We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

The insane world of bilateral international aviation regulation

It was recently announced that after talks between the British and Hong Kong governments, Richard Branson’s Virgin Atlantic Airways had won its long desired rights to fly from London to Sydney, Australia. In return for this, Hong Kong’s Cathay Pacific Airways were given the right to fly from London Heathrow to New York and other cities in the United States. Various observations were made about how an additional competitor on each route would increase competition and give passengers lower fares and more options.

While this is true as far as it goes, this is a pretty bizarre paragraph if you think about it. Why does the British government have to negotiate with the Hong Kong government before a private company can fly to Australia? In what parallel universe is the quid pro quo you must offer to get your airline permission to fly to Australia the permission for another airline from a third country to fly to New York?

And if additional competitors are good on routes, why were these airlines not allowed to fly on them already? And why did Singapore Airlines, Delta Airlines, and Continental amongst others object strenuously to the deal?

To answer these questions, we have to look at just how international aviation is regulated. This is bizarrely anachronistic. This most global of industries is regulated by a web of bilateral treaties between nations that dramatically limits competition. And to find this out, we have to look back into the dim depths of the past, to 1944. In the late days of the Second World War, it was widely recognised that international air travel was going to be a much bigger deal after the war than it had been before, and the instincts that created the new framework were, to put it mildly, protectionist.

The basic principal of the Chicago convention was that treaties allowing airlines to fly between countries would be bilateral in nature, negotiated between governments, and that airlines would generally only to allowed to fly either from or to their home country.

That is, flights between countries A and B would only be provided by airlines from countries A and B. In most instances each of these countries would have a single state owned airline licensed to operate on international routes. These two airlines could operate the route in peace, without there being any pesky competition. (Under the auspices of the IATA, they were generally required to charge the same fares as each other, too). Even countries that had private airlines (for instance the United States) generally gave favoured regulatory treatment to certain designated “international” carriers (in the case of the US, mostly Pan Am and TWA). So airline seats were expensive and regulated.

However, things were a little different for extremely long haul routes. You see, aircraft could (and can) only fly so far. Every now and then they have to stop to refuel. And if they are stopping to refuel anyway, why can’t they pick up and put down passengers at the airports where they stop? If an airline based in country A is flying to country C but has to stop on the way in country B, then there are almost always treaties between A and B and also between A and C that apply. If such an airline flies from A to C via B, then it is almost always allowed to put down passengers who got on at A at country B, and on the return leg it can carry passengers from B to A. And it can carry passengers all the way from A to C and on the way back it can carry passengers from C to A. What it cannot do is pick up passengers in B and carry then to C to replace the passengers from A who got off at B. Therefore, in such circumstances an airline tends to have full planes on the leg from A to B, but half empty planes on the leg from B to C. It is not allowed to sell tickets to fill these empty seats.

Such an airline is at a serious commercial disadvantage to airlines that are actually based in country B. Airlines based in B will likely have treaties with A and C. They are allowed to fly passengers in both directions between A and B, and they are also allowed to fly passengers in both directions between B and C. Because flying passengers from A to C is just a combination of these two journeys, both of which are permitted, it can also carry passengers from A to C. Therefore, an airline based in country B can fill its planes with all three types of passenger, and is at a commercial advantage compared to the first airline.

Unsurprisingly, airlines in country A have never been very keen on this kind of arrangement. Therefore, they have always looked for stopover countries willing to grant them so called “fifth freedom” rights. (Also sometimes called “beyond” rights. Freedoms one to four, and the unofficial freedoms six to eight are listed here, although fifth freedom is by far the one the most discussed). Fifth freedom rights are the right to pick up passengers in one foreign destination and carry them on to another foreign destination. Given that it is surely in the interests of a country to have as many connecting flights to the rest of the world as possible, one would hope that lots of countries would offer such rights. On the other hand, those who have studied how governments actually work would be less likely to conclude this. Unsurprisingly, local airlines hate having foreign airlines that actually compete with them, and nationalised (and private but protected) airlines tend to have the ear of the government to a greater extent than do genuinely competitive market based airlines. So there is usually more reluctance to grant fifth freedom rights than there should be.

