Dear Prime Minister,
Within the next few weeks, you will have to make one of the most significant, and controversial, decisions of your premiership: whether to expand London’s airport capacity, and if so, whether to choose Heathrow or Gatwick. You will be faced with several thousands of pages of documents prepared by the Airports Commission which you appointed, and an even larger volume of submissions from interested parties.
Howard Davies, chair of the five-person Commission, concluded emphatically in his report published on July 1st that the right solution was to expand Heathrow. He overstates that case. The Commission has given too much weight to a model which includes an excess of detail, and makes too many assumptions about the distant future. It should have focussed on a limited number of central issues: the value of a hub, the costs of the rival options, the nature of airline competition, the value of landing slots and the cost of capital. Above all, the analysis suffers from a mechanical projection of the present into what is in reality a highly uncertain future.
Heathrow Airport is 15 miles west of central London. It has about 650 daily slots (a slot comprises one takeoff and landing), which equates to 470,000 aircraft movements, catering for 74m passengers a year. Forty-seven million passengers start or end their journeys at London and another 13.5m change planes there (and hence count twice). Heathrow became a hub airport when planes struggled to cross the Atlantic and remains a centre for “hub and spoke” operations, not only for British Airways but for the Lufthansa/United Star Alliance and the Delta/Air France/KLM Skyteam. Gatwick Airport, 25 miles south of the city, has around 350 slots, 262,000 aircraft movements and 39m passengers. Less than 5 per cent of Gatwick users are transfer passengers.
Heathrow has two runways and Gatwick one. (There are four at Frankfurt, JFK and Paris Charles de Gaulle). Heathrow has operated at full capacity for years and Gatwick is approaching it. Demand for air travel is growing, but little faster than GDP.
The Airports Commission reviewed proposals to build a third runway at Heathrow or a second at Gatwick. Your present dilemma arises, Prime Minister, because your predecessors procrastinated 50 years ago on plans to build an entirely new London airport, and at every decision point subsequently. Eventually they authorised what proved a white elephant at Stansted, today mainly used by Ryanair, the ultra-low-cost airline.
The elaborate modelling exercises undertaken for the Airports Commission are based on linear extrapolation of the present for the next 60 years. Everything will be the same, but bigger. Heathrow will continue to cater for financiers on the red-eye from New York, and salesmen departing for Shanghai; Gatwick for families in beach shorts off to Malaga. In future, more suits will want to use Heathrow and more holidaymakers will want to use Gatwick, the model projects. (In fact leisure travel predominates at both airports, but there is considerably more business and long-haul traffic at Heathrow than at Gatwick). If capacity is not increased, airlines will continue to raise prices and make large profits. The profits the model expects them to make from their London flights if there is no expansion are comparable to those the entire world airline industry has made in its history.
In calculating the benefits of the various options, the model looks at the difference between two very large numbers: the loss of these extra profits that the airlines would otherwise earn (producer surplus), and the gains to consumers from lower fares (consumer surplus). Rather oddly, the model appears to treat the loss of profit by foreign airlines that would follow an expansion of capacity as a “cost” in its assessment. That is a display of national altruism these carriers are unlikely to reciprocate.
Finally, the model reckons that because the time of the suits is more valuable than the time of the shorts, and the suits promote economic growth while the shorts spend its proceeds, developing Heathrow creates larger economic spillovers—benefits to people who are not themselves airport users.
You can safely disregard most of this convoluted analysis. It purports to describe, in immense detail, the evolution of air transport over the next 60 years, even predicting which routes airlines will choose to fly. But these are things no one can possibly know. Sixty years ago, commercial jet aircraft had just entered service. Low-cost airlines appeared in Europe only 20 years ago and even at the last UK air policy review in 2003 no one imagined that Dubai would now be the world’s busiest international airport.
But future change will be far more radical. By 2025, the earliest date at which more runway capacity might be available in the UK, it is probable that there will be commercial drones, driverless cars and fully electric vehicles. Those will transform both the economics of transport and its effects on noise and air quality, and will be precursors to further transport innovations we cannot yet imagine. In the face of such radical uncertainty, the rational course is to focus on known facts, be sceptical of grand projects, and promote developments which leave your successors a wide range of options.
The underlying demand is for air travel to and from London. Most British passengers will go to or from wherever the plane takes off or lands, and airlines will provide services to meet their demand from wherever capacity is available. British Airways has used some of the additional slots it has acquired at Heathrow to start flights to Chengdu and San Jose—and that is the connectivity everyone wants to encourage—but if travellers to these destinations are told their flights leave from Gatwick that is where they will go. If there is no expansion, larger aircraft will fly to fewer, but the most valuable, destinations. Heathrow is much more centrally located than Gatwick but, mainly for that reason, expanding Heathrow is more expensive and arouses much more opposition. Boris Johnson, and the Conservative and Labour candidates to succeed him as London’s mayor, have pledged to fight it. These are the main problems with the model.
London as hub
You made a mistake, Prime Minister, in including an objective of “maintaining the UK as an airport hub” in the Commission’s terms of reference. It is not obvious that this should be a goal at all. A businessperson travelling from New York to Moscow provides little benefit to the UK by changing planes in London. In fact the cheapest way of expanding London’s airport capacity is to encourage that traveller to transit through Amsterdam instead. Emirates now offers nine flights a day to Dubai from four regional airports, which gives a better experience to people from the North who want to visit friends and family in Australia and makes runway capacity for those planes from London to Chengdu.
