Research
Research
Anticipatory Fertility: Conceptions, Births, and Abortions around Employment Shocks
Joint work with Anna Adamecz, Márta Bisztray, Ágnes Szabó-Morvai and Andrea Weber
Journal of Human Resources, 2026
We study fertility responses to employment shocks using unique Hungarian administrative data linking firm-level mass layoff and closure events to individual-level birth and abortion records. We show that the main fertility response occurs in anticipation and differs by shock type. The number of conceptions leading to births increases in the year preceding a mass layoff (when maternal employment protection is available), but not before a firm closure (without employment protection). We interpret this result as evidence of anticipatory fertility behavior: women tend to schedule childbirth when job loss can be avoided.
Media appearances: Defacto,telex (in Hungarian ) Die presse (in German)
Family foster care or residential care: the impact of home environment on children raised in state care
Joint work with Gábor Kertesi
Journal of Population Economics, 2025
In the media: CEU podcast, Telex defacto article
This study investigates how family foster care, compared to residential care, affects the outcomes of adolescents raised in state care. Using Hungarian administrative panel data, we contribute to the literature by examining previously unmeasured adult outcomes, conditioning on a rich set of variables observed in childhood. We show that adolescents raised in a foster family have substantially better outcomes as adults. Compared to similar peers in residential care, they are more likely to complete secondary education and have lower probabilities of using mental health medication and spending extended periods without either working or studying. For girls, teenage childbirth and abortion are less likely. Oster bounds indicate that selection on unobservables would need to be implausibly strong to explain away the estimated effects. Further evidence suggests that local child protection systems with more-abundant foster parents lead to better outcomes for children raised in state care.
Media appearances: CEU podcast, Telex defacto (in Hungarian)
The Effect of Physical Education Time on Students' Body Composition
I analyze the effect of physical education (PE) time on the body composition of students. Previous literature has established that spending more time at PE classes is largely ineffective in reducing children's body mass index (BMI). However, BMI can increase or stay unchanged even when body composition becomes healthier, as physical activity can reduce fat mass and build muscle mass at the same time. Therefore, relying solely on BMI as an outcome can be misleading in assessing the true potential of PE time on students' health. In my analysis I utilize a new outcome, body fat percentage, to overcome this problem. To identify the effect of PE time, I use the introduction of daily PE classes in Hungarian primary and secondary schools in 2012. The staggered implementation of the policy created a large variation in time ever spent with PE between multiple subsequent age cohorts, making it possible to compare cohorts with different past PE time, while controlling for year fixed effects. Using data on the whole population of Hungarian 5th to 12th grade students I estimate OLS models on the following outcomes: mean BMI, mean body fat percentage, and obesity defined by a BMI threshold and by a body fat percentage threshold. The estimated effects of past PE time are statistically insignificant and close to zero for all outcomes. Heterogeneity analysis reveals that increased PE time left the outcomes unchanged in specific subgroups of students (by age, income, gender, and school infrastructure) as well. The findings support the idea that increasing PE time alone is not sufficient to tackle childhood obesity.