Parents commonly open financial accounts for their children during childhood, providing a financial foundation when they come of age. Using data from the French wealth survey, a nationally representative dataset recording individual-level asset holdings including for children, we examine the effect of sibship size on financial wealth held in children's names. We exploit two complementary identification strategies: a difference-in-differences design and an instrumental-variable strategy based on multiple births at second parity. We find suggestive evidence that an increase in sibship size reduces children's financial wealth, with effects that differ by birth order and are stronger for younger children. These effects operate partly through a reduction in overall household resources rather than a simple reallocation across siblings. Our findings suggest that the quantity-quality trade-off extends to financial investments in children, and that inequalities in wealth accumulation may emerge during childhood, well before major intergenerational transfers occur.

We study how union dissolution affects individual wealth accumulation and whether these effects differ between men and women. Using Dutch administrative data and a stacked event-study design, we estimate the short-term causal effect of union dissolution on individual net wealth trajectories. We find that union dissolution generates an immediate and substantial gender divergence in wealth: women experience a decline in wealth, whereas men experience an increase on average. The gap is primarily driven by housing wealth and substantial ownership. Men increase both assets and liabilities, with asset gains dominating, while women experience larger asset losses than liabilities reductions. We document three mechanisms shaping post-divorce wealth trajectories. First, inter vivos gifts provide informal insurance for a wealthy subgroup. Second, women experience a persistent decline in saving capacity relative to men. Third, repartnering mitigates long-run wealth losses for women. Our findings show that union dissolution is not merely a redistribution of existing resources, but a key moment in the production of durable gender wealth inequalities over the life course.

This study examines the impact of offspring loss on household wealth accumulation, savings behavior, and intergenerational wealth transmission. Using a combination of theoretical modeling and empirical analysis, we explore how the sudden death of an adult child alters economic decision-making in affected families. We develop a dynastic bequest model incorporating wealth-in-utility preferences, and the probability of dynastic continuation through grandchildren. Our theoretical framework predicts that wealth trajectories diverge significantly based on whether a child survives, whether grandchildren exist, and household structure (single-child vs. multi-child). Empirically, we employ a matching triple-difference regression strategy using Dutch register data to estimate the causal effects of child loss on wealth accumulation. We leverage exogenous variation from unexpected deaths to isolate the impact of offspring loss on bequests. Our findings contribute to the broader debate on bequest motives and the role of dynastic preferences in wealth accumulation, highlighting the differential responses of single-child and multi-child households. The results have implications for understanding intergenerational wealth inequality, particularly in the context of ageing societies.

This dissertation thesis highlights that wealth accumulation trajectories are deeply shaped by family dynamics and life events, from childhood to end of life. Combining French wealth survey and Dutch register data with quasi-experimental methods, I document how individual wealth trajectories respond to changes in sibship size, marital status, and dynastic continuity. Wealth inequalities appear early in life through differences in financial assets held directly by underage children in France. An additional sibling reduces eldest children’s wealth, consistent with the quantity–quality trade-off model and constrained resources at the household level. Union dissolution widens gender wealth gaps: women experience persistent losses, while men’s wealth increases on average. These diverging paths reflect inequalities in assets’ reallocation and in saving capacities, with repartnering and inter vivos gifts partially offsetting losses for some women. The last chapter studies how the death of a child affects parental wealth at old age. Wealth accumulation responds to the disruption of dynastic continuity, mostly with a substitution effect between illiquid and liquid assets.

Ce travail s'intéresse à la communication politique sur Twitter des députés français de la XVe législature (2017-2022). La mobilisation d'une méthodologie de classification supervisée de séquences textuelles appuyée sur un modèle de langage par représentation - CamemBERT - permet de produire des analyses de contenus fines. Le genre de l'élu(e) a un effet fort quant au contenu des textes postés sur les réseaux sociaux numériques. Ce discours genré est exemplaire d'une domination masculine du champ politique. L'analyse du contenu de ces tweets capture un espace politique sémantique parcouru par des rapports de force et où s'expriment des processus de légitimation. A l'aune d'une typologie des trajectoires professionnelles préalable au mandat législatif (Ollion, 2021), je montre que les discours répliquent des socialisations différenciées au champ de la politique parlementaire se traduisant en légitimités hiérarchisées dans l'arène politique.

This work focuses on the reactions to COVID-19 pandemic and sanitary policies on divorce-related queries on Google search engine in the United States. I also nowcast the crude divorce rates for 2020 thanks to Google Trends data. Event studies and triple difference analyses tend to identify a prominent lockdown effect over a pandemic effect. The lockdown renders a decrease in relative search volumes related to divorce queries, except for financial features of divorce which shows an important increase. The averaged crude divorce rate for 2020 that I estimate thanks to Google Trends data remains at a stable level, and is even estimated slightly higher than in 2017 and 2018 when unemployment data are included in the panel data model.