Improving Information from Manipulable Data

TL;DR

This paper explores how decisionmakers can improve outcomes by intentionally underusing data to counteract manipulation, reducing information loss and enhancing allocation accuracy.

Contribution

It introduces a formal framework showing that committing to underutilizing data mitigates manipulation effects and improves decision quality.

Findings

Underutilizing data reduces the impact of manipulation.

Commitment to data underuse improves allocation accuracy.

Formal framework for data manipulation and decision commitment.

Abstract

Data-based decisionmaking must account for the manipulation of data by agents who are aware of how decisions are being made and want to affect their allocations. We study a framework in which, due to such manipulation, data becomes less informative when decisions depend more strongly on data. We formalize why and how a decisionmaker should commit to underutilizing data. Doing so attenuates information loss and thereby improves allocation accuracy.