PhD thesis defence by Edoardo Frattola
This thesis consists of four independent chapters covering topics related to family, gender, and education economics.
In the first chapter, I investigate whether parental retirement affects the timing of adult couples’ fertility decisions and whether the effect is heterogeneous across family policy regimes in Europe. Exploiting variation in eligibility rules in a regression discontinuity design with panel data, I find a positive causal effect only in countries with weak family policies and strong family ties (i.e. in Mediterranean countries), where the result is driven by an increase in the availability of informal childcare. This finding suggests that raising the minimum retirement age might have unintended negative consequences on fertility rates by delaying the offspring generation’s childbearing decisions.
In the second chapter, coauthored with Marco Tonello, we leverage newly available rich administrative data to study the heterogeneous evolution of fertility and newborn health during the pandemic. We focus on Tuscany, a representative region of Italy, which was one of the first countries to experience the severe impact of the COVID-19 outbreak in early 2020. Our findings indicate a decline in the number of births relative to the pre-pandemic trend in late 2020 and early 2021, roughly nine to twelve months after the pandemic onset. However, starting in March 2021, birth numbers consistently exceeded the pre-pandemic trend, resulting in a cumulative "baby bump" compared to the counterfactual scenario. This aggregate increase conceals significant heterogeneity across sociodemographic groups, with positive deviations entirely driven by native, educated, and employed parents. During the same period, newborn health indicators showed no signs of deterioration and, if anything, slightly improved.
In the third chapter, coauthored with Andrea Cintolesi, we analyse the dynamics of gender representation in leadership positions, focusing on whether male leaders, under public pressure, are more or less inclined than their female counterparts to appoint women to executive roles. Using a regression discontinuity design based on mixed-gender municipal elections in Italy from 1993 to 2019, we find that male mayors are significantly more likely to appoint women to their executive committees. Notably, this effect is contingent on public sensitivity to gender issues, being present only in progressive regions and municipalities with high female voter turnout. Moreover, it does not extend to appointments in less visible roles, such as those on the boards of municipal state-owned enterprises. This study emphasises the conditional nature of male leaders’ support for gender equality and offers broader implications beyond the political sphere.
In the last chapter, coauthored with Niccolò Cattadori and Elena Lazzaro, we estimate the causal effect of university department excellence funding on several outcomes, including faculty hiring, educational offerings, student attraction, and scientific productivity. For the first time, we link administrative data on the careers of all Italian academics to their publications, department educational activities, and student enrollment. Using a dynamic difference-indifferences design, we find that excellence funding significantly increased faculty size (a 6.4% rise in funded departments), but had no impact on educational offerings in terms of new degree or PhD programs. The funding led to a 13.3% increase in student enrollment after three years, suggesting that the excellence designation likely acted as a quality signal, making the departments more attractive to students, particularly high-achieving ones. The funding also boosted scientific output, with a 10.1% rise in publications. While a larger faculty size contributed to this increase, large STEM and Life Sciences departments also saw a rise in individual productivity, likely due to agglomeration effects and improved research infrastructure. No such effect is found in Social Sciences and Humanities departments.
The event will take place in hybrid modality.