The Algorithmic Barrier: A Framework for Artificial Frictional Unemployment and Information Asymmetry in Automated Recruitment Systems

Ibrahim Denis Fofanah

Papers from arXiv.org

Abstract: The United States labor market has entered a period in which high job vacancy rates and prolonged unemployment persist together. Classical theory attributes such conditions to skills mismatch or geographic immobility, but neither fully explains a pattern now widely reported: qualified candidates are rejected at the earliest, automated stage of hiring, before any human sees their application. This paper introduces Artificial Frictional Unemployment (AFU), a framework describing how deterministic automated screening rejects qualified candidates through semantic misinterpretation rather than genuine skill gaps. We situate the phenomenon within labor economics and information asymmetry theory and formalize the mechanism by which legacy Applicant Tracking Systems (ATS) turn hiring into a high-precision classification problem that inflates false negatives. The contribution is primarily conceptual. To make the mechanism concrete, we report a controlled proof-of-concept simulation comparing keyword-based screening with vector-space semantic matching under identical conditions. The simulation shows how lexical variance alone can produce false negatives; it is not a measurement of how much real-world friction is artificial, which we leave to future field studies. Building on the framework, we outline JobOS, a candidate-side architecture that illustrates how semantic competency mapping could operate alongside existing hiring infrastructure. Framing automated recruitment as labor market infrastructure, rather than a firm-level convenience, exposes a correctable source of matching inefficiency with consequences for workforce participation and the use of human capital.

Date: 2026-01, Revised 2026-07
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://arxiv.org/pdf/2601.14534 Latest version (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:arx:papers:2601.14534

Access Statistics for this paper

More papers in Papers from arXiv.org
Bibliographic data for series maintained by arXiv administrators ().