Just to repeat the key definition: A fifth freedom right allows an airline to carry passengers between two countries, neither of which are the home country of the airline.

But none the less, one or two countries in useful places have consistently offerered fifth freedom rights to most foreign airlines. One such country is Singapore, which had (and has) a government that smart enough to realise that it is in the country’s influence to be an aviation hub. Most airlines were and are permitted to fly into Singapore to put down and pick up as many passengers as they like, and then continue on their way. (This hasn’t hindered Singapore from developing a large and successful airline of its on – in fact it has probably helped). Thailand has always been fairly liberal, too, which is why most flights from Australia to Europe go via either Singapore or Bangkok.

And, in the aftermath of the second world war, there was another approach, which was that the country that won the war (ie the United States of America) could force the defeated powers to agree to unequal treaties with it. In particular, the United States signed a very one sided aviation treaty with Japan, which gives two US airlines (originally Pan Am and TWA, but now United and Northwest) essentially unlimited fifth freedom rights from Japanese airports. The US also signed a treaty with Germany, which was also quite one sided, and gave rights to Pan Am in particular that allowed them to create a hub in Frankfurt.

However with the other victorious powers the US was not able to obtain equally favourable treatment. In particular, the treaty signed with Britain (Known as Bermuda II) was quite restrictive for US carriers. Initially, it only allowed 2 US airlines, and two British carriers to fly from the US to Britain. Once again, these were initially Pan Am and TWA.. (They are now United and American). On the British side, for a long time only one airline (BOAC, the predecessor airline to British Airways) made use of the rights in a significant way. Other airlines used the rights occasionally but nobody was big enough to make a go of it until the emergence of Richard Branson’s Virgin Atlantic as a major competitor to BA in the 1990s. The most important impact of the treaty was to prevent the large number of American airlines that would like to compete with BOAC/BA on the all important trans-Atlantic routes from doing so.

In a series of negotiations, the Bermuda II treaty was modified in little ways. Most importantly, other US airlines were allowed to land at Gatwick and eventually Stansted, but the all important rights to land at Heathrow continued to be restricted to four airlines. By preventing airlines from flying into Heathrow, the present treaty restricts airlines other than the “big four” from gaining anything other than marginal profitability on routes to the UK. The reason for this is that business class passengers are responsible for an inordinate portion of the profitability of long haul airlines, and business class passengers tend to avoid London airports other than Heathrow. An aircraft with few business class passengers can be full, but not especially profitable. At the moment, Continental, Northwest, US Airways and Delta all have flghts from the US to Gatwick, and all would like to be able to fly to Heathrow, but none of them can. (American Airlines and Continental have at times offered trans-Atlantic flights from Stansted, but both have ultimately withdrawn the services). On the other side of the treaty, British Midland Airways (which has an extensive series of domestic and European routes out of Heathrow) would also like to be able to fly between Heathrow and the US.

At the end of the cold war, the post war aviation treaty between the US and (formerly West) Germany was clearly anachronistic and needed to be renegotiated. In addition, the prevailing opinion amongst the bodies negotiating these things had shifted (slightly) in favour of more liberal policy. Starting with Germany, the US started negotiating what it referred to as “Open Skies” agreements with European countries. These were agreements allowing any US airline to fly from any US airport to any airport in the foreign country, and which also granted full fifth freedom rights to airlines of both countries. US carriers could (for instance) fly to Frankfurt, pick up and put down passengers, and then fly on to Moscow. Or, if they had signed these agreements with both Italy and Germany, they could fly from New York to Frankfurt and then on to Rome, essentially flying internal routes within the European Union. Which is another problem comes from.

You see, fifth freedom rights de facto mean something different in the case of the US to what they do in the case of Europe, because Europe consists of lots of little countries and most flights are international. A series of these “open skies” agreements within Europe would allow US carriers to fly internal flights within Europe. As the US is a big country, internal flights within it qualify as domestic flights, and granting fifth freedom rights to foreign carriers does allow them to fly to foreign destinations from the US, but not to fly internal flights within the US. In the case of Germany, accepting such a deal led them to gain quite a bit – increased access to US airports – but did not cost them much, as the existing treaty was heavily biased towards American interests anyway.

