The Garden Hose
Both trace how physical infrastructure and scarce capacity shape networks people rely on.
For months, an airline flew an empty aircraft from Heathrow to Cardiff six times a week. It was the rational thing to do. What it was protecting was permission to use a runway. A Heathrow-commissioned study estimated that scarcity added over £200 to an average long-haul return before the pandemic.
In March 2007, British Mediterranean Airways was reported to have been flying from London Heathrow to Cardiff six times a week since the previous October. No seats were sold on it. None were ever offered. The aircraft took off from one of the busiest two-runway airports in the world, flew to Wales carrying nobody, and came back.
The airline was not confused. It had been flying to Uzbekistan, the Uzbek market had collapsed, and the take-off and landing times it had used for that route were now sitting idle. Under the rules, times you do not use are times you lose. So BMed flew to Cardiff for months to hold on to something. Its chief executive, David Richardson, explained it without embarrassment: "The Uzbek market collapsed, but we knew we would want to use those timings again this summer. It wasn't ideal, but we wanted to keep hold of it."
In February 2007, while the empty flights were still running, BMI bought the whole of British Mediterranean Airways for £30 million. The deal brought the airline and its route network into BMI. A related transfer sent 51 Heathrow slots to British Airways.
For £30 million.
The same headline sum attached to the acquisition and the slot transfer. It was not quite an airline for nothing: BMI's chairman Michael Bishop explained that BA had pre-existing rights over the slots and that their transfer was a condition of the deal. But permission to use a runway at particular times of day carried a price comparable to the reported price of the airline itself. This piece is about what that permission actually is, why nobody owns it, how it came to command a price comparable to the company holding it, and who pays. The answer to the last one is you, through a premium researchers can estimate, on a line that appears on no ticket.
Twice a year, somewhere in the world, about 1,300 people check into a conference centre to divide up the sky.
The IATA Slot Conference has been running since 1948. The 158th was held in Bangkok in June 2026; the 159th, shaping schedules for 2027, is scheduled for Budapest on 17–19 November 2026. Representatives of more than 250 airlines and the coordinators of over 300 capacity-constrained airports sit at tables and negotiate, face to face, over who lands where and when. It is one of the largest events the airline industry stages, and it is, in effect, a long-running marketplace in aviation, for a commodity that the law is adamant is not a commodity.
What is being negotiated is a slot: a permission. Under the regulation that governs allocation in Britain - Council Regulation (EEC) No 95/93, inherited into UK law after Brexit and amended several times since - a slot is "the permission given by a coordinator... to use the full range of airport infrastructure necessary to operate an air service at a coordinated airport on a specific date and time for the purpose of landing or take-off". The Department for Transport glosses it identically in its own 2024 consultation: permission to use the runway, the terminal, the gates, at a named moment.
Permission. Not a lease, not a title, not a right in land. The word is doing deliberate work, because the holder has conditional access to infrastructure, not ownership of it. American regulators are blunter: the Federal Aviation Administration describes a slot as an operating privilege subject to its control rather than a property right.
In Britain the permissions are issued by Airport Coordination Limited, a company registered in Staines in April 1991, limited by guarantee and with no share capital, which describes itself as a not-for-profit body founded on independence and non-discrimination. It coordinates around eighty airports worldwide. At Heathrow, Gatwick, Stansted, Luton and City - all Level 3, meaning demand so exceeds capacity that no aircraft may move without an allocated slot - ACL is the gate.
And the rules it applies look, on paper, like a competitive system. Unused slots go into a pool. Half of everything in that pool must be offered first to new entrants, so that incumbents cannot pull up the ladder behind them. The baseline for historic rights is the 80:20 rule, known in the trade as use it or lose it: fly a series of slots at least 80% of the time across a season and you keep it next season, fall below without qualifying relief and the series returns to the pool. Exceptions matter: Parliament considered temporary provisions in June 2026 allowing airlines to hand back up to 10% of slots for the summer and winter 2026 seasons without losing historic rights, in response to disruption from the Middle East conflict.
None of this was designed to create an asset. Grandfather rights - the principle that if you flew it last year you may fly it this year - began as scheduling housekeeping among a handful of flag carriers in an era when there was no shortage worth arguing about. It was a convention for avoiding collisions in a timetable, not a settlement of property.
But conventions calcify, and the shortage arrived.
Take the system at its word and ask how much capacity it actually distributes.
More than 99% of slots at Heathrow were allocated on historic right in the Department for Transport's consultation figures. Across all UK Level 3 airports it is around 92%, at Gatwick 98%, at Heathrow above ninety-nine.
