parents’ implicit and explicit biases through an online survey. I can thus correlate the biases of parents and their children. Participating children exhibit some explicit but little implicit biases, and results of the pilot study suggest that exposure to the treatment book may be associated with a reduction in these biases. Implicit biases are, on average, significantly greater among the parents than the children.