Who Owns the Channels

Media Concentration and Counter-Architectures

Walker Briefing is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.

A democracy can reform the way it raises political money, redesign the chamber its representatives sit in, and rebuild the machinery by which it counts votes, and still discover that the questions worth reforming never reached the public in the first place. Before a population can deliberate, it has to learn what is at stake. The channels through which it learns are owned. Who owns them, and how few of them there are, settles in advance which questions arrive carrying the weight of public attention and which never surface at all. This is the upstream condition. It sits above the remedies this series has already examined, because it conditions the formation of the public will those remedies presuppose.

The architecture that does this conditioning in Australia is not a single thing. It has two layers that are easy to fold into one and should not be. There is an origination layer, the proprietors who produce professional journalism, concentrated in a way that has few parallels in the comparable democratic world. And there is a distribution layer, the platforms that now carry most of journalism’s reach, separately owned, separately concentrated, and frequently at odds with the first. The primary subject here is the origination layer. The distribution layer is named as the second concentration the same logic reaches, and held there. What follows is a description of that architecture as it stands in the middle of 2026, and an examination of whether anything on offer can change it.

Begin with who. As of mid-2026 the dominant commercial groups are News Corp Australia, Nine, and Southern Cross, the entity that emerged when Southern Cross Austereo and Seven West Media merged, a combination the Australian Communications and Media Authority approved subject to divestment commitments, and which now carries the former Seven assets including the Seven Network and, in Western Australia, The West Australian. Kerry Stokes, through his listed vehicle, remains the largest shareholder in the merged group at around a fifth of it. A financing arrangement backed by Gina Rinehart’s Hancock interests, reported in May 2026 and lodged with the exchange, supports a holding of roughly nine and a half per cent. These are filings, not motives, and the distinction governs everything that follows. The public broadcasters, the ABC and SBS, sit alongside the commercial groups; Australian Community Media and a scattering of independent and digital-native outlets hold the remainder. News Corp Australia is part of the United States-listed News Corporation, which is a structural fact relevant later to the question of foreign ownership and an imputation of nothing.

Now the harder question, which is how much. The claim that three entities control the national narrative is the kind of sentence that feels true and turns out to depend entirely on the instrument used to measure it. Measured by metropolitan newspaper circulation, it is close to true: News Corp Australia alone is reported to hold something on the order of two-thirds, a figure traceable to the company’s own audience data and to comparative academic work, and consistent with its publishing seven of the ten highest-circulating mastheads. Measured by free-to-air and subscription television revenue, a frequently cited figure put three corporations at about eighty per cent, though that number predates the 2026 merger and should be read as the order of the concentration rather than a current measure. Measured by online reach, the claim does not hold cleanly at all. The widest-reaching online brands include the ABC, news.com.au, nine.com.au, the Guardian, and a set of digital-native outlets, a more crowded field than the print one. So the claim holds or fails on the metric, and anyone who makes it owes you the metric. Concentration here is high by international comparison: the Global Media and Internet Concentration Project at Carleton placed Australia among the most concentrated it surveyed in 2024, and earlier comparative work, on data predating the rule changes, found its newspaper ownership the most concentrated of the countries studied. Those are two findings from two projects, measuring different things, and they are worth more kept apart than fused into one slogan.

The architecture acquired its present shape through decisions that can be named. The most consequential was the 2017 repeal of the two-out-of-three cross-media rule, which had stopped one proprietor from owning newspaper, radio, and television in the same market. That repeal is a matter of legislative record, and the consolidations that followed it, the union of Nine and Fairfax in 2018 and the regional tightening completed by the 2026 merger, came after it. In most regional markets the merger reduced three commercial groups to two. The same pattern reaches Western Australia, where the masthead that covers the state now sits inside the national combination. I name that as an instance of the national architecture and not as a grievance, because the writer’s own location is not the subject of the analysis.

There is a second concentration the print and television figures miss entirely. The Digital News Report: Australia 2025 found that for the first time social media, at around twenty-six per cent, overtook online news websites, at around twenty-three per cent, as Australians’ main source of news, with television still ahead at about thirty-seven per cent and Facebook the single most-used platform for news, reaching close to two in five. The platforms that now carry most of journalism’s reach are owned separately from the proprietors who produce it, and they are concentrated in their own right. The same survey found that traditional news brands remain the most-accessed source type on every major platform, which is the fact that stops this from becoming a story about news being replaced. So neither slogan is right. The making of professional news is concentrated in a few proprietors; its delivery is concentrated again in a few platforms, separately owned; and the online world, varied as it looks, is still fed mostly by those same newsrooms.

