时间：2026年7月6日下午2：00—5：00
地点：邵逸夫科学馆206会议室
题目：Washed Out by the Crowd? Accountability under Sequential Review
内容摘要：When a sequence of later reviewers revises an initial analysis, it may improve the organization’s final decision while making it harder to tell how much credit belongs to the first analyst. This paper studies accountability in such sequential review processes. An analyst chooses hidden effort, which makes a binary initial report more likely to be correct. Later reviewers then react to the report and to their own information, and the organization can reward the analyst only through the record this process leaves behind. The main result shows that any downstream record has incentive value only insofar as it distinguishes two branches: histories that follow an initially correct report and histories that follow an initially mistaken one. Effort matters through the probability of entering these branches; downstream review matters through how much the final record still separates them. In a benchmark sequential-learning environment, this logic gives a simple contrast between final-performance rewards and report-survival rewards. When later reviewers can overturn early mistakes, review improves decisions but gradually washes the analyst’s effort out of final correctness. First–last agreement, by contrast, can remain informative about whether the initial report was right. When later reviewers instead defer mechanically to the first report, the ranking reverses: final correctness preserves the initial effort imprint, while agreement carries no incentive content. At finite horizons, the comparison is governed by the probability that review repairs an initial mistake. Which reward is better comes down to one thing: how likely review is to repair an initial mistake.
主讲人介绍：叶思茗，现为乔治城大学经济学博士。她的研究方向为微观经济理论，研究兴趣包括决策理论、社会选择与信息经济学。她的论文发表于Journal of Economic Theory。