Which leaves the pool. In the summer 2022 season, out of 152,613 slots at Heathrow, the number that reached the pool was sixteen. Twelve went to new entrants. The 50% new-entrant rule was honoured in full, and it distributed twelve slots across an airport running more than 1,300 flights a day.
The front door
152,613
Summer slots available
16
Allocated through the pool
12
Allocated to new entrants
The 80:20 rule was meant to be what refills that pool, and BMed's empty Cardiff service shows why that mechanism can fail. Eighty per cent is not a demanding threshold when the thing being defended is worth more than the aeroplane defending it. Flying nobody to Wales can be rational behaviour. The operating cost becomes an insurance premium against losing the slots.
So the front door allocates almost nothing, and access has to come from somewhere else. It comes from incumbents selling to each other, through a legal mechanism framed as an exchange rather than an outright sale.
What they can do, under Article 8a of the regulation, is exchange one slot for another. The market was built inside that gap, and the device is called an artificial swap. An airline that wants a valuable early-morning arrival first acquires something worthless, typically a late-evening slot nobody wants, and then formally exchanges the worthless one for the valuable one. The coordinator processes the exchange; a separate payment settles the difference in value. The less valuable slot can subsequently be returned to the pool.
This is not a loophole anyone is getting away with. It has been described from the bench.
Monarch Airlines collapsed in October 2017 with no aircraft in service and no passengers. ACL took the view that a carrier which had stopped flying was no longer entitled to slots and moved to return them to the pool. Monarch's administrators went to court, because the slots were the most valuable thing left in the estate, and they won. Floyd, Newey and Asplin LJJ held that Monarch remained an air carrier and kept its allocation, and the judgment set out the operating principle of the entire market in one sentence: "while a slot cannot be the subject of an outright sale, it continues to be possible to exchange slots on the basis that the recipient of the more valuable slots will make a payment".
A dead airline, with nothing to fly and nowhere to fly it, successfully asserted a claim on permission to land aircraft it did not own, so that its administrators could convert that permission into cash for creditors. The court was told in terms that they intended to swap the slots for much less valuable ones and take a payment reflecting the difference, and found nothing improper in it.
Therefore the honest description is not that Britain allocates runway capacity at Heathrow. Britain allocated it once, to whoever happened to be flying, and has been watching it change hands privately ever since.
To see what that capacity is worth, you need the moment the market discovered its own price, and that moment has a date: 30 March 2008.
Until then, access to Heathrow from the United States was not a market at all. It was a treaty. Bermuda II, signed on 23 July 1977, initially restricted the designated British and American Heathrow-US operators to British Airways, Pan Am and TWA. United and American later replaced Pan Am and TWA, and Virgin Atlantic gained access in 1991. Some other countries' carriers also held onward traffic rights. Access remained tightly restricted by agreement between two governments until 2008.
Bermuda II expired on 30 March 2008, replaced by the EU-US Open Skies agreement. Suddenly any qualifying carrier could fly Heathrow to America, provided it could get in. Continental Airlines got in by buying four daily slot pairs for a reported $209 million, a figure recorded in the European Commission's impact assessment. Roughly $52 million per daily pair, in 2008 money, for permission that had been handed to the incumbents for nothing.
That is one of the clearest documented benchmarks in this market, and it is not the number everybody quotes.
The number everybody quotes is $75 million, attached to a 05:30 arrival Oman Air acquired in February 2016. Follow it through the record and it comes apart in three stages. On 16 February 2016, ch-aviation reported that Air France-KLM had sold two Heathrow pairings, one of them to Oman Air at $75 million. Nine days later, Business Traveller reported that the seller had not been Air France-KLM at all: "News has emerged that it was Kenya Airways who sold Oman Air its prized early morning slot at London Heathrow. Initially it was reported that Oman Air acquired the slot from the Air France/KLM Group." Then Aviation Week reported the shape of the transaction: Oman Air was taking two Boeing 787-8 Dreamliners from Kenya Airways on a three-year wet lease alongside the slot, as Kenya Airways scrambled for cash.
So the most cited figure in British aviation was reported against the wrong counterparty, circulated without a published contract or price breakdown, alongside reports of a wider agreement involving two widebody aircraft. That does not establish that the aircraft leases were included in the $75 million; it does make the figure harder to interpret as a clean comparison.
That confusion is what an opaque market makes possible. ACL records who holds a slot and who operates it, and does not record, regulate or publish what was paid. There is no comprehensive public price register or requirement to advertise availability and cost. The DfT's own verdict is unusually undiplomatic: with no requirement to advertise either availability or cost, the secondary market "lacks the basic features of a transparent modern marketplace". The department also notes why capacity sits unused in plain sight - airlines "prefer to hold on to even 'loss-making' slots rather than selling them, in order to hinder competitors from obtaining slots".
