TPC Reviews Compass Failure

• any contribution made by restricted access by Compass to airport and terminal facilities;

• the impact of the failure on Compass passengers with pre-paid tickets; and

• any evidence of potential breaches of Parts IV andV of the TPA.

Contribution of Compass to its Own Failure

On the information available to it, the Commission has concluded that shortcomings in the entry strategy and management of Compass itself made a significant contribution to the failure of the airline. Most importantlY, Compass’ entry into the domestic market in 1990. The strategy also misjudged the competitive advantages of the incumbent airlines, their likely responses to the additional capacity introduced by its entry and the capitalisation required to achieve viable entry under those conditions. In addition, the severity of the recession worsened the market environment for all three major airlines, resulting in fiercer competition than might have otherwise occurred.

Compass’ entry strategy involved providing a single class service on major trunk routes at fares of 80% of the standard economy of the incumbents and 50% for back-of-the-clock services. This strategy assumed Compass had a cost advantage and the incumbents would not respond competitively.

Cost Competitiveness

The Bureau of Transport Communications & Economics (BTCE, 1991) pointed out that a new competitor can gain access to most available economies of route density by flying a small number of key routes. Hence Compass was probably not disadvantaged by entering initially with a relatively small network.

The BTCE estimated Compass’ direct operating costs per available seat kilometre (ASK) at 5. 3c, compared to 8. 5c for the incumbents. The main reason for this cost difference was the larger aircraft employed by Compass. In addition, it had the benefit of a larger average stage distance and a higher rate of aircraft utilisation. Compass’ overheads were estimated by the BTCE to be 2. 7c perASK compared to 5c for the incumbents. That is, Compass’ total operating cost per ASK was about Sc comparedto 13. 5c for its competitors. Compass also had cost advantages arising from its relatively productive use of flight crew and ground staff.

However, these cost differences in favour of Compass should not be misinterpreted. On those routes where Compass was competing directly with other major airlines, their aircraft would have been larger on the average than for their networks a. s a whole and the difference in cost would have therefore been smaller. Part of the remaining difference in cost per ASK on those routes was attributable to the fact that the multi-class service provided by Ansett and Australian allowed fewer seats in an aircraft of given size and involved higher inflight service costs than the one-class service provided by Compass.

In any case, Compass’ apparent cost advantage perASK did not translate into higher profitability. Profitability depends on the relationship between cost and revenue per passenger. Both cost and revenue vary due to differences in the quality and type of services provided, as well as the size and efficiency of the aircraft used.

Ansett and Australian were catering to the first and business class market as well as to the economy and discount markets. To some extent their higher operating and overhead costs reflected the additional cost of services to this high yield market. However, by providing those services the incumbents were able to charge premium fares which earned a more than proportionate increase in revenue and margin contribution per passenger.

Assisted by their sophisticated yield management systems, the incumbent airlines were able to segregate business, economy and discount passengers in their aircraft and to match Compass discount fares on certain seats which would otherwise have flown empty and therefore had a relatively low marginal cost.

In contrast, Compass was providing a single class ‘no frills’ service based on discount fares, catering principally for leisure and non-business travel. Although it did not incur the additional costs of services for premium passengers, neither did it earn the relatively high revenues earned by its competitors from catering to that market. When the incumbents matched discounts for selected seats, Compass was forced to offer progressively deeper discounts for the majority of passengers on its wide bodied aircraft to maintain load factors and cashflow.

The decline in profitability was accompanied by a decline in revenue yield. After an initial increase in yield during the first two months of operation, yields declined consistently each month. Initially, average revenue per passenger of $165 had been earned. However, at the time of its collapse Compass was earning approximately $100 per passenger. The prospectus forecast for 1990-91 was based on a yield of $184.

It is interesting to note how the relationship between yield, passenger numbers and load factor changed over time. Over the 1 1 months of operation, as new routes were serviced and fare discounts stimulated demand, the number of passengers carried increased by approximately 250%. However, revenue yield decreased steadily during the period February to November 1991.

Compass’ passenger numbers increased steadily through 1991 as a result of this strategy, and its load factors also increased from February onwards. However, the strategy also resulted in a steady decline in revenue yield during the year. Thus, despite its apparent cost advantage, Compass was making a smaller contribution to the recovery of overhead costs than its competitors. This was an important factor in the eventual failure of the company.

