Lorenz Goette

We propose a novel experiment that prevents social learning, thus allowing us to disentangle the underlying mechanisms of social influence. Subjects observe their peer’s incentives, but not their behavior. We find evidence of conformity: When individuals believe that incentives make others contribute more, they also increase their contributions. Conformity is driven by individuals who feel socially close to their peer. However, when incentives are not expected to raise their peer’s contributions, participants reduce their own contributions. Our data are consistent with an erosion of norm adherence when prosocial behavior of the social reference is driven by extrinsic motives, and cannot be explained by incentive inequality or altruistic crowding out. These findings show scope for social influence in settings with limited observability and offer insights into the mediators of conformity.

Background: Roughly one quarter of short-term temporary deferrals (STTD) of blood donors are low-hemoglobin deferrals (LHD), i.e. STTD due to a hemoglobin (Hb) value falling below a cutoff of 125 g/L for female and 135 g/L for male donors. Since voluntarily donating blood is a prosocial activity, donors may perceive deferral as social exclusion, which can cause social pain, decrease self-esteem, and lead to antisocial behavior. However, little is known about the causal impacts of LHD on donor return. Study Design and Methods: We conducted a quasi-experiment with 80,060 donors invited to blood drives in the canton of Zurich, Switzerland, between 2009 and 2014. Within a narrow window of Hb values around the predetermined cutoff, the rate of LHD jumps discontinuously. This discontinuous jump allows us to quantify the causal effects of LHD on donor return, as it is uncorrelated with other unobserved factors that may also affect donor return. Results: We found different behavioral reactions to LHD for female and male donors. Female donors do not react to the first LHD. However, after any repeated LHD, they are 13.53 percentage points (p <0.001) less likely to make at least 1 donation attempt within the next 18 months and make 0.389 fewer donation attempts (p <0.001). Male donors react to the first LHD. They are 5.32 percentage points (p = 0.139) less likely to make at least 1 donation attempt over the next 18 months and make 0.227 (p = 0.018) fewer donation attempts. After any repeated LHD, male donors are 13.30 percentage points (p = 0.004) less likely to make at least 1 donation attempt and make 0.152 (p = 0.308) fewer donation attempts. Conclusion: LHD have detrimental impacts on donor return, especially if they occur repeatedly – suggesting that avoiding false LHD and helping donors to better cope with them helps to maintain the pool of prospective donors.