In the case of the UK, however, there was quite a strong objection to granting such fifth freedom rights. The UK was not willing to grant US carriers permission to fly to London, then pick up passengers to fly on to Paris or Frankfurt unless in return British airlines (notably British Airways) could pick up passengers in New York and then fly them on to Los Angeles or San Francisco.

So we have had an impasse. US and UK authorities have met with each other a number of times in recent years to attempt to negotiate a more liberal agreement than Bermuda II. The US has been absolutely insistent that any new agreement must allow an unrestricted number of US airlines to fly to the UK. The UK has been absolutely insistent that any new agreement must allow British airlines some domestic rights in the US (notably the right to pick up passengers on the east coast and put them down on the west coast, and vice versa). A new agreement making things even slightly more liberal would be vastly in the interests of consumers, but no progress has been made whatsoever. The US negotiators are unable to put what the British negotiators want on the table, for two reasons. Firstly, US airlines are highly unionised, in most instances financially quite precarious, and have considerable political influence, and for this reason allowing foreign carriers to carry domestic passengers is something considered unacceptable to these political interests. Secondarily, when it negotiated the “open skies” agreements with other countries, the US gave these countries the right to renegotiate their deals if the US concluded a more favourable deal with any other country.

British negotiators have slightly more flexibility, but it doesn’t seem to matter. Although British Airways are no longer a state owned company, they might as well be to the extent that the UK Transport ministry negotiates on their behalf. In the eyes of the ministry BA’s interests are paramount. It would certainly be in the interests of consumers and the British economy as a whole to simply open access to Heathrow without any concessions from the US, but negotiators are not willing to do this. Therefore, the last several attempts to negotiate a new British-American aviation deal have ended acrimoniously after a couple of days with nothing resolved.

But, there is one other way that additional airlines can enter the Heathrow-US market. This is for some airline from a third country with landing rights in the US to be granted fifth freedom rights by Britain. If this happens then that airline can start flying between Heathrow and the US without the US-UK treaty being renegotiated. And this is what has just happened with Cathay Pacific. But as to why, we need to look at the Australia-UK market.

There are three airlines that compete for the premium traffic between the UK and Australia. These are Qantas, British Airways, and Singapore Airlines. British Airways owns a substantial equity stake in Qantas, and the two airlines have coordinated their schedules and do not compete with each other on the route. (When this coordination occurred a few years back, various concessions were made, one of which was that other Australian and British airlines would be given the right to operate the route if the issue came up). Singapore Airlines has the rights to fly to both Australia and London, and has coordinated its schedules so that the layover in Singapore is very short, and passengers can get from Australia to London and vice versa is the shortest possible time.

There are many other airlines that sell tickets from London to Australia, but these do not generally compete for the premium traffic. Generally the schedules of their flights from Australia to Britain are not coordinated, and passengers have to spend a few hours in the terminal in Asia or the Middle East somewhere. This is fine for passengers who either don’t mind spending a few hours in a terminal in return for a cheaper fare or who like to break the journey for a day or two in an Asian city, but this does not generally appeal to business passengers who are generally trying to make the journey in the shortest possible time. For these airlines, carrying Australia-London traffic is about filling empty seats rather than using it as a core business.

However, for some time Richard Branson’s Virgin Atlantic has been interested in entering the Australia-London market in a serious way, to compete with Qantas/BA and Singapore Airlines. It has had no difficulty in obtaining the right to land in Australia, but it has had significantly more difficulty finding somewhere along the way to stop that would be willing to offer the necessary fifth freedom rights. Ideally, Virgin wanted to extend flights from one it its existing Asian destinations, the most suitable being Hong Kong. (Singapore was out, partly because Virgin did not fly there already, and partly because Singapore Airlines is a shareholder in Virgin Atlantic and competing this directly would cause problems in the relationship between the airlines). However, Hong Kong has not traditionally granted fifth freedom rights with as much willingness as has Singapore. Therefore, the British negotiators had to put something on the table in return for Hong Kong granting these rights to Virgin. And the thing that Cathay Pacific wanted was fifth freedom rights from London in order to fly America-London routes.