Finance and Management

Compass raised $62m in equity capital but had a number of initial claims on those funds which limited the availability of working capital, resulting in liquidity problems from the very start of operations. Initial security deposits on aircraft were $12. 7m and increased to $24. 4m during 1991. The commission on the public subscription was estimated at $2. 5m and there were necessary purchases of property and equipment estimated at $38m in the prospectus. In addition, lease payments on aircraft comprised 30% of monthly cash outlays. (This could have been reduced by at leasta third if used A300s or by more than half if ex-Eastern 757s had been acquired. Ed)

Compass experienced difficulty in paying major creditors early in 1991 and even at that stage it was unsuccessful in seeking to raise debt financing from major banks to overcome its liquidity problems. The period May to December 1991 saw Compass relying heavily on prepaid ticket sales to maintain day to day liquidity. Prepaid sales showed an increasing trend from May to November, with particularly sharp increases in May and August.

Large numbers of presold tickets were released onto the market by Compass in May, which was a time of tight liquidity. The TPC understands that the sale of prepaid, low yield fares at this time was specifically targeted at raising cash. The fares were struck at a yield of 6-7c per seat kilometre, which was above the BTCE’s estimate of direct operating costs but below its estimate of total cost. Management reports indicate that approximately $5m of prepaid sales were planned in May but actual sales during that month amounted to $25m. While this may have satisfied Compass’ short term cash requirements, the forward bookings contributed to poor financial performance in the actual months of uplift. Furthermore, having undertaken the commitment to provide these services, it would have reduced Compass’ revenue flexibility on some ofits most attractive routes asits capacity to provide services to high yield traffic would have been compromised.

Compass appears to have used sophisticated financial and management accounting systems but there is some doubt about the accuracy and timeliness of available management information and the use made of it. The commission understands that Compass did not have an effective yield management system in place.

Competitive Strategy

Compass sought to compete on the basis of filling large one class aircraft with discount fares 20% below the standard economy fares of the incumbents. Howeve1; it did not seem to anticipate that its discount fares would be matched by Ansett and Australian, even though its entry added significantly to the number of seats available on the major trunk routes. Moreover, Compass did not use yield management techniques effectively to contain the decline in its yields, partly because its information system was not up to itand also because its strategy and facilities were not geared to the high yield traffic. As a result, Compass’ yield fell in relation to direct operating costs during 1991 reducing its contribution to overheads. The yields of the incumbents also fell during 1991 but not as far as Compass. In spite oftheir higher costs per ASK, they consistently achieved higher contributions to overheads than Compass during 1991 largely because of their more diversified services and effective yield management.

The Competitive Responses of Ansett and Australian

Faced with excess capacity and a depressed economy, all airlines offered deep fare discounts during 1991. This stimulated a substantial increase in demand for air travel in spite of the recession. Although this strategy maintained, and even increased their load factors, it also caused a steady decline in their revenue yields and ultimately contributed to the failure of Compass.

There are two alternative hypotheses which might explain the competitive responses of Ansett and Australian in 1991 to the entry of Compass into the market. One hypothesis is that they pursued predatory pricing and other strategies to drive Compass out. The second is that they pursued legitimate competitive strategies to maximise profits (minimise losses) under conditions of excess capacity and general economic recession.

The following sections attempt to evaluate the validity of these two hypotheses in terms of the observed market conduct in 1991. However, as background for that analysis the requirements of competition law regarding predatory conduct are set out below.

Predatory conduct and the Trade Practices Act

The Trade Practices Act (TPA) does not define or prohibit predatory conduct as such. Rather, Section 46 of the Act prohibits corporations with a substantial degree of market power from taking advantage of that power for the purpose of:

• eliminating or damaging a competitor in that or another market;

• preventing entry to that or another market; or

• deterring or preventing competitive conduct in that or another market.

To prove a breach of S. 46 it is necessary to establish that a corporation has the necessary power in a market, and that the corporation took advantage of that power for a proscribed purpose. Neither the TPA nor the limited Australian case law on misuse of market power gives clear guidance on how predatory conduct may be distinguished from vigorous competition. The TPC recognised this in its 1990 guideline on the misuse of market power. It also indicated that in assessing whether such conduct may infringe S. 46 it would take account of all of the evidence which may establish the existence of one of the proscribed purposes. The question of whether price is being set below average or incremental cost will be only one of the factors to be taken into account.

Predatory Conduct

Predatory pricing involves setting prices below cost to damage or drive out competitors, so that prices and profits can be increased in the future. Non price predatory conduct involves strategies to raise the costs of rivals or lower their revenue in order to damage them or deter entry into the market.

