The situation is even less simple than you think

Writing on the CNE Competition Blog, my co-Samizdatista Brian Micklethwait responded to my post on the present anti-trust investigation of British Airways, by in response to a mention on my part of “landing fees”, bringing up the question of whether there should be a free market in airport landing slots. Brian clearly has in mind something like an airport charging a fee for aircraft to land that is driven by demand, and setting it at a point such that the supply and demand curves meet, and thus allocating airport capacity using market pricing signals.

While I agree completely that this would be the optimal way of allocating airport capacity, things are never going to work like this. In aviation industry terms, I didn’t mention the question of landing slots at all. I mentioned “landing charges”. Slots are seen as something else. Even when people talk about a free market in landing slots, they still aren’t talking about landing charges specifically. Landing rights, landing slots, and landing fees are all separate from one another, and it wouldn’t occur to many or most people in the aviation industry that they are connected. The situation is an awful mess, and if it is ever going to be untangled, it is necessary to understand it.

For this reason, I am going to attempt to explain it. This is going to be rather nerdy. If Transport Blog still existed, it would be a fine post for that blog. However, it does not, so I will do it here. In order to be allowed to fly a particular route, the first thing an airline must have is “route authority”, which is the legal right to fly a particular route. For an international route, this is normally granted by a treaty signed between the two countries involved. (In aviation speak, such a treaty is normally referred to as a “bilateral”). Such a treaty will normally specify the number of airlines that can fly between the two countries, and which cities or airports they can fly between. It will also often state how many flights per week are allowed and it might even name the airlines. A treaty that allows any airline from either country to fly as often as it likes is called an “open skies” policy. These exist in some markets (eg between the US and Germany) but not in most. For a domestic route (and all flights within the EU are treated this way) an airline must be granted a licence to fly the route. In many markets airlines are licensed for specific routes (and this used to be the usual arrangement) but in deregulated markets (also often referred to as “open skies”) any airline can fly any domestic route. (“Any airline” normally excludes foreign airlines. Any US airline can fly essentially any domestic route, and any EU airline can fly any EU route, but US airlines cannot usually fly routes within the EU, and European airlines can never fly domestic routes within the US).

The EU has been attempting to take over the right to negotiate bilateral aviation treaties from individual countries for a while now, based on the (I think sensible) argument that if the EU is a single market for internal aviation it should also be one for external aviation. They have reached an agreement in principle with the US that under a new “Open Skies” treaty, any US or European airline will be able to fly from anywhere in the EU to anywhere in the US. (Airlines from all over the US and EU would have to be treated equally with respect to the allocation of slots at airports, too). One consequence of this would be the end of Bermuda 2 and an end to the restriction on the number of US or UK airlines that can fly between the US and Heathrow. Another would be that French and German airlines could start flying from Heathrow to the US in addition to US airlines. At least, they could if they could get the slots at the airport, which is another story.

For once an airline has legal route authority, an airline must then obtain slots at the airports it intends to fly between. A slot is the right to land or take off at an airport, but these are granted to allow flights at the same time on a regular basis, usually at the same time once a week. (Therefore, to have a daily flight an airline would require seven take-off and seven landing slots). If an airport can handle 40 flights per hour and only 20 aircraft want to land, then slot allocation is not necessary. Most of the world’s airports do not have to worry about slot allocation, or only have to worry about it at peak hours. However, most of the busiest airports do. Heathrow is just about the most congested airport in the world, so it certainly does.

Grandfather rights usually apply to slots. If an airline has a slot to use this week, it will have the right to use it again next week, and this right will continue ad infinitum. In effect, a slot is the right to use an airport at a particular time, in perpetuity/ Therefore, airlines that have been operating at a particular airport for a long time are allowed to continue to operate there. New slots become available when existing airlines stop flying routes, and when airports are expanded or made more efficient so as to add more flights. When new slots become available, the allocation process is usually non-financial. Airlines with route authority are allocated a share of available slots based upon how long they have been waiting, how many flights they operate already, how large are the aircraft they intend to operate, and that kind of thing. . Basically, though it is allocation by queuing. It is possible for a new airline to obtain a large allocation of slots at a busy airport like Heathrow – it just takes time. (For instance, Virgin Atlantic have been slowly accumulating slots since commencing flights from Heathrow in 1991. They now have a lot of slots, but it has taken them 15 years to reach this position).

