BAA plc: A report on the economic regulation of the London airports companies (Heathrow Airport Ltd, Gatwick Airport Ltd and Stansted Airport Ltd)

Summary of report (html format)
Full text (pdf format)

Adobe Acrobat Reader can be downloaded from http://www.adobe.com

Summary

Under the references made by the Civil Aviation Authority (CAA) on 28 February 2002 (see Appendix 1.1), we are required to recommend the maximum levels of airport charges at Heathrow Airport Ltd (HAL), Gatwick Airport Ltd (GAL) and Stansted Airport Ltd (STAL), for the period of five years beginning on 1 April 2003. (This is the fourth regulatory period since privatization of the airports, referred to as Q4.) HAL, GAL and STAL are all subsidiaries of BAA PLC (BAA). We are also required to consider whether any of the three airport companies have, at any time from the date of the last references to us in December 1995 up to the date of the current references, pursued a course of conduct (in relation to matters specified in the references) which has operated, or might be expected to operate, against the public interest.

During the current regulatory period (Q3), airport charges per passenger at Heathrow and Gatwick combined have been required to increase by no more than RPI-3 and at Stansted by no more than RPI+1. The current reference follows a substantial review by the CAA of the appropriate basis on which to set the maximum level of airport charges. In the references to us, the CAA recommended a formula of RPI+6 at Heathrow, RPI+5 at Gatwick, and slightly above RPI+6 at Stansted (but did not expect STAL to be able fully to recover the price cap set). This represented increases in airport charges per passenger (in 2000/01 prices) between 2002/03 and 2007/08 from £5.37 to £7.25 at Heathrow; from £4.03 to £5.09 at Gatwick; and from £4.47 to £6.10 at Stansted.

During the inquiry, we had to consider a number of general issues relevant to setting the maximum level of charges, as well as a number of more specific issues relating to the future costs and revenues of the three airports.

The most controversial aspect of the CAA's recommendations was its proposal for a 'dual-till' approach. Currently, airport charges are set on a 'single-till' basis, which takes into account the revenues and costs of BAA's highly profitable commercial activities (primarily rental income from retail and other concessionaires). The CAA proposed that they should instead be set to cover the costs of aeronautical activities, ie those regarded as essential for the operation of the airport, and monopolizable by the airport, users therefore having little choice but to use them. This would exclude not only activities such as retail and catering outlets, but also surface access to the airports (such as the existing Heathrow Express) unless required as a condition for planning permission of new facilities (for example, the extensions to Heathrow Express and Piccadilly lines). Other things being equal, the dual till would lead to significantly higher airport charges over the longer term, but in practice the CAA proposed to set charges for Heathrow in Q4 at a level that was consistent with a single-till approach, since it regarded that as the reasonable upper limit for price increases in this period. The CAA proposed that the maximum level of charges at Gatwick and Stansted in Q4 would, however, be on a dual-till basis, and that Heathrow should recover in subsequent regulatory periods the difference between the single-till and dual-till approach at Heathrow in Q4.

Airlines all objected strongly to the dual-till approach, regarding it as unnecessarily increasing charges, to the benefit of BAA at the expense of airlines and, through the higher fares that would result, at the expense of passengers. Among their other arguments, they regarded commercial activities as an inseparable part of the aeronautical business; and there were also concerns that it gave BAA the incentive to favour commercial development at the expense of aeronautical development.

The main benefits claimed by the CAA and BAA for the dual-till approach related to effects on investment incentives, efficient utilization of scarce runway capacity, and deregulation of non-monopoly activities. BAA's investment programme-about £8 billion from 2002/03 to 2012/13 including about £3.5 billion on Terminal 5 (T5) at Heathrow-is important to the airlines. However, we found no evidence that the single till had led to any general under-investment, nor in current or foreseeable circumstances, any basis for an expectation that it would do so over the next five years; nor was it clear that the dual till would lead to better aeronautical investment in future. The dual till could improve the efficient utilization of capacity, but the benefits are unlikely to be more than marginal, even at Heathrow, where there is significant excess demand. The case for the dual till is, indeed, even weaker in Q4, where the CAA proposed to apply it only to the less congested airports: GAL and STAL.