There is broad agreement that to distinguish predatory from competitive conduct, it is necessary to establish whether market conditions exist under which predation would (would not) be feasible and to assess whether there is a rational commercial explanation for the observed conduct, other than predation against Most commentators suggest a comparison between the predator’s price and its average total cost and average variable cost as one indicator ofa possible predatory motive. However, it is recognised that in certain special circumstances (such as the need to avoid redundancies, or to clear excess stocks of special perishable goods and services) pricing below average variable cost could be justified and consistent with competitive behaviour. However, in most cases pricing below average variable cost is difficult to justify on commercial grounds and may be indicative of predatory intent.

Profit Maximising Competition versus Predatory Conduct

Ansett and Australian maintain that given 1991 conditions of economic recession and excess airline capacity due to the entry of Compass, they could have sought to maximise profits (minimise losses) either by:

(a) Matching market discounts for otherwise empty seats while seeking to retain business/standard economy traffic to support revenue yields; or

(b) Retaining their fare structures and high yield traffic as far as possible while seeking to maintain load factors by withdrawing services and capacity.

They say (a) was a sound commercial strategy in depressed economic conditions with excess airline capacity, relatively high fixed costs and low marginal cost for seats that would otherwise fly empty. However, they say (b) would not have been as effective because it would not make efficient use of aircraft, staff and facilities and under it Compass would have expanded capacity and obtained an even larger market share. Strategy (a) combined with effective yield management, was said to be profit maximising. It achieved maximum contribution to overheads after covering direct operating costs, and it took account of competition from the other major airline as well as from Compass.

However, the alternative hypothesis, which also appears consistent with the observed behaviour, is that both incumbents may have recognised (separately) that a reduction in capacity was necessary to return to profitable trading in the medium term and that the self interest ofeach would be served by the elimination of the most vulnerable supplier, Compass. Given the under-capitalisation of Compass, its liquidity problems and its focus on the low yield end ofthe market, both airlines may have seen advantage in a short term competitive strategy of matching Compass to keep its revenue earnings at or near direct operating cost, recognising that failure to cover its overheads would be fatal within a relatively short period.

Such a strategy would not have been rational, however, if there was reason to believe that the domestic aviation market is contestable; that is, if Compass failed, other newly established or experienced airlines were likely to enter the market. Moreover, as noted above, the predatory purpose would be difficult to infer from observed market conduct which is also consistent with hard, profit maximising competition.

The airlines also maintain that their non price strategies (CRS, travel agents links, business lounges and frequent flyer incentives) are profit maximising by either improving efficiency or adding value to their services. They could be regarded as predatory, however, if the strategies also impose disproportionate costs on their rivals or exclude them from making similar value adding investments.

Assessment of Price Competition in 1991

Market conditions during 1991 were such that predatory pricing by Ansett and Australian could have been feasible. Their market shares, combined with natural and institutional entry barriers, gave them the necessary market power and both were in aposition to use their larger networks to cross subsidise losses on Compass routes from higher margins on non Compass routes. While this would have tended to reduce their short run profits, it might have been a rational strategy if it succeeded both in eliminating Compass and preventing any other entrant from taking its place.

Comparisons of revenue yield data for Compass and Australian (insufficient data was provided by A nsett) on a selection of Compass routes during 1991 show that Australian maintained yield figures significantly above those of Compass throughout the period, although both showed a declining trend. This would largely reflect Australian’s strategy of pursuing high yield traffic combined with the selective use of matching discounts to fill empty seats and Compass’ alternative strategy of providing a low yield one class discount fare service.

However, Compass’ revenue earnings did not make the same contribution to overheads per ASK as that of Australian on the routes where they completed. Thus, Compass’ low yield strategy achieved growth in passenger numbers and load factors at the cost of a much lower contribution to overheads than was achieved by Australian on those routes. This analysis tends to support the view that Australian was pursuing a pricing strategy to maximise the revenue contribution to overheads after covering direct operating costs.

This data does not refute the alternative hypothesis that, given Compass’ discount strategy and financial position, a strategy of matching its discounts would keep its contribution to overheads sufficiently low to force it into liquidation within a relatively short period. It is also relevant to note, that all three airlines were not covering their overheads in 1991 and that the level of price competition on Compass routes was not sustainable in the medium term. Nevertheless, the predatory purpose inherent in this alternative ‘hypothesis’ would be difficult to infer from this analysis, as the observed market conduct is also consistent with rational profit maximising behaviour in the prevailing market conditions of 1991.