The volumes are tiny. In summer 2019, 2.4% of Heathrow slots traded. At Gatwick, 0.8%. This is not a liquid market; it is a rare and private one, which is why a single disputed transaction from 2016 has been doing the work of a price index for a decade.
Where the value stops being deniable is on balance sheets, and there it appears only in half. Under international accounting standards an airline that buys a slot capitalises it at what it paid, and slots judged to have indefinite useful lives are not amortised, but are tested annually for impairment. An airline handed a slot for nothing has no cost to capitalise, so the original allocation does not generate a purchase price to record. Purchased slots therefore show up at tens or hundreds of millions. Inherited slots show up at zero. British Airways holds roughly half the slots at Heathrow, but the carrying value of purchased slots does not measure the market value of its entire holding.
Lenders are not fooled by the accounting. In November 2025, Apollo-managed funds and affiliates completed a $745 million senior secured financing backed by Virgin Atlantic's Heathrow slot portfolio. Etihad paid Jet Airways $70 million for three Heathrow pairs in 2013 and leased them straight back, using a sale-and-leaseback arrangement to release cash while retaining access. Banks have no difficulty valuing permissions that are not ownership of the runway.
And the bill lands somewhere specific.
In June 2022, Frontier Economics reported to Heathrow Airport on slot scarcity and ticket prices. Its modelling estimated a congestion premium on Heathrow fares: around 20% before the pandemic, worth over £30 a head on a short-haul return and over £200 a head on a long-haul return. Across point-to-point passengers, roughly £2 billion a year before the pandemic. These are modelled averages, not a fixed surcharge on each ticket. Heathrow commissioned the work and had an interest in showing that airlines capture scarcity rents; the estimate should be read with that incentive in view.
The estimated rent generated by constrained runway access runs into billions. It accrues through airline fares, rather than through payments to the airport for slot transfers. Heathrow receives nothing when a slot changes hands: not a fee, not a percentage, not a levy. Its airport charges are capped by the Civil Aviation Authority under a regulated asset base model. The concrete is regulated. The permission to use the concrete is not.
One qualification, because the strong version of this claim is wrong. The state is not entirely absent. Ordinary taxes can apply to airline profits and slot transactions. What is absent is a dedicated public charge capturing the scarcity value when the permission changes hands.
It is tempting to read all this as enforcement failure: a rule saying slots are not property, defeated by clever lawyers.
That is not what happened. Nobody ever decided to create this asset, which is exactly why nobody ever decided who should own it.
The scarcity at Heathrow is not geological. It is a planning condition. The airport is permitted 480,000 air transport movements a year, a ceiling attached to the Terminal 5 consent and set to limit noise over west London rather than to reflect what the airspace could absorb. Before the pandemic it was running at about 99% of that. The cap is a decision taken on behalf of people living under the flight path, and on its own terms a perfectly defensible one.
Now look at where the most valuable permissions in Europe actually sit.
Heathrow's night quota period runs from 23:30 to 06:00, and the airport is limited to 5,800 night flights a year. Around 80% of Heathrow's night flights fall between 04:30 and 06:00 - an average of sixteen aircraft a day. The morning bank, the transatlantic and long-haul arrivals that command some of the highest prices in this market, lands inside the window the noise rules exist to suppress. Kenya Airways' slot was a 05:30 arrival. Oman Air wanted it precisely because it is at 05:30.
The morning bank
Ninety minutes inside the night.
The restriction and the value are the same fact. A cap on a thing everybody wants creates a rent, and the rent is not created by the airline holding the slot. It is sustained by capacity constraints, including the refusal to permit a 480,001st flight. That refusal was made by the state, in public, for public reasons, and the windfall it generated was distributed by asking one question: who was flying here already?
If you want to know whether the government understands this, look at what it did in May 2022. After closing UK airspace to Russian carriers, ministers discovered that Aeroflot, Ural Airlines and Rossiya still held slots they could not use but could still cash in. So the government legislated to stop them selling. The Foreign Secretary, Liz Truss, put it plainly: "We've already closed our airspace to Russian airlines. Today we're making sure they can't cash in their lucrative landing slots at our airports." The Transport Secretary, Grant Shapps, priced the intervention: it would prevent Russia "cashing in on up to £50 million".
The British state passed a sanction to prevent the sale of a thing that, in British law, cannot be sold, and put a number on it in the press release. ACL subsequently announced that the affected carriers' winter 2022 historic slots would return to the pool for reallocation.
