---
title: "Owner-level news-market concentration measures 1.30x higher than outlet-level measures"
type: "claim"
status: "seedling"
source_url: "https://arxiv.org/abs/2610.40164"
source_author: "Peter Carragher; Kathleen M. Carley"
source_date: "2026-10-29T00:00:00.000Z"
source_title: "Who Owns Digital News, and Who Gets Cited: Measuring Media Concentration Across 77 Countries and Seven AI Systems"
source_venue: "arXiv:2610.40164v1 [cs.CY]"
source_quote: "outlet-level measures understate concentration, with owner-level indices a median 1.30× higher"
source_tier: 1
source_sha: "33521fd5dc14fbd653efcf1482cf3bbd35bd9ea16d212128846c8b0cea92e1f6"
independent_roots: 1
audit_status: "capture-verified"
provenance: "Promotion from 10-inbox/raw/2026-10-31-outside-arxiv261040164-who-owns-digital-news-and-who.md, 2026-10-31"
date_created: "2026-10-31T00:00:00.000Z"
writer_model: "claude-sonnet-5"
origin: "batch"
derived_from: ["10-inbox/raw/2026-10-31-outside-arxiv261040164-who-owns-digital-news-and-who.md"]
tags: ["media-concentration","news-ownership","antitrust","hhi"]
---


Carragher and Carley (Carnegie Mellon, arXiv:2610.40164, submitted 29 Oct 2026) built a global news-ownership dataset — 9,761 domains, with reliable coverage in 77 countries — by linking digital news outlets to their ultimate corporate owners via Wikipedia/Wikidata, then cross-validated it against independent US (Medill) and UK (PINF) ownership directories and against Eli Noam's earlier cross-national census ([[entity-eli-noam]]; the paper reproduces his four-firm newspaper ranking at ρ=+0.465 and his [[entity-herfindahl-hirschman-index|HHI]] ranking at ρ=+0.407 across the 29–30 countries his 2008–2013 study covered).

Recomputing standard concentration indices — including HHI — at the owner level instead of the outlet level changes the measured answer: across the dataset's countries, owner-level concentration is a median factor of 1.30x higher than the equivalent outlet-level figure. Treating affiliated outlets (shared parent company, but separate mastheads or domains) as independent competitors therefore understates how concentrated a national digital-news market actually is. The gap is a methodological artifact of unit choice, not a claim about any single market or merger — it is the paper's baseline finding, which the rest of the paper's AI-citation results are then measured against at both levels.

> [!note] Seek's commentary:
> Noam's census counted mastheads by hand across two or three decades; this one counts domains by scraping Wikidata. Same arithmetic error both generations made without the scrape — count the children, not the parent — and both generations needed someone to go looking for the parent before the error showed itself as a number. 1.30x is what the oversight was worth, median, worldwide.

## Further verification
A single unrefereed preprint (no independent replication yet) underlies this and two sibling notes — see [[entity-herfindahl-hirschman-index]] for the shared methodology and [[question-corroborate-carragher-carley-2026-ai-overview-concentration-scenario]] for the fourth finding from this same source that the vault's single-source cap routed to the question queue instead of a claim-note.