So that is what happened. Hong Kong and Britain did a deal, and as a consequence Virgin Atlantic can fly from London to Australia, and Cathay Pacific can fly from London to New York.

However, lots of other airlines are upset with this. United, American, and BA are mildly annoyed because they will face another transatlantic competitor. Northwest, US Airways, Continental, Delta and British Midland are more annoyed than this, because they really don’t like it when they are not permitted on a route from their own country when an airline from an entirely neutral country is allowed on it. And Singapore Airlines are absolutely livid. They have long coveted Heathrow-US routes at least as much as Cathay Pacific have. Traditionally, Singapore has been far more unilateral than Hong Kong in offering rights to other airlines, and in return for this they are now locked out of a route that a traditionally less liberal rival is allowed into. The arguments will go on for a while, but while the deadlock in negotiations between the US and UK goes on, none of these complaining parties are going to get much satisfaction.

And so here we have a demonstration (in gruesome detail) of what happens when bilateralism rules. It isn’t pretty. One would hope that blind Freddy could see that consumers and economies would be far better off if every airline that wants to fly across the Atlantic or from England to Australia could do so. Regulation has harmed consumers horribly, and has only helped a few vested interests (and it may not have done that). What is needed is for the whole bilateral framework to be thrown out and replaced with a new multilateral framework.

However, I am not holding my breath.

That said, a decade ago international telecommunications was regulated with a bilateral framework very similar to that of airlines, and that is now largely gone. That is a very different industry that has undergone much more technological change, but the example perhaps gives us reason to hope.

8 comments to The insane world of bilateral international aviation regulation

  • Mark

    I travel from Texas to London for business 3-4 times a year. And I fly Continental to Gatwick. With a non- stop flight and the Gatwick Express into Victoria Station it is by far the fastest travel arrangement I can make. And Gatwick is far less chaotic than Heathrow.

    As an aviation nut, I’ve followed the airline industry and regulation for some time. Thanks for putting together a concise essay about the whole mess.

  • mad dog

    Perhaps the problem is a simple “supply and demand” problem. There are less “landing slots” in most big international airports than people wanting to use them. Hence the “constriction of competition” referred to, I suppose.

    Maybe the constraints in the story would have been more understandable if it had been “Iran Air” asking for a few more slots into New York. An aircraft seems to have become a “psuedo weapon” of mass distruction and so its use is ever more regulated.

    I suppose this might be the precursor of many such examples where the free market will be additionally constrained for reasons of “national security”.

  • Rob

    Absolutely fascinating post; I had no idea any of this was going on. And what a complicated mess it is!

  • John Thacker

    mad dog’s comments don’t quite seem to be on point. If he had actually read the essay, he would have noted that in fact landing rights have NOT in fact been “ever more regulated.” The most stringent regulations date from 1944, as noted. In fact, the trend has generally been towards less regulation, though in small steps.

    Perhaps before a US-UK deal can be negotiated, there will have to actually be a US-Canada deal reached allowing each country’s airlines to fly domestically in the other. Such a preliminary agreement might be easier to pass, and would start to weaken the power of the airlines and their union allies.

  • Dave

    Good essay and interesting reading – I’d seen something similar from an MBA student on my wife’s MBA who worked for Continental who reviewed this subject matter from a slightly different angle.

  • Robert

    There is a terrific site, which I have lost, that has the full text of almost every int’l aviation treaty. I believe it’s Danish. Can anybody post it?

  • Maz

    youre a googlewhack
    with acrimoniously porcupines

    see http://www.googlewhack.com/tally for more info



  • bill n

    Are routes tied to particular airlines or do the routes belong to the government. For example if Northwest was to be bought by another airlines now that they are in bankruptcy, could the routes be transferred or are they linked specifically to Northwest? What about individual routes? Could Northwest sell a Japan/Singapore route for instance?