You click a link about a rent increase, a chemical spill, a council vote. Three paragraphs in, the text fades to grey and a box appears. Subscribe to continue reading. * The instinct is to be annoyed at the outlet. Hold that for a second, because the more useful question is not *why is this behind a wall* — journalism costs money, that part is easy. The useful question is **who decided that this particular story goes behind the wall, and what stayed free instead.** That decision is made by humans, it follows a logic, and once you can see the logic you can see the shape of what you are allowed to know for free. * ### The three kinds of wall * **Hard paywall.** Nothing without a subscription. You get a headline and a login box. Financial and trade publications lean this way, because their readers expense it. * **Metered paywall.** A few free articles a month, then the wall. This is the classic model — it lets casual readers in, catches habitual ones, and keeps the outlet visible in search results. * **Freemium / dynamic.** Some articles are free forever, some never are, and increasingly an algorithm decides per reader — how often you visit, where you came from, whether you look likely to pay. Two people can click the identical link and one hits a wall the other never sees. * That last one matters more than it sounds. It means "is this article paywalled" is not always a fact about the article. It is sometimes a fact about you. * ### The number that explains the whole system * The Reuters Institute's Digital News Report 2026 found that across its "pay basket" of 20 countries where paywalls are common, **17% of people pay for online news** — down a point from 18% the year before. In most markets, 10–20% appears to be a hard ceiling. Norway (40%) and Sweden (32%) are the outliers everyone cites precisely because they are outliers. * Sit with the other side of that fraction. Roughly **four out of five people** in the richest media markets on earth are reading only what is free. Whatever is outside the wall is not "some of the news." It is the news, for most people, permanently. * Meanwhile the paying side is doing fine. The New York Times reported 13.4 million digital subscribers after adding 280,000 in the second quarter of 2026, with digital subscription revenue up 16.4% to $408 million for the quarter. That is a functioning business. It is also a business whose growth does not depend on the other 80% ever reading a word of it. * ### The proof that the wall is a choice * Here is the cleanest evidence that a paywall is a commercial decision rather than a law of physics: outlets switch it off when they decide the information is genuinely urgent. * As Hurricane Sandy approached the US east coast in October 2012, The New York Times and The Wall Street Journal dropped their paywalls. After the Boston Marathon bombings in April 2013, the Boston Globe, the Times, the Journal and the Washington Post removed all or part of theirs. During Hurricane Harvey in 2017 the Times, the Post, the Journal, the Los Angeles Times and the Houston Chronicle all came down — the Times explaining it wanted its hurricane coverage read as widely as possible because of the public-safety stakes. Smaller papers do it too: the Tallahassee Democrat suspended its wall for Hurricane Ian. * And in March 2020, local outlets across the US dropped paywalls on pandemic coverage — then, according to Northwestern's Local News Initiative, gained digital subscribers anyway. * Every one of those decisions is defensible and humane. But stack them up and they say something the industry rarely says out loud: **we know how to make this free when we think you need it.** Which means that on every ordinary day, a judgement has been made that you do not need it quite that much. A zoning board, a water contract, a police budget — those are also public-safety stories, on a slower clock. * ### There is another model, and it has numbers * The Guardian has no paywall at all. It asks instead. In the year to 31 March 2026, digital revenue from recurring supporters and one-off contributors reached **£125 million, up 17%**, with digital recurring supporters up 13% to about **1.3 million** — a record, and with more than half of those supporters now outside the UK. * We are not going to pretend that settles the argument. £125 million funds a serious newsroom, but it is a fraction of what subscription-first publishers pull in, and the model works partly because a large majority of readers pay nothing while a devoted minority pays for them. It also depends on a structure most outlets do not have — the Scott Trust — which removes the pressure to return a dividend to anybody. * The point is narrower and harder to dodge: an outlet can be well-funded, ad-supported, globally read, and free. So "we had to put up a wall" is a description of one strategy, not a description of gravity. * ### What actually ends up on each side * This is where it gets uncomfortable, and where you should watch for yourself rather than take our word. * The expensive work — the months-long investigation, the FOIA fight, the reporter who sat in a courtroom for three weeks — is the work most likely to sit behind the wall, because it is the work that converts casual readers into subscribers. The cheap work — the aggregated write-up of someone else's scoop, the opinion column, the reaction piece, the story assembled from a press release — is the work most likely to stay free, because it costs little and earns advertising impressions. * So the free tier of the internet skews toward commentary and rewriting, and the paid tier skews toward original reporting. Not by conspiracy. By arithmetic. * The consequence is a two-tier information system, and it does not divide along the lines people assume. It is not smart versus stupid, or engaged versus apathetic. It is roughly **£10-a-month versus not**, applied to the question of who gets to know what their own government did. * ### Getting to the reporting without pretending the money isn't real * Four legitimate routes, none of which involve stealing anything. * **Your library card.** Many public libraries buy digital access to major newspapers and hand it to cardholders for free. This is the single most underused piece of public infrastructure in the news economy, and it exists specifically so that ability to pay does not decide who is informed. * **The wire copy.** If the story originated with a wire service, the same reporting is often running free on a dozen smaller sites that licensed it. The byline changes; the facts do not. * **The outlet's own newsletter.** Publishers routinely give away in email what they charge for on the site, because an email address is worth more to them than one pageview. * **The primary document.** Court filings, company results, regulator rulings and government reports are usually free and public. The article is the interpretation. The filing is the thing. * And if you can afford to pay one outlet, pay one. Reporting that nobody funds does not become free — it becomes press releases with bylines, which is the most expensive news of all. * ### The question to carry around * Next time a wall drops in front of you, do not just ask whether the article is worth the money. Ask what that outlet is giving you free on the same day. * If the accountability reporting is behind the wall and the outrage is in front of it, you have learned something about that newsroom's model — and something about why the public conversation sounds the way it does.