Research
I am an assistant professor at the Toulouse School of Economics, a CESifo Research Network Affiliate, and an IZA Associated Member.
I am a behavioral economist working on education and family economics. I use field experiments and panel data to study how the people around us, and parents in particular, shape what students aspire to and choose, and how we come to remember what we once wanted, as well as the consequences of these behaviors for individuals and society, such as social (im-)mobility and gender gaps in STEM.
I completed my PhD at UC Berkeley in 2022 and spent the following year as a postdoctoral fellow at the briq Institute on Behavior & Inequality.
Memory, identity, and belief about oneself
Selective memory of past aspirations
Abstract
Selective memory can lead people to mistake life outcomes as what they always wanted. Using data on past fertility desires from a panel tracking fertility preferences and actual fertility for 3,936 Kenyans from their early twenties to their thirties, I provide monetary incentives to remember and to be reminded of past desires. I find that 29 percent of respondents have more children by their thirties than once desired; that despite incentives, respondents with excess fertility misremember past desires in the direction of current fertility and do not want to be reminded of them, consistent with motivated memory; and that selective memory affects the intergenerational transmission of fertility preferences.
The price of identity: overoptimism and congruence concerns
Abstract
We study how social identity distorts both beliefs and preferences in investment decision-making. We ran framed field experiments in online sports betting with Kenyan and UK soccer fans during the 2021/22 Premier League season, eliciting subjective probabilities and incentivized bets on matches that sometimes involve a participant’s supported team and sometimes do not. Supporters overestimate their team’s win probability by 10–18 percent relative to neutral bettors and allocate about 20 percent more of their budget to bets on their own team than to otherwise similar matches where they are neutral. Embedding these data in a portfolio-choice model, we separate the role of distorted beliefs from that of identity-driven preferences. Distorted beliefs explain 30–44 percent of the excess investment in favored teams, while the remaining 56–70 percent reflects an aversion to profiting when one’s team loses, equivalent to valuing payoffs from such outcomes by 17–27 percent less. Exploiting variation in both recent and season-long performance, we show that this preference to avoid “disloyal” gains is stronger when teams perform poorly. Using the estimated model, we calculate welfare losses from identity-driven distortions of roughly 0.5–2.7 percent of betting budgets, with the exact magnitude depending on whether identity preferences are seen as welfare-relevant or as mistakes.
The illusion of stable fertility preferences
Abstract
Fertility preferences have long played a key role in models of fertility differentials and change. We examine the stability of preferences over time using panel data on Kenyan women’s fertility desires, expectations, actual fertility, and recall of desires across three waves over nine years. Although desired fertility is quite unstable, most women perceive their desires to be stable. Few expect their desired fertility to increase over time but such increases are common, and when asked to recall past desires, most respondents report previously wanting exactly as many children as they desire today.
Selective memory of past educational goals
Social influence on educational and career choices
Under parents’ watch: the effect of parental observability on educational choices
pdf·July 2026·Previously circulated as “Parental pressure and educational choices”·Coverage: Teach Economy, Berkeley Social Science Matrix
Abstract
This paper tests for parental observability as a driver of students’ post-secondary educational choices and as a mechanism behind socio-economic disparities in college attendance. I conduct a field experiment in Germany randomizing whether students’ stated aspirations are shared with parents. Parental observability strongly influences stated college aspirations and field rankings, triggering opposite reactions at different school types and doubling the socio-economic gap in college aspirations from 13 to 26 percentage points. While aspirations stated under parental observability better predict actual choices, aspirations elicited in private better predict satisfaction, persistence, and performance in the chosen field. Students’ stated motives point to the desire for parental recognition, rather than open conflict avoidance or salience of parental constraints, as the concern behind these adjustments.
Under pressure? Experimental evidence of parental and peer observability on students’ occupational aspirations
pdf·November 2026·With Lea Cassar, Denise Feigl, and Christina Felfe de Ormeño
Abstract
We provide causal evidence on how perceived social pressure shapes students’ occupational aspirations, tracing when such influences first emerge across school years. In a randomized control trial with 2,840 secondary school students, we vary the observability of stated occupational aspirations to parents or peers. Making aspirations visible reinforces stereotypical choices, increasing the STEM gender gap by 76% under parental and by 39% under peer observability, with effects already emerging in sixth grade. Notably, beliefs about occupations shift after aspirations were elicited under parental observability, raising caution about using them as causal determinants of career choices. Our findings highlight the need for interventions addressing social dynamics early, before aspirations adapt to pressure.