Publications by PSE researchers

Role model interventions are often designed to foster students' pursuit of specific careers and are commonly employed in science, technology, engineering, and mathematics (STEM). Here, we drew on social-psychological theories of intergroup relations to hypothesize that role model interventions might also unintentionally shape students' beliefs and preferences concerning the broader social system-their ideologies. Specifically, role model interventions may lead students to view the (inequitable) status quo in STEM as natural and acceptable. A systematic narrative review (35 articles, 42 studies) examined these hypothesized side effects. This review indicated that the ideological side effects of role model interventions were rarely considered in the literature on role models. Although limited, the few relevant findings revealed both undesirable side effects of role model interventions on students' ideologies (e.g., greater endorsement of the status quo) and effects that are-from our perspectivedesirable (e.g., greater awareness of gender bias in STEM). This review demonstrates that role models can be a double-edged sword and serves as a call to evaluate role model interventions based on criteria beyond motivation.

In September 2018, the French government implemented a suite of measures to reduce wage gaps between women and men. Among these measures, a requirement on all private companies with more than 50 employees to calculate their professional equality index. They must attain a minimum threshold on this index, or face sanctions. We analyse the efficacy of this index in highlighting pay inequality between women and men. After first explaining the rules for calculating the index, we analyse the results for companies in 2020 and compare them with male-female pay inequalities measured using other indicators. We observe that the pay gap indicator as measured by the index tends to minimise the true level of inequality, resulting from the exclusion of certain employees, from the option for companies to declare the index to be incalculable, and the methodologies chosen.

Using rich historical surveys on job tasks and advanced machine learning techniques, we study which jobs can be moved from the office to home over three decades in France. The share of jobs that can be done from home has increased steadily from 14% in 1991 to 45% in 2021. At the same time, actual Working From Home (WFH) remained limited to less than one fifth of its full potential before the Covid-19 crisis and is still below it in 2021. The growth of WFH is largely unrelated to the evolution of job tasks, implying that the main obstacles to WFH have not been technical constraints. Low-skilled employees in particular have jobs that have long been largely teleworkable but they were barely teleworking before the Covid-19 crisis and remained still below 50% of their full potential during it. This pattern is not explained by differences in workers' desire to telework. It implies that the well-known large inequality in access to WFH along the earnings distribution cannot be attributed only to feasibility constraints and is potentially inefficient.

We study the earnings responses to six large payroll tax and income tax reforms in France. We find evidence of full pass-through to workers in cases where there is a strong and clear relationship between contributions and expected benefits. By contrast, we find a limited pass-through of employer payroll taxes to workers for reforms with no tax-benefit linkage, and close to full pass-through to workers for income tax reforms nominally incident on employees. Together with a meta-analysis of the literature, we interpret these results as evidence that tax-benefit linkage matters for incidence of payroll taxes, a claim long made by the literature but not backed by empirical evidence to date. Absent tax-benefit linkage, our results suggest that the individual-level incidence of payroll taxes aligns with their statutory incidence.

While there is no gap in math performance at the beginning of Grade 1, a gap in favor of boys appears and widens during the first year of primary school. Using standardized national assessments administered during Grade 1 (CP) to more than 2.5 million pupils in France between 2018 and 2022, we show that this relative drop in girls’ performance is observed for all the cohorts and most of the exercises assessed. The greatest drop-off occurs among the best-performing girls at the start of Grade 1 (those in the top 1 % initially). These girls lose an average of nearly 7 percentile ranks at the start of second grade compared with boys in the same initial percentile. The emergence of a gender gap in math performance during Grade 1 is observed across all social categories and family compositions, and throughout the country. Girls lose slightly less ground compared to boys in classes where the top student in math is a girl, and in priority networks public schools (REP or REP+). However, characteristics of the school environment explain only a small part of the overall dynamics, suggesting that girls are losing ground compared to boys in every strata of society.

Payroll taxes are usually designed to fund social insurance and not to contribute directly to redistribution. Over the last fifty years, France has modified dramatically the schedule of payroll taxation, turning it into the most progressive part of its tax system. Using administrative data and detailed microsimulation model of labor income taxation, we show that pretax wage (or labor cost) inequality measured by the P90/P10 ratio has increased by 15.4%, while net wage inequality has actually decreased by 18.9% over the 1967–2019 period. This reduction in wage inequality can be largely attributed to the policy mix of reductions of employer payroll taxes for low wage earners joined with minimum wage increases. We discuss whether this unusual French experiment carries lessons for other countries.

We show in a large-scale field experiment that a brief exposure to female role models working in scientific fields affects high school students’ perceptions and choice of undergraduate major. The classroom interventions reduced the prevalence of stereotypical views on jobs in science and gender differences in abilities. They also made high-achieving girls in Grade 12 more likely to enrol in selective and male-dominated STEM programs in college. Comparing treatment effects across the 56 role model participants, we find that the most effective interventions are those that improved students’ perceptions of STEM careers without overemphasizing women’s under-representation in science.