Watch the architecture meet the limit of its own logic, because that is where its workings become visible. For two decades the business model that funded Australian journalism, advertising sold beside the news, has been migrating to the platforms that distribute the news without paying to make it. The News Media Bargaining Code, enacted in 2021, was the origination layer’s attempt to charge the distribution layer rent. Its design is more revealing than its result. No platform was ever formally designated under it. Designation was a power the Treasurer held and never used; it worked as a threat, and the threat moved money. Over roughly three years Google and Meta struck confidential deals with publishers large and small, worth, by estimates that cannot be audited because the deals were private, somewhere between about six hundred million and one billion dollars. One concrete and documented effect: the ABC used its share to fund around sixty regional journalist positions.

Then the threat met its bluff. In 2024 Meta declined to renew, and signalled that if compelled it would remove news from its Australian services, exactly as it had done in Canada from August 2023 when faced with equivalent law. That is the confrontation, and it is unresolved as I write. The government’s answer is the News Bargaining Incentive, exposure-draft legislation that would levy two and a quarter per cent on the Australian revenue of the large platforms, above a revenue threshold, and let them reduce the charge to nothing by striking deals instead. It is intended to apply from July 2026. As of late May 2026 it is draft, not law, and it has drawn the predictable fire: around the twentieth of May, United States technology trade groups lodged submissions arguing it should be withdrawn as a discriminatory tax in breach of the free trade agreement, with the prospect of trade retaliation raised in the background. The Australian government has said it intends to proceed. The window in which this particular instrument will either hold or fail is open now, and it is narrow.

The present collision is not the first. It is the latest instance of a pattern the documented record will support and the contested record cannot be made to carry. In December 1975, during the campaign that followed the Dismissal, journalists employed by News Limited went on strike for two days over what they called bias in the company’s coverage, the first such strike by Australian journalists over editorial matters rather than pay, and seventy-five of them signed a letter to Rupert Murdoch calling The Australian a propaganda sheet. Those facts are robust, and they came from inside the building, which is what gives them their weight. The further claim, that a proprietor determined the election’s outcome, is contested interpretation, argued by some historians and rejected by others, and denied by Murdoch himself; it is not mine to assert, and I do not. The same discipline governs the rest. Kevin Rudd’s description of News Corp as a cancer on democracy is his characterisation and is reported as his. The 2021 Senate committee that found the regulatory environment weak, fragmented, and inconsistent, and recommended a judicial inquiry, was a committee majority writing over a dissent, and is reported as that. What is not interpretation but record is this: the inquiries that recommended structural reform, Finkelstein in 2012 and the Senate committee in 2021, were not acted on, while the deregulation that increased concentration was. That asymmetry is the thing to hold.

If the architecture is to be changed, four kinds of counter-architecture are on offer, and they share a difficulty that has to be admitted before any of them is described. There is no clean causal evidence that diverse ownership produces diverse content. The empirical literature that has gone looking for it describes the effects of ownership rules on content as inconclusive at best, and some studies find that consolidation did not reduce, and occasionally increased, local coverage. This does not mean concentration is harmless. It means the case against it stands more securely on structural and democratic ground, the concentration of agenda-setting power and the exposure of a whole system to one proprietor’s choices, than on a demonstrated effect on what gets printed. Keep that in mind, because the four are not equally evidenced, and it would be a cheat to lay them out as though they were.

The first class is ownership limits and concentration caps, the cross-media and within-medium ceilings whose repeal in 2017 began this story. The logic is strong and widely accepted. The direct evidence that a cap delivers the diversity it is justified by is weak. The second class is structural separation, prising content production apart from distribution, or media ownership apart from unrelated commercial interest. It is a recognised competition remedy, but the evidence specific to news pluralism is thin, and most of the precedent is borrowed by analogy from telecommunications and broadcasting carriage rather than tested in news markets. These two classes carry the strongest logic and the least proof.