Comparisons for Australian across a selection of its Compass and non Compass routes show that its yields were significantly higher on non Compass routes (byas much as 7 c per RPKin some months) than on Compass routes although there was a declining trend in yields on both types of route. However, Australian was actually making a comparable contribution to overheads after meeting direct operating costs on both Compass and non Compass routes. This would seem to indicate that operating costs as well as revenues were higher on Australian’s non Compass routes and that competition from Compass appears to have had little effect on the relative profitability of Australian’s services. The result is not consistent with cross subsidisation of Australian’s Compass routes by its non Compass routes.

Similar comparisons between A nsett and Compass were not possible with the limited data provided by A nsett. While the Commission was permitted to view (but not retain) Ansett accounting and operational data which appeared to be broadly consistent with the competitive strategies and performance shown by the Australian data for 1991, it is not in a position to express a view on the relative performance of Ansett in the marketplace during 1991. (This is perhaps an area of the TPC’s operations that need to be strengthened. Under the circumstances of open deregulation and the issue of possible predatory behaviour, it is essential that the public watchdog, ie; the TPC and PSA, have the legal right and the full backing of the Minister to access any relevant data tantamount to their investigations. Ed)

Based on this limited analysis, the Commission was unable to find indications of predatory pricing and (subject to further monitoring, information and analysis) has reached the preliminary view that the pricing behaviour of the airlines in 1991 appeared to be consistent with hard competition to minimise losses in depressed economic conditions with excess airline capacity. It is also noteworthy that Compass itself chose not to pursue any complaints of predatory pricing to the TPC when it was invited to do so on at least two separate occasions in 1991.

Assessment of Non Price Competition in 1991

Based on experience in the US, the areas of non price competition in the airline industry which would appear to have the potential to involve either cost raising or yield diminishing predatory conduct include:

• value added services for high yield passengers;

• ownership of CRS;

• ownership and control of travel agents; and

• access to terminal and airport facilities.

The latter could both add to costs through excessive charges and diminish yield by denying access to essential services.

To attract high yield passengers, airlines add value to their services, thus incurring costs that must be recovered from high fares. A nsett and Australian value added services include high service frequency, first and business classes, business lounges, valet parking, and frequent flyer clubs, which they claim are in demand by high yield passengers and are necessary to serve and retain that market. Compass did not seek to enter that market until late 1991 and rather than seeking to offer comparable services, offered discounts to the business market which actually lowered yields (eg 3 for 1 ticket sales).

Based on the limited data available for this study, the Commission found no indication that value adding strategies were being used in a predatory way. However, it will continue to monitor this area of competition between the airlines.

It has been suggested that ownership links between large travel agencies and the major airlines may bias the advice and business of the agent towards the relevant airline. During the course of this limited study, the Commission found no indication of systematic bias. However, to place that issue in perspective, it should be noted that commercial incentives were offered by all three airlines in 1991 in an effort to channel business in their favour. Australian offered commission overrides for sales above certain levels, A nsett offered free travel bonuses and Compass for atime offered commissions well above market rates.

Joint ownership of the Galileo CRS by Ansett and Australian (through Southern Cross Distribution Systems) had the potential to disadvantage Compass in its access to sales through travel agents who use Galileo and not the Sabre system used by Compass. Compass was unable to show last seat availability information on the Galileo system through Sabre because of technical problems in establishing the link and to some extent this would have had an adverse effect on its access to business through travel agents. The TPC recently authorised a proposal to combine the Fantasia and Galileo systems under one corporate structure on conditions which should ensure the provision of unbiased, non discriminatory information and the provision of last seat availability information to agents who participate in either system.

Compass claimed that lack of adequate terminal facilities under its sub lease with Australian had an adverse effect on its ability to compete in the market generally and in particular for the profitable business market. It also alleged that Australian did not satisfy its obligations under the terminal head leases to provide Compass with equivalent facilities and failed to do so for the purpose of damaging Compass and reducing competition.

The facilities provided to Compass were certainly less than ideal. It appears to have had inadequate baggage handling and passenger assembly facilities at some airports, Sydney in particular, and was required to operate from the international terminal in Perth and from substandard, temporary accommodation in Adelaide while renovations to Australian’s terminals were completed.

There may be reason to question whether Australian sought to overcharge Compass for the terminal facilities provided under sub leases as apossible cost raising strategy. The arbitrator’s decision to reduce the amount Australian charged Compass for terminal facilities would tend to support that view, although Australian is appealing that decision. The standard and efficiency of the ground services Compass could provide its passengers also appeared to suffer because it was not in control of its own terminal facilities.

