IMAGE: L. LEDUC/GETTY IMAGES
THE POP-UP arrives without warning. One moment a reader is scanning an article about a school-board meeting; the next, the text disappears behind a grey box offering a monthly subscription. For the newspaper’s digital audience, this is the paywall. What happens next, across more than 55m such events between 2020 and 2024, offers the clearest picture yet of what online news consumers truly value, and whether those values can keep a newspaper in business.
Gregory Martin, Shoshana Vasserman and Cameron Pfiffer, three economists, obtained the complete digital traffic and subscription records of a large metropolitan daily in the United States. Their working paper, published by the National Bureau of Economic Research, tracks 605m article visits and every subscription purchase over four years. The newspaper uses a meter: non-subscribers may read a small number of articles (two to four) in any rolling 30- or 60-day period before the paywall blocks further access. Because the remaining allowance is never displayed, and the rules change unpredictably, a reader cannot easily guess when the wall will appear. For the researchers, this quasi-random arrival provides a way to isolate willingness to pay.
The results upend a common assumption among media executives. If one measures reader interest by clicks alone, the most popular articles are syndicated advice columns and entertainment pieces. But when a reader must reach for a credit card, the calculus shifts. Articles about public health, local politics and the local economy generate subscription rates double those of entertainment coverage. Health stories proved especially powerful during the covid-19 pandemic, which fell inside the sample period. In-house reporting consistently outperforms wire-service copy on both traffic and conversion. Yet the correlation between what drives visits and what drives subscriptions is weak: a gap that has implications for how a newsroom should be staffed.
The authors group the newspaper’s writers into eight beats based on the topics they cover: Sports, Entertainment, Local News, Health, Business, Local Events, Editorial and Crime. Using the paywall variation, they estimate how much readers are willing to pay for the average article from each beat. A Local News piece—covering schools, zoning or city hall—commands roughly twice the subscription value of an Entertainment article. Under a hypothetical revenue model that maximises subscription income, the newspaper would shift staff from soft news to hard news beats. Under a model that maximises traffic, the opposite would happen. An à la carte pricing system, where readers pay per article, would produce a similar allocation to subscriptions but generate less total revenue than the existing all-access monthly plan. The readers who subscribe value the bundle, and not just single stories.
This finding might seem to vindicate the subscription model. If readers are willing to pay for accountability journalism, then the market could in principle support it. But the numbers suggest otherwise. Even the highest-value beats—Local News, Health, Business—do not generate enough subscription revenue to cover the salaries of the journalists who write for them. At current staffing levels, every section runs a net loss from digital subscriptions. Adding more reporters would push marginal net revenue further into negative territory. The old cross-subsidy, in which classified ads and display advertising paid for investigative reporting, has evaporated. Digital advertising, though still present, does not fill the gap.
The paper quantifies the shortfall directly. Using a method that identifies investigative articles by their influence on subsequent coverage and their similarity to prize-winning journalism, the researchers estimate that roughly 1.3% of the newspaper’s output qualifies as investigative. To expand production of such pieces without incurring a net revenue loss would require an annual subsidy of about $11,000 per investigative article.
None of this is abstract. The newspaper in the study is typical of many legacy dailies: once a local monopoly in print, now owned by a private-equity-controlled holding company, with digital subscriptions accounting for about 40% of its subscriber base. Its experience mirrors an industry-wide collapse. Aggregate weekday circulation of American dailies fell from 44.4m in 2011 to 24.3m in 2020; advertising revenue dropped by more than $26bn over the same period. The internet unbundled news, and readers who want only sports or celebrity gossip can now get them elsewhere, for free.
Walter Lippmann saw the problem a century ago. “Nobody thinks for a moment that he ought to pay for his newspaper,” he wrote in 1922. “He expects the fountains of truth to bubble, but he enters into no contract, legal or moral, involving any risk, cost, or trouble to himself.” The difference today is that the advertising fountain has run dry. Readers will pay for local politics, it turns out—just not enough to keep the reporter on the beat. The subsidy question, once academic, now appears on every publisher’s spreadsheet. ■
