Yikun Jiang

Reputation versus Privilege Incentives in Online Knowledge Contributions: Evidence from a Field Experiment and Structural Modeling. 

• User-generated content is crucial to online content platforms, yet how platforms can design incentives to stimulate user content contributions remains unclear. This paper distinguishes between reputation incentives (the desire for social recognition) and privilege incentives (the pursuit of platform functionalities unlocked at reputation thresholds). I design and implement a large-scale field experiment involving 12,182 users on a leading online question-and-answer platform, in which an experimental increase in reputation (via an anonymous upvote) simultaneously shifts both incentives. The treatment magnitude corresponds to a 19.6% increase in the median sampled user’s reputation. I find that the treatment significantly increases an individual’s probability of contributing additional answers by around 15% relative to baseline. The treatment effect emerges within the first three weeks, resulting in a persistent cumulative difference between the treatment and control groups. The effect is slightly stronger for users close to earning new privileges and weaker for more experienced users, while the quality and effort of subsequent answers remain stable. To quantify these incentives, I structurally estimate a model of contribution decisions, finding that users value an additional privilege at approximately 55% of the value of an additional upvote. Finally, I leverage a large language model (LLM) to assess the quality of approximately 200,000 historical answers by sampled users. Structural simulation using LLM-based quality assessments shows that undervaluation of high-quality contributions by new users substantially discourages participation. Together, these findings provide the first experimental isolation of reputation versus privilege incentives, integrate structural and experimental evidence, and inform the scalable design of non-monetary incentives on digital platforms.