We investigate the impact of rising inequality on social instability through a novel multi-period laboratory experiment. In each period, members of two groups are randomly matched to play a battle of the sexes game, in which payoffs are either zero for both or positive but unequal. At the beginning of the game, the implied payoff inequality is low, but it increases over time. Our findings shed light on the effect of inequality on social stability. While under low initial inequality, unequal conventions - where one group is systematically better off  - tend to emerge, increasing inequality causes destabilization. This is initiated by the group disadvantaged by the unequal convention and is particularly pronounced when not only the relative, but also the absolute position of the disadvantaged group is worsening. These findings are consistent with a simple model incorporating disadvantageous inequality aversion. Finally, we show that history matters; responses to current inequality depend on past experiences of inequality and stability.

[PDF]

We study gender gaps in a multi-stage hiring process, focusing on recommendation and recruitment. Across two experiments - one with academic economists and one with a college-educated sample - we show how recommendation letters can cause gender hiring gaps. While recruiters favor female candidates, consistent with efforts to diversify hiring, some groups of recommenders choose less effective letters for women, undermining this advantage. Strategic but erroneous beliefs about letter effectiveness and gender views of recommenders explain these  choices. A two-sample prediction exercise using  letters from the academic job market supports the experimental finding that gender views drive the language used in recommendations.

[PDF]

Social sanctions sustain social order by reinforcing widely accepted principles. Political polarization may weaken this mechanism by fragmenting these principles, yet causal effects are hard to identify: observational data cannot separate the impact of polarized preferences from exposure to polarization. We test the effectiveness of social sanctions in a representative U.S. experiment (N = 2,400) that exogenously varies the degree of polarization. Participants allocate money between politically opposed recipients privately and publicly. When allocation choices are public, they can be punished by partisan Observers whose values are drawn from one of three distributions: tight (low variance), loose (high variance), or polarized (U-shaped). We find that polarization makes public allocations less equitable because participants (correctly) expect they may be punished even when acting fairly. Our results thus show that polarization causally undermines the disciplining role of social sanctions.

[PDF]

Even though reporting mistakes could substantially improve work processes and productivity within organisations, employees often hesitate to do so. This paper studies the role of  fear (of being fired) and futility (i.e. reports being inconsequential) in explaining such employee silence. Drawing on a principal-agent framework with career concerns, we formalise mistakes as noisy signals of both agent quality and the work environment and show that optimal reporting decisions are affected by fear and futility considerations. We then use a novel experiment to exogenously manipulate the degree of fear and futility and test our theoretical predictions. In a 2x2 between-subject design, we vary the anonymity of reporting and the likelihood of organisational response. Results show that reducing fear and futility are complementary actions. Tackling both significantly increases reporting by about 20pp. This improvement in communication is accompanied by better organisational income, highlighting the value of improved reporting structures for firms and employees.

[PDF]

This paper investigates how individuals understand fairness in the presence of negative income shocks. I combine evidence from two dictator-game experiments. In the first, income shocks are randomly assigned within the experiment; in the second, I use information on participants’ real-world income losses during the Covid-19 pandemic. Across both settings, shocks significantly affect distribution decisions. Individuals allocate more to those who lost income, regardless of whether they themselves or another person were affected. This effect cannot be explained by need considerations alone: even if two individuals have the exact same income at the moment of the distribution decision, participants allocate more to the person who suffered a shock. This highlights that the mere experience of shocks matters and is in line with a theoretical framework where past earnings serve as fairness reference points. Finally, I find that if information about shocks or reference points is incomplete, individuals use this ambiguity in a self-serving way.

[PDF]