Five people were killed when a cargo jet ran off a runway in Miami in 2026. None of them were on the aircraft. There is a system built for exactly this — a bed of crushable concrete that stops an overrunning aircraft. That runway does not have one, and the reason is not that the airport was cutting costs.
Every commercial runway in the United States must end in a safety area: 1,000 feet long, 500 feet wide, flat, clear and empty. That is the whole standard. Not how fast an aircraft might arrive, not how much energy the ground must absorb — just how much emptiness there is, checkable with a tape measure and an annual inspection.
The number is old; what made it urgent was a crash. In 1999 an airliner overran a wet runway at Little Rock and eleven people died. Six years later Congress ordered every commercial airport to bring its safety areas up to standard by the end of 2015, and most of them did. Which leaves the airports that physically cannot.
Many American airports were laid out in the 1940s and 1950s, when they sat in empty fields. The fields became suburbs, and a thousand feet past the runway there is now a motorway, a river, a neighbourhood or a cliff. So in the 1990s the aviation authority funded a different answer: a bed of lightweight cellular concrete engineered to crumble under a landing gear. The wheels sink; the aircraft ploughs to a stop in hundreds of feet instead of thousands.
The first was laid at a New York airport in 1996. By 2026 there are beds at 122 runway ends at 70 airports, and they have stopped 26 aircraft carrying 497 people.
The airport had skipped nothing. Runway 30 in Miami has the full 1,000-foot safety area the rule requires, verified in the annual federal inspection. And then comes the sentence this turns on, and it is the airport's own: the bed is intended as an alternative, not as an additional layer on top of a safety area — and no airport in the country has both.
That is not an excuse. It is a description of how the rule is written.
The rule measures compliance, not risk. The safety area is a distance. The bed is a machine. The authority made them interchangeable by declaring the bed an equivalent level of safety — which is what lets an airport with no room be counted as compliant. But equivalent means substitute: adding a bed to a runway that already has the field ticks no box that exists.
And the money follows the box. Nothing forbids a compliant airport from building one anyway, but the authority pays for beds where the standard cannot be reached. So the airport that could not reach the standard is handed a machine that grabs aircraft. The airport that met it is handed a lawn.
Put beds everywhere. It fails three ways. Speed: a standard bed is designed to stop an aircraft entering at 70 knots — a figure drawn from thirteen years of data showing about nine in ten overruns happen at or below it. The jet in Miami left the runway at roughly 112. A bed would have slowed it, not stopped it.
Subscription: a bed is not a structure. Little Rock laid one in autumn 2000 and replaced it in 2018 — eighteen years, then you buy it again. And a cost that gets left out: in the Little Rock investigation the engineers who build these beds calculated that for that aircraft, on that night, it would have shed about 15 knots, still not stopped inside the safety area, and the loads would likely have collapsed two landing gear legs. A layer that protects people on the ground can injure the people in the aircraft.
Measure energy instead of distance, and a field and a bed stop being alternatives — they become quantities you can add up. Or declare a landing distance shorter than the concrete you own, and buy compliance with a timetable. Or move the road out of the line of the runway. Or write the second layer into the rule wherever there are people just past the end.
Each one is real, and each one is priced.
Since June 1999, no passenger or crew member has been killed on board an aircraft in a runway overrun in the United States. Not one, in more than a quarter of a century — and that last death was Little Rock, the crash that made the safety area urgent. It is a real achievement, and the safety area is part of it.
But look at who has died in those same years. A six-year-old boy in a car on a Chicago street, when an airliner came through the fence in 2005. Five people in Miami in 2026. The risk did not disappear. It moved past the edge of the thousand feet — and that edge is not the airport fence. One of the vans in Miami was parked well inside the airport. The rule's blind spot starts where its measurement stops.
When you are told that something meets the standard, you have been told one of two very different things. That it is safe — or that it is not required to be any safer.