Nor do we see significant benefits from any deregulation of BAA's commercial activities, most of which relates to the leasing of space for retail and other commercial activities. In so far as airport charges affect fares, BAA's current relatively high profits from its commercial activities are mitigated by the fact that they are applied to the benefit of passengers. The dual-till approach would break this link and may therefore require increased regulation of such activities. The dual till could also risk unduly benefiting commercial activities at the expense of aeronautical activities, which may not attract sufficient funds or management priority.

Against those, at most, limited benefits, we see significant disadvantages from the dual-till approach. We believe it is difficult sensibly to separate commercial and aeronautical activities. BAA's rental and other commercial revenues at the three London airports would not be generated without aeronautical facilities-commercial and aeronautical facilities are better, therefore, in our view, and more realistically regarded as one business. Since the successful development of commercial revenues requires airlines to attract passengers to the airport, the benefits of commercial activities should also in our view be shared with airlines and airline users.

The increase in airport charges resulting from the dual till could, on the basis of some figures we saw, increase the net present value of revenues to BAA over the longer term by between £3.2 billion and £3.7 billion. The CAA argued that this would merely be a shift of economic rents from airlines to the airports, but was unlikely at congested airports to make a material difference to airfares or passengers: it did not therefore see such a transfer of rent as of concern to the regulator in this case. We do not believe average fares would be unaffected at either congested or uncongested airports. We cannot, however, be indifferent to such a transfer of resources to a regulated utility whether the effect was on airlines or their passengers, potentially undermining regulatory credibility and creating regulatory uncertainty.

It is also difficult, in practice, to allocate either investments or operating costs between aeronautical and commercial activities. To the extent that some of the judgements that have to be made are arbitrary, future disputes about cost allocation could also harm relations between the airport and its users.

We recognize that the current level of prices particularly at Heathrow are very substantially below market-clearing prices or the long-run incremental cost (LRIC) associated with new capacity. But setting market-clearing prices is not consistent with international obligations, and the net effect of setting charges equal to LRIC (which would transfer some £10 billion of value from airlines and passengers to BAA) would be very detrimental to users. Our approach does, however, capture some important elements of a modified LRIC approach.

One argument put to us for the dual till was that it would raise prices towards market-clearing or LRIC-based levels. However, in the circumstances which Heathrow faces for the foreseeable future, the overall balance of effects on users would remain adverse, and the approach significantly less acceptable than the one we recommend. We do not therefore share the view of the CAA that the dual till is best calculated to meet the objectives of the Airports Act 1986 (Airports Act); in our view, with these objectives in mind, the single-till approach should be retained for Q4.

Among other aspects of the general approach we had to consider:

(a) In principle we believe that the 'system approach' to setting the maximum level of airport charges-with respect to the rate of return of the system as a whole rather than that of each airport-should be retained under current conditions. However, on the basis of our forecasts this makes no difference in the period we have to consider due to STAL's increasing profitability and the current projections of investment requirements of the airports.

(b) The CAA proposed a price path commitment for future quinquennia. One element of that commitment would be a significant increase in prices when T5 came on stream to reflect the high costs of T5, that element of prices being maintained in future quinquennia. There would also be a commitment to fix the element of charges relating to the current capital base. Such a commitment is intended to improve incentives to invest, and reduce the uncertainty as to the future return on projects which depend on the outcome of future reviews. In our view, however, there can be no meaningful commitment, since neither we nor the CAA can prejudge the outcome of future reviews; we also believe that, even if such a commitment could be given, it may be undesirable since it would prevent users sharing in the benefits of future cost reductions. In our view, a preferable, alternative approach to promoting adequate incentive to invest is to allow for assets in the course of construction (AICC) subject to a series of triggers relating charges to progress particularly of T5, but together, where necessary, with an element of smoothing of return between quinquennia, reducing reliance on future large increases in charges.

(c) During Q3 there was also allowance for AICC on T5. That investment could not proceed due to delay in the planning inquiry for T5. There was also a significant advancement of revenue for such smoothing purposes. We believe recovery of these revenues and of the effects of under-investment in Q3 should in principle be allowed for in Q4.

