Optimal Selection Using Algorithmic Rankings with Side Information

TL;DR

This paper investigates how noisy rankings and binary signals about candidate availability affect optimal selection strategies in online matching platforms, revealing counterintuitive effects of improved ranking accuracy.

Contribution

It introduces a model analyzing the impact of ranking noise and availability signals on selection outcomes, highlighting surprising effects of increased ranking accuracy.

Findings

Higher ranking accuracy can lead to suboptimal social outcomes.

Increased accuracy may cause selection of lower-ranked or busy candidates.

Counterintuitive effects of ranking improvements are identified.

Abstract

Motivated by online platforms such as job markets, we study an agent choosing from a list of candidates, each with a hidden quality that determines match value. The agent observes only a noisy ranking of the candidates plus a binary signal that indicates whether each candidate is "free" or "busy". Being busy is positively correlated with higher quality, but can also reduce value due to decreased availability. We study the agent's optimal selection problem in the presence of ranking noise and free-busy signals and ask how the accuracy of the ranking tool impacts outcomes. In a setting with one high-valued candidate and an arbitrary number of low-valued candidates, we show that increased accuracy of the ranking tool can result in suboptimal social outcomes. For example, increased accuracy may mean that agents may be more likely to make offers to busy candidates, and (counter-intuitively) may be more likely to select lower-ranked candidates. We further discuss conditions under which these results extend to more general settings.