Enjeux : En septembre 2018, le gouvernement français a pris un ensemble de mesures visant à réduire les inégalités femmes-hommes, parmi lesquelles la mise en place de l’Index de l’égalité professionnelle. Depuis 2019, toutes les entreprises privées de plus de 50 salariés doivent calculer cet indicateur et obtenir une valeur minimale à cet Index, à défaut de quoi elles s’exposent à des sanctions. Cet Index introduit, pour la première fois dans les politiques de lutte contres les inégalités liées au sexe, une obligation de résultat pour les entreprises et non seulement une obligation de moyens. Présentation : Cette étude a évalué l’impact de la mise en place de l’Index de l’égalité professionnelle sur les inégalités entre femmes et hommes dans l’entreprise. Pour ce faire, elle s’appuie sur les données déclarées par les entreprises et les données administratives détaillées concernant les rémunérations et heures travaillées par l’ensemble des salariés du secteur privé. Enseignements clés : 1. La couverture de l’Index est très imparfaite Seulement 50 % des entreprises (44 % des PME-PMI et 93 % des entreprises de plus de 1 000 salariés) normalement assujetties déclarent une note sur 100 points. Seul un quart des salariés du secteur privé sont concernés par le calcul de l’indicateur d’écart salarial de l’Index. 2. La mise en place de l’Index n’a pas d’effet détectable sur les inégalités femmes-hommes dans les entreprises concernées Les entreprises de moins de 50 salariés (non assujetties à l’Index) et celles de plus de 50 salariés (assujetties à l’Index) ont des trajectoires remarquablement parallèles en termes d’inégalités femmes-hommes entre 2010 et 2020. On n’observe pas de changement de tendance pour les entreprises assujetties après la mise en place de l’Index en 2018. 3. L’Index tend à invisibiliser les inégalités réelles entre femmes et hommes Les bons résultats obtenus par les entreprises pour les écarts de rémunération sont liés à deux choix méthodologiques discutables faits pour le calcul de l’indicateur d’écart de salaire de l’Index : l’application d’un seuil de tolérance (tous les écarts inférieurs à 5 % sont ramenés à 0) l’utilisation des hommes comme catégorie de référence pour normaliser les écarts (les écarts dans chaque groupe sont divisés par le salaire moyen des hommes, ce qui les réduit quand les femmes sont moins bien payées que les hommes mais les augmentent dans le cas inverse). 4. Les entreprises qui déclarent leur Index ne sont pas plus vertueuses – au sens de leur performance en matière d’égalité professionnelle – que celles qui ne déclarent pas Les données administratives utilisées dans l’étude permettent de mesurer les inégalités femmes-hommes à la fois pour les entreprises qui déclarent un Index et celles qui ne le déclarent pas : les entreprises qui ne déclarent pas leur Index ou déclarent ne pas pouvoir le calculer ne sont pas plus inégalitaires que les autres. Elles reportent en revanche des notes à l’indicateur d’écart salarial supérieures à celles recalculées à partir des données administratives. On observe que pour l’indicateur simple et transparent que constitue la part des femmes parmi les dix plus hautes rémunérations, les déclarations des entreprises concordent avec leurs résultats reconstruits alors que cela n’est plus vrai pour l’indicateur complexe et ajustable d’écart salarial. Cela suggère que l’utilisation d’indicateurs complexes et peu transparents peut permettre aux entreprises d’invisibiliser leurs inégalités réelles.

We show in a large-scale field experiment that a brief exposure to female role models working in scientific fields affects high school students’ perceptions and choices of undergraduate major. The classroom interventions reduced the prevalence of stereotypical views on jobs in science and gender differences in abilities. They also made high-achieving girls in grade 12 more likely to enrol in selective and male-dominated science, technology, engineering and mathematics programs in college. Comparing treatment effects across the 56 role model participants, we find that the most effective interventions are those that improved students’ perceptions of science, technology, engineering and mathematics careers without overemphasising women’s under-representation in science.

Even though females currently outnumber males in higher education, they remain largely underrepresented in math-related fields of study, with no sign of improvement during the past decades. To better understand which students drive this underrepresentation, we use PISA 2012 data on 251,120 15-year-old students in 61 countries to analyse boys’ and girls’ educational intentions along the ability distribution on math assessment tests. We analyze the percentages of boys and girls intending to pursue math-related studies or careers as a function of math performance. First, we show that for both boys and girls, there is a positive and linear relation between the probability of intending to pursue math and math performance. Second, the positive relation is stronger among boys than among girls. In particular, the gender gap in student intentions to pursue math-related studies or careers is close to zero among the poorest performers in math and increases steadily with math performance. Third, as a consequence, the gender gap in math performance, to the detriment of girls, is larger among students intending to pursue math than in the general student population.

Conditions de travail, qualité des emplois, choix managériaux et d'organisation, santé et sens du travail : un état de la situation française.