Mike Toms
The first structured system for the economic regulation of airports was established by the UK Government in the 1986 Airports Act. Although the system was designed to be workable for both publicly and privately owned airports, its political context was a clear intention by the then Conservative government to privatise airports wherever possible. The intention of the regulation system was to capture the best features of private ownership and management, whilst ensuring that airports did not unreasonably exploit their market power. An appreciation of this philosophical framework is important to understanding why the system operates as it does.
The system is not intended simply to prevent airports making excessive profits; it aims to encourage cost efficiency in operations, and innovation in management and business development. The regulatory framework was built upon the structures previously created for the two major preceding privatisations, British Telecom (BT) and British Gas (BG). In both cases the emphasis was on 'light-touch' regulation, where having set a price formula for five years, the regulator would leave the company alone between reviews. (In practice, this deteriorated rapidly, and both industries were subjected to regular and increasing intervention.)
The key elements of the airport system were:
The decision to create a single separate specific sector regulator for each privatised industry was in itself an interesting step. The Government had had the option of using pre-existing legislation (the Fair Trading Act and the Treaty of Rome), and existing regulatory bodies, to control the industry. These laws and institutions are powerful bodies, so why create another set of controls? The reasoning was never full articulated, but with the benefit of hindsight, it appears that:
So has the system stood the test of time? Have the five pillars of regulation remained standing over 15 years?
The decision not to break up BAA and introduce 'competition' between the London Airports was controversial, and has re-emerged at regular intervals. (Interestingly, the demand for break-up has tended to come from academics, political sources, and potential investors, rather than from airlines. With the exception of Virgin, British airlines have remained conspicuously neutral on the subject of airport ownership.) The notion of competing airports has a superficial attraction, particularly to free market economists in the 'think tanks'. It has been tested again periodically.
In 1991, the Monopolies and Mergers Commission (the predecessor to the Competition Commission) investigated the possibility of a "break up" in its five-yearly review of charges. It rejected the notion, albeit with a dissenting note from the Commissioner associated with right wing political and economic thinking.
In 1996, the Commission again considered requests that BAA be broken up, but concluded that BAA had not been guilty of any misconduct that justified this drastic action.
In 1999, the Chancellor, Gordon Brown, announced that the scope for competition between airports would be investigated, by a team led by the Deputy Prime Minister. Given the Chancellor's pro-competition views, it was expected that break-up was a real possibility.
However, having taken evidence from the Civil Aviation Authority, a joint team of the Departments of Transport, Trade and Industry and the Treasury, concluded that break-up would not offer benefits to consumers that outweighed the damage. It therefore concluded that break-up should not be pursued at that time. (This qualification is interesting, since if left in the air the possibility of break-up sometime in the future. This can be interpreted as a Sword of Damocles, designed to ensure the good conduct of the company; as such it is an effective piece of shadow regulation.)
Most recently, the Competition Commission again considered airport ownership, as part of its 2002 investigation. Again the notion was rejected, although the Commission suggested that the issue might need to be reopened after the Government has published its plans for the development of the airports in a policy statement (or 'White Paper') expected in late 2003.
These repeated debates have illustrated a number of clear themes;
The question of break-up and competition will therefore no doubt re-emerge in the future, and the threat provides a useful constraint on any inclination by the airport operator to abuse any dominant position it may enjoy.
Since 1986 there have been three attempts to extend direct price control to smaller regional airports.
In 1994, the Government investigated whether BAA's Scottish Airports should be subjected to RPI-X. BAA made a strong case against price control, and offered a voluntary package of measures, including a three year RPI-3 formula, to head off the threat. The Government accepted these undertakings and decided against any action.
In 1994, Easyjet sought price controls at Luton airport to prevent substantial price increases. The Government rejected this request on grounds that the airport was not profitable (!) and, interestingly, on the grounds that Easyjet could, if it wished, transfer its operation to Stansted.
Conversely, in 2001, the Government's Better Regulation Task Force recommended that the Government consider the lifting of price controls on BAA's airports to encourage direct negotiation with airlines and eliminate regulatory gaming. The increased general power to remedy monopoly abuse contained in the 1998 Competition Act could provide an effective backstop. Unfortunately, the Government did not act on this recommendation.
This position can be set against the decision of the Australian Government, to abandon RPI-X type price control at the major airports, for a trial period of five years, to test whether normal commercial negotiation will produce a more effective outcome. In doing so, they had recognised an important feature of airports, which is the countervailing economical and political power of small groups of highly organised and litigious customers.
The RPI-X approach to regulation was first adopted for airports as an extension of the principle originally devised as a temporary method of regulating telephone charges levied by British Telecom, until competitors emerged in the telecoms industry.
