Research

Working papers:


A worker within a firm, or a researcher within the academia, is required to both cooperate with colleagues in team-projects and to compete with them for career progressions. Hence, within workplaces, individuals need to adapt when switching between tasks characterized by different levels of competitiveness and cooperativeness. In this paper, we experimentally investigate possible spillover effects on males’ and females’ cooperation measured by indefinitely repeated Prisoner’s Dilemmas, distinguished by two different levels of the competitiveness-cooperativeness index (CCI, Demuynck et al., 2022). Additionally, as the importance placed on competitiveness might differently impacts males’ and females’ attitudes towards the task, in our Decomposition treatment we separately present its zero-sum component and its common interest component.

While our results support the efficacy of the CCI in measuring the competitiveness and cooperativeness of the Prisoner’s Dilemma, the impact of the choice architecture intervention is different for whether the Prisoner’s Dilemma is highly competitive (CCI High) or highly cooperative (CCI Low). Our Decomposition treatment enhances cooperation in highly competitive setting (CCI High), whereas no such effect is observed in highly cooperative context (CCI Low). In addressing our main research question, we find that females are more likely than males to spill over their cooperative behaviour when switching from a low competitive environment to a high competitive one.




We experimentally test how information about the number of caught tax evaders, by interacting with individuals’ prior beliefs, affect the decision to underreport taxes. Specifically, our results indicate that when individuals receive the information about the number of people caught evading taxes and perceive this as higher than prior beliefs, they evade less. When, instead, individuals consider the number of caught evaders as low with respect to their beliefs, they evade more. These findings suggest that when subjects are informed on how many people have been found evading taxes they infer the audit probability, rather than the tax evasion rate. Finally, we observe no salience bias effect when considering individuals to whom we highlighted information about others’ norm violation nor when looking at those to whom we emphasised the probability of being audited.




Does the gender of the supervisor matter for career trajectories and compensation of both male and female workers? This paper exploits a fine-grained longitudinal personnel data on workers from an Italian insurance company over the period 2014-2021 and identify the gender composition of supervisor-subordinate dyads. Employing an individual fixed effect model, we show that while male and female supervisors evaluate similarly the performance of male and female workers, female supervisors are less likely to award one-off bonuses to both genders compared to their male counterparts. Additionally, both male and female workers are less likely of receiving promotions from employee to middle-manager when their supervisor is a woman compared to when their supervisor is a man. We interpret these findings as suggestive either of female supervisors adhering more strictly to masculine gender stereotypes associated with leadership roles or facing heightened scrutiny, leading to fewer promotions and bonuses being granted.