(d) Among more detailed aspects of the formulae, we believe BAA should still be permitted to pass on 95 per cent of the costs of additional security requirements given the uncertainty as to the future level of such requirements. As proposed by the CAA, we recommend that the formulae should apply to charges as if users pay the full price before discounts; and that revenues from non-passenger flights should be removed from the cap (subject to a condition that charges should be no more than for passenger flights).

We have estimated a cost of capital for the three airports of 7.75 per cent. This reflects in part the extra risk to BAA of the investment in T5, but which also in our view equally affects the cost of capital of all three airports. We have also taken into account scope for savings in operating costs through higher productivity, and lower pension costs. We have not adjusted BAA's forecasts for capital expenditure: even if there is scope for lower costs on some projects, there is in our view likely to be a demand for any cost savings to be spent on additional projects. We have, however, recommended a number of triggers relating airport charges to progress particularly on T5.

We have recommended formulae of RPI+6.5 at Heathrow and RPI+0 at Gatwick. We have based Stansted's cap on its current yield gross of marketing expenditure of £4.20 at 2000/01 prices throughout Q3, but, for the purposes of the projections, STAL is assumed to be unable to increase its net charges to that level within Q4. On this basis, however, each airport earns almost exactly its cost of capital over the ten years as a whole. The effect of our recommendation is to result in charges by 2007/08 of £7.82 at HAL and £4.08 at GAL (both in 2000/01 prices). STAL's net charges are expected to remain below the cap of £4.20. (Our figures are not directly comparable with the CAA's proposals, however, since they include the currently separate charge for transfer baggage infrastructure at Heathrow, and exclude cargo flights at Stansted.)

A wide range of public interest issues were raised with us. Some airlines were critical about aspects of BAA's investment performance, including its not promoting additional runway development at Heathrow and delays to additional pier developments. However, there were in our view good reasons for BAA's approach, in particular in relation to planning constraints and uncertainty following the events of 11 September. Underlying the complaints to us, however, is a perception of inadequate consultation between BAA and the airlines. The consultation process in Q3 has been hindered by uncertainty as to the outcome of the T5 inquiry and as to additional runway developments at one or more of the airports. We nonetheless have significant reservations about the process of consultation and as those uncertainties are resolved believe improvements both to consultation and complaints procedures (which BAA itself proposes to bring about) should be put into operation.

We also regarded it, however, as unsatisfactory that airlines appear to have to use five-yearly reviews as the main means to pursue complaints, many of which should, more properly, in our view, have been addressed when they arose. We acknowledge the significant constraints on the CAA resulting from the Airports Act, which hopefully can be resolved when that Act is brought into line with other legislation. But we also suggest that the CAA examine whether current procedures under section 41 of the Airports Act can be made more effective in resolving disputes between airports and users and whether it has more scope to be involved in relationships between airlines and users aside from its formal role under that section of the Airports Act. On one detailed issue, we have found aspects of the imposition of a levy on taxi drivers at Heathrow to be against the public interest, and we have put forward an appropriate remedy.

A further main issue related to BAA's quality of service. The CAA had been concerned with this and believed that BAA's revenue from airport charges should reflect the quality of service provided. To achieve this it proposed adding a term in the RPI-X formula which would partly relate charges to quality of service. We do not believe the average level of service in Q3 has been poor. We have, however, noted a variability of quality of service provided; that although service standards were agreed, they were not comprehensive; that variability in standards has been reflected in variability of charges in only a few areas; and that users have not known what quality of service they were entitled to expect. We are in no doubt that, in a fully competitive market, airport charges would vary according to the level of service provided, but as regards Heathrow and Gatwick, during the last quinquennium, this occurred only to a minimal extent. We find that in failing to conduct themselves so as to make prices paid sufficiently reflect the levels of service provided, HAL and GAL have pursued a course of conduct which may be expected to operate against the public interest. The adverse effect of this conduct is that charges do not adequately reflect the quality of service provided to the extent that would occur in a competitive market and in consequence that there is an absence of the financial incentive to provide the combination of price and quality of service that would obtain in a competitive market.

We have recommended that HAL and GAL be required to pay specified rebates to the airlines whenever their quality of service fails to meet such performance standards as may from time to time be specified by the CAA (guidelines for which we have set out in our report) together with other requirements relating to the operation of its Quality of Service Monitor, which we regard as an important part of monitoring of quality of service.

Full text

Back to the top