Therefore the question of who holds these permissions stops being administrative, because for some countries the slot is the connection.
Kenya Airways sold its valuable early-morning Heathrow slot. That is the detail the $75 million story usually loses. A national flag carrier's prime arrival time at a major long-haul destination was converted into liquidity during a cash crisis, and by 2024 the Kenyan press was reporting that the sale had damaged the airline's competitiveness on the route. It could not count on recovering an equivalent morning slot from a nearly empty pool.
In August 2024, Romania's state-owned TAROM sold its Heathrow slots to Qatar Airways. The daily Bucharest-Heathrow service, flown by an Airbus A318, ended with the switch to the winter timetable on 27 October 2024. Qatar put a Boeing 777-300ER into the same times on its Doha route. Romania's national carrier had been in restructuring, with state aid approved by the European Commission on conditions that included cutting routes. The country's flag-carrier link to London became a Doha flight, and the proceeds went into an airline's restructuring.
By October 2024 South African Airways was reported to be weighing the same move on slots it had been leasing to Qatar Airways and British Airways, to raise capital. An SAA representative offered the reasoning that every distressed carrier eventually reaches: "Even during our peak operations to London, we were only utilising one of these landing slots."
Each of those decisions is defensible on its own terms. Together they describe something odd: a country's scheduled access to the British capital is an asset on a private balance sheet, it is worth more to a solvent foreign carrier than to the airline holding it, and it moves in the direction of capital. The treaty restrictions have been replaced by an auction with no auctioneer, in which historic allocations were entrenched by the 1993 regulation and the bidding is private.
This is also why expansion is a more complicated proposition than it looks. A third runway would lift the cap from 480,000 movements to around 756,000, and capacity from roughly 84 million passengers to 150 million. In November 2025 the government backed Heathrow's own scheme over a rival from the Arora Group, at about £33 billion for the runway and £49 billion for the wider project, with a development consent decision targeted before the end of this parliament and the runway perhaps operating in 2035. The government launched consultation on a draft Heathrow Expansion National Policy Statement in June 2026. The CAA has since confirmed recovery of early expansion costs up to £320 million in 2024 prices, subject to safeguards.
Adding a quarter of a million movements a year to a market where sixteen slots reached the pool in a season does not merely reduce scarcity. It could erode the premium that scarcity sustains, putting pressure on balance sheet values, collateral, and the barrier that keeps challengers out - while the incumbents also pay the regulated charges that fund the concrete. Their enthusiasm for more Heathrow has been notably cooler than one might expect from businesses being offered room to grow.
The reform proposals have the same shape. The DfT's consultation canvassed tightening the usage rule from 80:20 to 90:10, building a slot register with a mandated trading platform, and restricting what can be done with newly allocated slots. It is a serious document that names the problem accurately. Its treatment of new capacity does not amount to auctioning away the existing historic allocations. In June 2026, Parliament was also debating slot-reform powers in the Civil Aviation Bill, with ministers defending the predictability of historic rights. Reform remains politically live; proposals and enabling powers are not a completed redistribution.
Auctioning has been the economists' answer since the DotEcon study for the European Commission in 2001, and has never been implemented in Europe. The objections are substantive rather than merely self-interested: fleets and networks have been planned for decades on the assumption of historic precedence, confiscation would land on balance sheets banks have lent against, and a bidding war for London could be won by the deepest state-backed pocket rather than the best operator. The Commission's 2011 attempt at reform died in the Council. The rule survives because everyone with standing to object has a reason to, and passengers paying the estimated premium have no slots to bargain with.
On a typical morning, between half past four and six, Heathrow says around sixteen aircraft arrive over west London. They are flying in the hours the noise rules were written to protect, and their arrival times are among the most valuable in British aviation. Somewhere in those fares is the scarcity premium: the Heathrow-commissioned study estimated more than £200 on an average long-haul return before the pandemic, not a fixed charge paid by everyone aboard.
British Mediterranean Airways understood this in 2006 and flew an empty aeroplane to Cardiff six times a week rather than let the times lapse. It was right. The related slot transfer carried the same £30 million headline figure as the acquisition.
The aircraft belongs to someone. The airport belongs to someone. The ninety seconds belongs to nobody.
The runway will not see a penny of it.
You’ve looked beneath the surface.
Both trace how physical infrastructure and scarce capacity shape networks people rely on.
A connection through “The hidden bottleneck”: Find the physical and institutional constraints beneath apparently limitless systems.
A connection through “The hidden bottleneck”: Find the physical and institutional constraints beneath apparently limitless systems.