It is true that the more air traffic there is to and from London the more flights and destinations there will be from London, but transfer traffic is the least valuable use of scarce airport capacity. Nor do we know the extent to which hub and spoke operation will be replaced by more direct point to point travel stimulated by greater demand and more efficient planes. The claims made for the importance of Heathrow’s role as international airline hub are weaker than you have been told.
The cost of a runway itself is less than £1bn. Land, stands, gates and terminal facilities make up most of the costs, estimated at £18bn at Heathrow and £8bn for Gatwick. The Heathrow costs are much higher partly because expansion requires that homes be demolished and a waste facility relocated (and cleaned up). The plans suppose that suits need more terminal space per head than shorts. Since the proposed runway crosses the M25 and demands new access routes, associated roadworks at Heathrow will add an extra £5bn to the total for an all-in cost of £23bn. Gatwick expenditure can be phased because terminal growth there can be piecemeal but the Heathrow plan requires a new terminal immediately the runway opens.
Current airport charges
On average, airlines pay the airport £23 for each (arriving or departing) Heathrow passenger, while the equivalent figure at Gatwick is £9. Heathrow passengers contribute another £12 per head in car parking charges, profits on retail concessions, and other revenues; those at Gatwick, only £8 or so. Operating costs (excluding depreciation) are £15 per head at Heathrow, £9 at Gatwick. Thus each Heathrow passenger makes a net £20 contribution to capital costs and each Gatwick one around £8.
The intention is that the costs of either expansion (though perhaps not the Heathrow roadworks) will be financed by the airports themselves while (somewhat curiously) the regulatory regime will allow these costs to be recovered in their entirety from all users of whichever airport is selected for expansion. So if Heathrow is selected charges are likely to rise by around 50 per cent; if Gatwick, the increase will be more in relative terms, but less in absolute terms.
The value of slots
Discussion of the market in Heathrow slots is confined to one footnote. Currently, landing slots at Heathrow are a valuable commodity, while at other London airports (including, until recently, Gatwick) they have been available on demand. Continental (now United) Airlines paid over $200m in 2008 for four Heathrow landing slots, but these were at prime times of day and the price seems excessive. British Airways bought BMI for £172m, mainly for its 42 landing slots, and slots have changed hands more recently at around £15m-£20m. These figures emphasise the attractiveness of Heathrow but nevertheless make it difficult to see how it could be worth the additional cost of £15bn (£40m per slot) to create the 350 new slots at Heathrow rather than Gatwick.
Fares and profits at London airports
As was implied by the Airport Commission model, and is evident from the slot valuations above, inadequate capacity at London airports boosts airline profits by restricting capacity and competition. Prices are determined more by the presence or absence of competitive pressure than by differences in underlying costs. The premium for slots at Heathrow seems mainly to be the result of anxiety by airlines to attract transfer passengers rather than a reflection of the greater convenience of Heathrow’s location for UK users of these airports.
The “economic rents” which airlines are extracting as a result of past policy failures, which will increase steadily until new runway capacity becomes available, should be used to defray the costs of expansion and not allowed to boost airline and airport profits. This means that your decision on expansion must be linked with carefully crafted policies about how airport charges fund future airport development. It is disappointing that neither the Airport Commission nor the Civil Aviation Authority (the economic regulator) appear to have thought this issue through.
The cost of capital
The key to an evaluation of any long term infrastructure proposal is the anticipated life of the project and the relevant cost of capital, an issue which the Commission has failed to address. London airport expansion is essentially investment in the ability to land planes near the capital, an asset which will be useful as long as air travel of the kind we know continues.
But what figure should be used for the cost of capital? There are several possibilities. First, the risk free rate at which the British government can borrow. The longest dated index-linked gilt, maturing in 2068, currently yields minus 1 per cent. Second, the “green book” rate prescribed by the UK Treasury, used in the Commission’s cost-benefit analysis, and fixed arbitrarily in 2003 at 3.5 per cent in real terms. Third, the anticipated cost of the Private Finance Initiative-type deal which it is anticipated will be employed to fund airport developments—around 5 per cent—6 per cent in nominal terms. The final possibility is to use the weighted average cost of airport capital of between 5 per cent and 6 per cent in real terms, as set by the CAA, Britain’s airport regulator.
Each of the last three plays a role in the Commission’s calculations, but there is no discussion of the appropriate choice or of the compatibility of the different assumptions made. And arguably it is the first—the risk-free long-term borrowing rate—which is most relevant.
Which rate is used makes all the difference. At minus 1 per cent over a life of 60 years, the annual cost of the proposed capital expenditure at Heathrow is around £250m. At 3.5 per cent, the figure is £800m. And at 5.5 per cent the annual cost becomes £1.35bn. If (optimistically) operating costs were maintained at £15 per passenger, a 5.5 per cent cost of capital would make the cost of providing fresh capacity at Heathrow around £55 for each of the extra 35m arriving or departing passengers, a cost which would be spread over all Heathrow passengers and which would make it the most expensive airport in the world by some distance. The equivalent figure for Gatwick is £23—still expensive, but probably worthwhile. British Airways would be the largest contributor to the costs, and would gain business opportunities but lose pricing power in any expansion—wherever it takes place but especially at Heathrow. BA opposes growth at its principal hub.
A revenue stream assured by a government agency and secured on landing charges at a London airport should not expect a return on capital substantially in excess of a risk-free borrowing rate. If the capital costs of Heathrow expansion could be substantially reduced and its actual financing costs were also trimmed, that project would merit further consideration. Otherwise, a second runway at Gatwick appears simpler, cheaper, less risky and less politically unpalatable.
This article was first published in Prospect Magazine on October 15th, 2015.