The Commission has reached the preliminary view that the standard of terminal facilities available to Compass, and its lack of control over those facilities was a factor in its inability to compete effectively in the airline market. The Commission notes, however, that a lower standard of terminal facilities and services would have been more of an competing for the high yield business market. Compass did not seek to enter this market until late in 1991. Moreover, access to terminal facilities was not the sole, or even the most important, factor in the failure of Compass, but its ability to compete and to manage its yields more effectively could have been improved with access to better facilities.

The primary responsibility for such decisions must remain with the company directors in the light of their assessment of the company’s financial position and its market prospects. But agencies such as the Australian Securities Commission (ASC) and the Australian Stock Exchange (ASX) could consider the scope for improving their capacity to identify and monitor rapidly deteriorating situations in business. It has not been possible to examine that issue during this study and it should be addressed by other inquiries being conducted by the ASX and the Commission.

In the light of the Compass experience, there is also a need to examine any scope for policy options to improve the existing protection available to passengers in the deregulated airline market. Options could include improved travel insurance, or the use of trust accounts for such advance purchase payments. The effect of such options on the competitiveness and efficiency of the aviation market would need to be considered. The Federal Bureau of Consumer Affairs is currently preparing a paper on these and other options for consideration by the Government.

Conclusions

The Commission has reached the following general conclusions from its study of the failure of Compass Airlines: (a) Shortcomings in the entry strategy and management of Compass appear to have been the most important contributing factor to the failure of the airline. Importantly, Compass’ entry strategy misjudged the operational, economic and regulatory obstacles to large scale entry into the domestic market in 1990. It also misjudged the competitive advantages of the incumbent airlines, their responses to the additional capacity introduced by Compass and the capitalisation required for successful entry in these market conditions (b) Other contributing factors were the impact of the recession, the presence of excess capacity in the market, the competitive responses of Ansett and Australian and the problems experienced by Compass in gaining access to appropriate terminal facilities. (c) Because of their powerful market positions as the established airlines, Ansett and Australian have the capacity to place a new entrant airline at a relative disadvantage, particularly in relation to the high yield passenger market. It is quite a different question, however, whether their competitive conduct during 1991 constituted a breach of the Trade Practices Act. During this study the Commission has not found indications of predatory pricing within the meaning of the Act on the part of the incumbent airlines. The airlines’ ownership of CRSs, travel agent links, provision of business lounges and frequent flyer incentives are structural features of the market. While they may make it more difficult for smaller entrants to compete, no indications were found that they resulted in misuse of market power in terms of the Trade Practices Act.

(d) On the information available, there is reason to believe that Compass was disadvantaged to some extent in its ability to compete by the nature and terms of the access it was given to terminal facilities under its sub lease with Australian. This disadvantage was not as great as it might have been because Compass did not seek to compete for high yield traffic until late in 1991. (e) To the extent that Compass suffered some damage from inadequate access to, and control over, appropriate terminal facilities, the long term leases between the Government and the incumbent airlines were a contributing factor, particularlyin view of the lack of appropriate alternative terminal facilities for new entrant airlines at major airports. Access to appropriate terminal facilities is also likely to be a problem for prospective new entrants to the industry in the short term, pending the construction of sufficient additional terminal facilities to alleviate the problem.

(f) Given the relatively small size of the Australian aviation market and the considerable risks associated with larger scale entry by newly established airlines, competition, efficiency and service quality in the domestic aviation market could be enhanced by permitting greater entry to the domestic market by Qantas and Air New Zealand. The effectiveness of competition among airlines serving the domestic market could be enhanced by extending the market to include trans-Tasman and New Zealand services in effect, giving Qantas, Air New Zealand, Ansett and Australian access to the entire Australasian market. This proposal has already been approved though still needs to be the subject of an agreement between the two governments before it can be given effect. (g) The failure of Compass caused substantial losses for investors, creditors and passengers and is also the subject of investigations under the Trade Practices Act and the Corporations Act. This experience highlights the need to develop practical options for minimising the costs imposed on consumers by such failures.

Questions can also be raised in hindsight 4bout the strategy of operating a one class discount service, the decision to operate large wide bodied jets and the lateness of the unsuccessful attempt to access the high yield passenger market.

These factors combined to force Compass to cease operations on December 20 1991. In hindsight it is now evident that the company was under capitalised from the start for the entry strategy it adopted. □

Readers are reminded that we carried a seven page analysis of the Compass collapse in our March 1992 edition.