Labels

2026-06-24

AI Summary

This study addresses the design of optimal classification mechanisms when labels are acquired through costly self-selection, balancing information transmission against signaling costs. Integrating mechanism design and information economics, the authors formulate a signaling game to analyze the optimal categorization structure when agents incur costs to obtain labels. The analysis reveals that overly fine-grained certification induces efficiency losses, while merging lower-tier categories substantially reduces first-order signaling costs with only limited sacrifice of higher-order decision value. The paper further establishes sufficient conditions for optimal classification under a finite number of categories. These findings demonstrate that moderate information aggregation—such as imposing a lower bound or limiting the number of categories—enhances overall system efficiency, offering theoretical foundations for practical rating and certification systems.

Abstract

Labels -- grades, credentials, scores, ratings, ranks -- do two things. They inform receivers, and they give agents something to chase. I study optimal classification when labels must be earned through costly self-selection. I show that exact certification is inefficiently fine: pooling a small bottom interval saves first-order signaling costs while losing only higher-order decision value. I provide sufficient conditions for lower censorship to maximize efficiency as well as for every optimal classification to use finitely many categories.

Problem

• optimal classification
• costly self-selection
• signaling
• labeling
• pooling

Innovation

• optimal classification
• costly self-selection
• signaling costs
• pooling
• finite categories