Slots are essentially intangible assets, and they are treated as such in many company accounts. Obviously slots at a very busy airport such as Heathrow are very valuable, but there is no direct market in slots, as they are not normally allowed to be bought or sold. Airlines can generally change the routes for which specific slots are used, but airlines that go bust, or cease using Heathrow, or reduce their number of flights there are supposed to give up the slots so that they can be allocated to other airlines through the usually allocation system (ie queuing). Of course, ways have been found to buy and sell slots regardless, through strategies like a larger airline buying a smaller airline, and then immediately disposing of all its assets except for the slots. However this is not very efficient.

All of this means that incumbent airlines continue to use Heathrow. Airlines that have been flying to Heathrow for 40 years are able to continue doing so. Lots of national carriers of obscure African nations fly to Heathrow, for instance. They gain little commercial advantage from doing so – if they moved their flights to Gatwick or Stansted their passengers would come with them – but they continue doing so because they and their passengers marginally prefer Heathrow, and the move would cost money. Their slots would be worth far more to major airlines in competitive markets, but they can’t easily sell them, and many of the airlines that might want to buy them do not have route authority anyway.

One potential solution to this is to simply make slots a tradable commodity, and establish a formal market in them. New slots could be allocated through auction, and airlines would be free to buy and sell slots at any price they could agree on. This would pretty rapidly ensure that the airlines flying to and from Heathrow would be those that could gain the greatest commercial advantage from doing so. Smaller airlines would be able to sell their slots, make money from doing so, and then continue their operations elsewhere. The slots would end up in the hands of those for to who they are the most valuable. New entrants at Heathrow would be able to obtain access to the airport by buying slots off existing owners or the airport itself. By being able to auction newly available slots, the airport owners would receive some price signals which would encourage them to run the airport efficiently.

This would be much better than the status quo, but even this arrangement is mighty peculiar when you think about it. It isn’t the type of arrangement that was described at the start of this post, in which supply is simply rationed through charges to use the airport. Why should airlines own the right to use landing slots in perpetuity? Once an airline has route authority and slots, it does have to pay to use the airport, a so called “landing charge”, which is tiny compared to the value of the slot itself, and is merely designed to reflect the operating cost to the airport of providing the services necessary for an airline to land. In the case of London’s three main airports (all owned by BAA) these charges are set by a government regulator, and they are required to be about the same for each airport.

If we could raise landing charges to the point at which the number of airlines that want to use Heathrow at rush hour is precisely equal to the number of landing slots available, we would end up with a situation where the cost of landing at Stansted in the middle of the day would end up close to the marginal cost of providing airport services to the airline, and the cost of landing at Heathrow at 7am would be very much higher. This would send much clearer price signals to the owner of the airport than would occasional auctions of the right to use the airport in perpetuityu, and the owner would then have proper incentives to increase efficiency and provide better service at the times when their revenues are higher.

Deregulating landing charges and allowing them to creep towards genuine market rates would be a step forward. If they eventually reached “fair” rates, then all those rights to use slots in perpetuity that airlines value so much would become worthless. And airlines are likely to greatly resist this, because the slots are listed as intangible assets on their balance sheets, and losing this would be negative for them, even without the additional competition. It would be like telling New York taxi drivers that the plates that they paid hundreds of thousands of dollars for are now worthless, and that the number of taxis on the roads will be henceforth regulated through tolls. It might mean that the system is more efficient going forward, but having your assets seized is none the less not fun.

Enshrining rights to use slots in perpetuity as fully tradeable property rights would make ultimately achieving the optimal outcome much harder, as it would make slots more valuable than they are now and would give their owners all kinds of rights they do not have now. Whilst this would be a far superior situation to what he have now, I am not as in favour of it as I was when I started writing this post. I think I am more for deregulating landing charges and saying “the hell with it”.

However, this is likely to be very difficult to implement. The aviation industry does not work like this anywhere, and never has. The vested interests of the industry are unlikely to be disposed to like it.