A kill fee is the money a publication pays a freelance writer for a story it commissioned, accepted delivery of, and then decided never to print. The work exists. The reporting happened. The phone calls got made, the documents got read, the sources got talked into saying something on the record. And then an editor writes back to say the piece is not going to run, and instead of the agreed fee, a fraction of it arrives. The standard fraction is 25 percent. Freelance contract guidance from The Open Notebook, a nonprofit resource for science journalists, describes the kill fee as a percentage of the contracted rate, with 25 percent common and stronger contracts pushing higher; some writers negotiate up to 50 percent. Which means the ordinary outcome of a killed story is that the person who did all of the work receives one quarter of the money, and the outlet receives a piece it never has to explain. Here is the part that matters for you as a reader: the kill fee is the mechanism by which stories disappear quietly. Not censorship in the movie sense. Nobody calls a newsroom and demands a spike. A story simply becomes inconvenient, or the editor who championed it leaves, or the section gets restructured, or legal gets nervous, or the news peg passes — and the piece goes into a drawer with a 25 percent receipt attached. No one publishes a list of what got killed. There is no correction, because nothing was printed to correct. * The clause exists for a real reason, and it is worth being fair about it. Commissioning is a gamble. An editor assigns a piece on a premise that might not survive the reporting. Sometimes a story is killed because the reporting did not hold up, which is exactly what you want an editor to do. Sometimes a writer turns in something unusable and three rounds of edits do not fix it. A publication that had to print every piece it commissioned would commission almost nothing, and the people hurt most by that would be freelancers. But notice the asymmetry in who carries the risk. The outlet risks a quarter of the fee. The freelancer risks the rent. And in most contracts the kill fee is not the end of the negotiation — it is the start of a second one, over rights. Some contracts return all rights to the writer on kill, which means the piece can be sold elsewhere and the work is only partly wasted. Others keep the outlet's exclusivity for a period, or claim the rights outright, which means the writer has been paid 25 percent for a story that now cannot be published anywhere. Read that clause before you read the percentage. It is usually worth more. * The clearest recent illustration of a story that did not die when it was killed involves a staff reporter rather than a freelancer, which is instructive in itself. Ronan Farrow began reporting allegations against Harvey Weinstein while at NBC News in 2016. NBC declined to air it. Farrow took the reporting to The New Yorker, where editor David Remnick published it on October 10, 2017 — five days after Jodi Kantor and Megan Twohey broke the story at The New York Times. In 2018, Farrow shared the Pulitzer Prize for Public Service with Kantor and Twohey. That case is famous precisely because it is the exception. Farrow had a name, a platform willing to take the piece, and enough leverage to walk. Strip those away and you get the ordinary version: a freelancer with no institutional backing, a killed story, a 25 percent payment, and a rights clause that turns reselling it into a legal question rather than a phone call. The stories that survive being killed are the ones attached to someone who could afford to fight. You do not have a list of the others. Neither does anyone else. * The word "kill" is doing a lot of work here, and journalism uses it loosely, so it is worth separating three things that get confused. A killed freelance piece is a contractual event with a payment attached. A spiked story is an internal editorial decision about staff work — no fee, just a piece that never runs. And "catch and kill" is something else entirely: buying exclusive rights to a story specifically in order to bury it, which is a suppression tactic wearing a publishing contract as a costume. The vocabulary overlaps. The ethics do not. * For years the whole arrangement ran on trust, which is a polite way of saying it ran on the freelancer having no remedy. That has changed slowly, and only in some places. New York City passed the Freelance Isn't Free Act — signed by Mayor Bill de Blasio on November 16, 2016 and in effect from May 15, 2017 — one of the first laws in the United States to give freelance workers an actual enforcement route rather than a moral argument. New York State followed. The state version added Article 44-A to the General Business Law and took effect on August 28, 2024. It requires a written contract for any freelance engagement worth $800 or more, itemizing the services, the rate and method of compensation, and the payment date. Where no date is specified, payment is due within 30 days of the work being completed. A freelancer who is not paid can recover the amount owed plus double damages, attorneys' fees and costs. If the state Attorney General brings a civil action, the hiring party faces penalties of $1,000 for a first violation, $2,000 for a second and $3,000 for a third or subsequent one, rising to as much as $25,000 where there is a pattern or practice of violations. Claims can be brought within six years of the conduct. Read those numbers next to the 25 percent and you can see what the law actually fixed and what it did not. The statute forces a written contract and forces the money to arrive on time. It does not require anyone to publish anything, and it sets no floor on what a kill fee has to be. The right to be paid promptly for a killed story is a real improvement over the right to send a fourth polite follow-up email. It is not the same as the right to have the story printed. * So what does any of this change about how you read? Three things. First, absence is not evidence of nothing. When a subject you would expect a publication to cover simply never gets covered, the honest answer is usually boring — nobody assigned it. But sometimes the answer is that it was assigned, delivered and shelved, and you will not be told which. Second, when a freelancer publishes something in a smaller or stranger outlet than the quality of the reporting would predict, that is often a killed piece finding a second home. The mismatch between the ambition of the work and the size of the venue is a signal worth noticing rather than dismissing. Third, the economics are the ethics. A newsroom that assigns widely on 25 percent downside is buying option value cheaply, and the people funding those options are the least secure workers in the industry. When you read a byline with no staff title under it, you are reading someone who took a risk the institution declined to take. * None of this means killed stories are suppressed stories. Most of them are simply stories that did not work, and an editor killing a piece that does not hold up is doing the job correctly. But the system has no daylight in it. The outlet knows what it killed. The writer knows. You are the only party to the transaction who is never told it happened — and you paid for the outcome either way, in a subscription, in an ad impression, or in the story you never got to read.