You should reject this paper: Dynamic agency, sequential evaluation, and learning in academic publishing

A variety of real-world situations take the form of repeated principal-agent problems with binary evaluation. I evaluate the principal's optimal evaluation effort and quality threshold for acceptance in the setting of dynamic agency with binary evaluation, focusing specifically on the evaluation by a top academic journal of papers submitted by economists. In the baseline model, the journal should statistically discriminate in favour of high-status economists, by setting an acceptance threshold that declines with their prior estimate of the economist's quality, while evaluation effort should likely increase with perceived author quality, as this provides incentives for authors to exert more effort, which is particularly valuable from the highest-quality authors. However, if a first good publication is more valuable than subsequent publications by the same economist, acceptance thresholds will tend to increase after publication success, relative to the threshold that would have followed a rejection.

Automated summary

This study evaluates the optimal evaluation effort and acceptance threshold for high-status economists' papers submitted to academic journals in a dynamic agency setting. The findings suggest that journals should statistically discriminate in favor of high-status economists by setting a declining acceptance threshold based on their prior estimate of economist quality, while evaluation effort should likely increase with perceived author quality.