When RPI-X was imposed on airports it was initially opposed by airlines, who preferred Rate of Return, or Cost Plus regulation. Subsequently airlines have learned to support RPI-X for the benefits it brings in terms of cost control and guaranteed prices. RPI-X is now solidly established as the preferred basis of regulation by all sides of the industry.
However, in the course of the 1991 and 1996 Reviews, the application of RPI -X has tended to deteriorate into a strict single-till Rate of Return based formula, with protracted arguments about the cost of capital, the definition of the asset base and financial protections.
The formula itself has become progressively more complex, with somewhat spurious precision replacing simplicity and transparency. The latest manifestation of this is the proposed incorporation of triggers in the formula, linking the value of X in individual years to delivery of individual elements of capital projects.
The Airports Act 1986 introduced a complaints procedure in which airport users could complain to the CAA about airport conduct that might be predatory, discriminatory or an otherwise unreasonable exploitation of monopoly. In addition, the Competition Commission was required to consider all public interest complaints at each five-yearly review.
Given the tendency of airlines to denigrate airport conduct, it is remarkable that there have been only 11 complaints to the CAA in the last 15 years. Complaints to the Commission have been made at each review, but with only minor consequences. Interestingly, while the flow of complaints dried up completely over the past five years, in the current (2002) review, the Commission received 87 separate complaints. Of these only two were upheld.
Airlines have argued that the CAA Complaints Process is too expensive, time consuming and bureaucratic, but an alternative interpretation is that the process has simply demonstrated that airlines have little confidence in the strength of their own complaints. It may be concluded that the existence of the complaints process has itself been an effective deterrent to unreasonable conduct by airports.
It is certainly true that the CAA has not encouraged airlines to complain, and the BAA has now stepped in to propose its own informal fast track internal complaints process. This provides an interesting example of self-regulation superseding imposed structures.
Interestingly, when the system of airports regulation was first established, little attention was given to service quality, and the issue did not even emerge in the first review in 1991. Paradoxically, at the next review, the Mergers and Monopolies Commission noted that the level of service quality had improved, but also bemoaned the absence of formal quality regulation as a structural weakness in the system. It therefore put the BAA on clear notice to engage with airlines on a voluntary system of Service Level Agreements with airlines, to avoid more formal regulation at the next review.
BAA did indeed discharge this obligation in full. However, at the current review, ever though the Commission has found service quality to be generally satisfactory, it has decided to implement a complex system of penalties for performance failure. It is not immediately clear how this reconciles with light touch regulation. It is however clear that it is a move towards the kind of interventionist quality regime that has served the railway industry so badly.
The two-regulator model, with the CAA taking decisions, advised by the Competition Commission, contrasted with other sectors in the UK in which the single-sector regulator prevails, with the Competition Commission used only on appeal. The airports model originally had two purposes:
This has been possibly the most interesting area of the regulatory structure. At face value it should have led to a fine complementary. In practice, the result has been quite different. In the 1991 Review, the CAA rejected the Commission's use of an 8 per cent cost of capital, giving RPI-3, and proposed 7 per cent, equivalent to RPI-8. In the end, middle ground was found at 7½ per cent, equivalent to RPI-4.5. In this instance the CAA had usurped the Commission's expertise on Cost of Capital.
In the 1996 Review, there was a broad consensus around 7½ per cent, leading to RPI-3, although the CAA did float options on the phasing of charges that had been rejected by the Commission.
In the current review, at the time of writing, large-scale differences of approach appear to be emerging between the two regulators. The CAA approached the review from first economic principles - normally the remit of the Commission whilst the Commission appears to be intent on simply defending the existing regulatory model with the aim of protecting the precedent it sets for all the industries it has to regulate. Differences have included such fundamental issues as;
This is resulting in a lone and intensive process with a highly uncertain outcome.
The situation points to some rules for regulation;
In undertaking analytical work, both regulators should have a mind to the fact that the rigour and quality of their analysis should be sufficient to satisfy not just themselves, but also the other regulator. The adoption of this principle by the CAA would certainly have eased the current review.
Where two regulators can differ as much as the CAA and Competition Commission appear to, the credibility of regulation itself is open to question. This is not helped when both regulators adopt confrontational styles, and engage in a manifest contest of academic arm wrestling. In that context, regulators could learn from the commercial companies they regulate, who have to manage their own business partner relationships towards agreement, not victory.
The regulation of airports in the UK was originally devised to create a climate in which airports act responsibly and commercially, and negotiate directly with strong and organised airlines within a fixed price ceiling, and without the distortions created by continuous regulatory interference. To this end, there was a simple price formula, reviewed through a simple review every five years.
The first review, in 1991 took 12 months.
The second review in 1996, took 21 months.
The current 2002 review has taken 32 months.
The first formula was a simple value of X.
The current proposed formula is a value of X, with six trigger points, and a service quality scheme covering a dozen parameters. In addition, the BAA operates under a web of agreements and undertakings which has grown at each review.
So much for light touch regulation.