Many negotiations---from legal disputes to labor contracts---are conducted through agents rather than directly by principals. While significant attention has been given to the misalignment of interests between principals and agents, little is known about how the process of selecting an agent affects bargaining outcomes. Across four preregistered experiments (N = 4,385), we show that principals systematically choose agents who go on to make overly aggressive offers. Principals on each side preferentially select agents whose beliefs about fair and achievable outcomes are especially favorable to their respective side. Thus, they send pairs of agents to the bargaining table whose expectations are strongly opposed and who are therefore more likely to reach impasse. Agent selection leads to worse outcomes than if principals chose randomly or negotiated on their own behalf, and outcomes could be improved by unilaterally selecting less aggressive agents. We document these patterns across distinct bargaining environments: an ultimatum-style settlement task and a real-time, open-ended negotiation.

Statistical discrimination relies on people inferring unobservable characteristics of group members based on their beliefs about the group. Across four pre-registered experiments (N = 9,002), we show that accurate information about the composition of top performers can induce incorrect beliefs about performance differences across groups when the groups are of unequal size. Because people fail to account for base rates, they underestimate the performance of individuals from smaller groups. As a result, when participants in our experiments receive true information about the gender composition of top performers in a male-dominated candidate pool, they are less likely to hire women, even when there are no gender differences in performance (Study 1). Similarly, they are less likely to hire better-performing non-White candidates when the racial demographics of the candidate pool reflect the US population (Study 4). We show that these choices reflect an error in statistical reasoning, rather than being motivated by a desire to discriminate against any particular group (Study 2). Despite leading to less accurate beliefs, participants disproportionately seek out information about top performers when given the choice, and discrimination thus persists when information selection is endogenous (Study 3).

Lack of trust is a key barrier to collaboration in organizations and is exacerbated in contexts when employees subscribe to different ideological beliefs. Across five preregistered experiments, we find that people judge ideological opponents as more trustworthy when these opponents support their opinions with self-revealing personal narratives than when they support their opinions with either data or with stories about third parties—even when the content of the messages is carefully controlled to be consistent. Trust does not suffer when explanations grounded in personal narratives are augmented with data, suggesting that our results are not driven by quantitative aversion. Perceptions of trustworthiness are mediated by the speaker's apparent vulnerability and are greater when the self-revelation is of a more sensitive nature. Consequently, people are more willing to collaborate with ideological opponents who support their views by embedding data in a self-revealing personal narrative, rather than relying on data-only explanations. We discuss the implications of these results for future research on trust as well as for organizational practice.