The third class, public-interest journalism funding, has the most practice behind it. The Nordic press-subsidy schemes have run since the 1960s, often by funding the second newspaper in a town so that one voice does not become the only one. Such funds tend to be modest: Nordic schemes on the order of a few million dollars a year, the proposed United Kingdom innovation fund somewhat larger, France’s subsidies larger again. But notice what that evidence concerns. It is about sustaining the supply of journalism, which is a different thing from reducing concentration, and the relatively stronger case for funding cannot be quietly transferred to justify the ownership caps of the first class. The fourth class is platform and distribution regulation, and here the evidence is at once the most current and the most chastening, because Australia’s own bargaining mechanisms are the leading example of it in the world. The verdict on them is mixed and documented: money moved, the instrument was never used as designed, and one major platform left at the first opportunity. Canada’s experience, where the platform blocked news outright, is the clearest evidence of the ceiling. There is a trap built into this class that the argument of this whole essay is designed to detect. Once “money moved to publishers” becomes the measure of success, the real question goes quiet: does the transfer restore the conditions for a self-governing public, or does it subsidise the incumbent newsrooms while the architecture stands?

One comparator earns its place by doing double work. The United Kingdom has just modernised its ownership rules to capture online media, extending its public-interest test for media mergers, in 2025, to the online outlets and news magazines the old regime, built for print and broadcast, never reached. That speaks directly to the gap Australia opened in 2017. The same United Kingdom reforms, having banned foreign-state ownership of newspapers in 2024, a ban that blocked an Abu Dhabi-backed bid for the Telegraph titles, then allowed foreign-state investors a passive holding of up to fifteen per cent, on the explicit reasoning that plurality protection has to be balanced against a struggling sector’s need for capital. That tension, between guarding plurality and admitting investment, is real, and any ownership cap has to answer it. The European Union’s Media Freedom Act, which reached most of its provisions only in August 2025, is the other marker: significant, ownership-focused, and too new to have produced any effectiveness evidence at all, which is itself a caution against borrowing it as a success story. On the recurring comparison of public funding, Germany’s per-capita support for public media runs at something like four times Australia’s, on figures from around 2018 and 2019, expressed in United States dollars, and measuring public broadcasting specifically. That ratio is worth stating as an order of magnitude and no more, and the comfortable claim that better-funded public media simply produce healthier democracies is a correlation wearing the costume of a cause, which it is not.

I have a stake in this, and I had better put it on the table myself. The counter-architecture Macgregor has argued for in print, and I have supported, the Isegoria programme set out in The Mechanics of Changing the World, distributes its elements across exactly these four classes. Its horizontal break-up and its cap on the capitalisation of any news-producing company fall in the first. Its vertical separation falls in the second. Its civic endowment, a citizen-directed pledge that would fund journalism without advertising, falls in the third. Its breaking up of the social-media cartels falls in the fourth. To audit the others and exempt this one would be precisely the failure this series exists to refuse.

So I will put the sharpest element under the hardest light. The programme proposes that no company producing news and current affairs hold a capitalisation above, in the book’s own qualifier, “say, twenty million dollars”. That figure is the most falsifiable thing in the whole scheme, a specific number in the one class where the evidence is weakest, and the finding is plain: there is no clean evidence that any ownership cap delivers the diversity of content it is justified by. The case for the cap cannot rest on a claim that caps demonstrably restore that diversity, because the evidence does not support it. It rests, if it rests anywhere, on structure: that a cap reduces the agenda-setting power any single proprietor can accumulate, and reduces a democracy’s exposure to the choices of one owner. That is a real argument, and a narrower one than the book’s confidence implied. The plurality-against-investment tension the United Kingdom has just had to confront applies to my own proposal as much as to anyone’s. The strongest version of the case for the remedy I support is structural, not empirical, and the integrity of the argument requires me to say so in the same breath that I make it.

What stands, when the architecture and its history and its four possible remedies are all in view, is a question rather than a programme. The channels through which Australians learn what is at stake are concentrated at the point where news is made and concentrated again at the point where it is carried, and not one of the available counter-architectures has clear evidence of restoring the whole. That is not a counsel of despair. It is the precise shape of the problem, and naming a problem precisely is the condition of ever solving it. The structural case against this concentration is strong even where the case for any particular cure is not, and that asymmetry is not a reason to look away. It is the reason the question deserves to be held open rather than closed prematurely around whichever remedy its advocates, this author included, happen to find most congenial.

There is a reason this question comes before the questions of the ballot that this series turns to next. A vote is cast inside an information environment, and so is the argument for reforming the vote. An electorate cannot deliberate its way out of an environment it does not control. Where that control sits, and whether it can be loosened, is the prior question. It is, on the evidence, still open.

Walker Briefing is a reader-supported publication. To receive new posts and support my work, consider becoming a free or paid subscriber.