Proposed combination of operations of P and O Princess Cruises plc and Royal Caribbean Cruises Ltd

No. ME/1652/01

Report under section 125(4) of the Fair Trading Act 1973 of the Director General's advice, dated 23 January 2002, to the Secretary of State for Trade and Industry under section 76 of the Act
 

The parties

P&O Princess Cruises plc (POPC) is a company listed on the London Stock Exchange. POPC was formed in 2000 by a demerger from the Peninsular and Oriental Steam Navigation Company.

POPC is concerned with the sale and operation of cruise holidays. Within the UK, POPC operates the P&O Cruises and Swan Hellenic cruise holiday businesses, with four and one ships respectively.

POPC also operates Princess Cruises, a cruise holiday business based in, and selling mainly to the USA. This operates ten ships with facilities, entertainment and crew aimed at the North American market. Finally, POPC operates Seetours in Germany with two ships under the AIDA and A'ROSA brands. These cater for German-speaking passengers.

For the financial year to 31 December 2000, POPC's worldwide turnover was £1,603m, of which £336.3m was in the UK. The value of its gross assets was £3,047.8m.

Royal Caribbean Cruises Limited (Royal) is a company incorporated in Liberia and listed on the NYSE. It has two major shareholders:

• Cruise Associates, a Bahamian general partnership, owns 25.1% of the shares. Cruise Associates is owned by trusts whose beneficiaries are the Pritzker family (Chicago) and the Ofer family.
• A. Wilhelmsen AS. a Norwegian corporation owned by the Wilhemsen family, has a shareholding of 24.1%.

Royal is a cruise holiday company which operates two principal businesses, Royal Caribbean International and Celebrity Cruises, which own 14 and 7 ships respectively. Both businesses are geared primarily to the North American market and carry a high proportion of North American passengers.

Royal has invested $300m in convertible preferred stock in First Choice Holidays. Under the articles of association, these carry normal voting rights, as if they had been converted, and account for approximately 17% of the voting rights in First Choice's shares. Royal has exercised an associated option to nominate a non-executive director to the Board, and has various other rights. Royal is therefore likely to have material influence over the commercial policy of First Choice and this has been taken into account in the assessment.

Royal recently entered into a 50/50 joint venture with First Choice Holidays called Island Cruises. This will operate one ship, to be transferred from Royal's fleet, aimed at the British market. Island Cruises will commence operations in 2002.

In the financial year to 31 December 2000, Royal had a worldwide turnover of £1,895m of which (see note 1) was in the UK. The value of its gross assets was £5,176.7m.

The transaction

The parties propose to effect a dual-listed company structure whereby there will be common control by a unified executive over the assets of both companies, such that the parties will become a single economic entity. Following the combination, POPC shareholders will have economic ownership of 50.7% and Royal shareholders the remaining 49.3%.

Jurisdiction

The transaction qualifies for investigation under the assets test in section 64(1)(b) of the Act. The ECMR does not apply.

Background

This paragraph has been excised at the request of the parties.

Assessment

Product Markets

The parties overlap in the supply of cruise holidays to various destinations around the world.

There have been no recent competition cases in the UK in which it has been necessary to consider the markets for cruise holidays. In Costa Corciere/ Chargeurs/ Accor (see note 2), the EC found the key features that attracted people to cruises were: the "holiday concept", service quality, destination/route and price. However, the decision of which cruise to purchase is complex (see note 3) and research implies that most potential customers need assistance (usually from a travel agent) in distinguishing between the cruises on offer. Customers themselves are not a homogeneous group and different characteristics will appeal to different groups. Thus some cruises are tailored towards retired people, some are "children free", some focus on on-board activities, while others focus on excursions.

There may be also a degree of segmentation according to the quality of the ship and the facilities offered. Some ships are more luxurious and offer more amenities than others and command a premium price as a result. An analogy might be made with the star rating system used for hotels. Set against this is the observation that each ship typically offers a range of accommodation packages. Finally, it is possible to consider that particular destinations may be distinct e.g. a cruise to Alaska may be regarded by consumers as a distinct product from a Mediterranean cruise. Arguably, then, there are sub-markets for different types of cruise, but for the purposes of this advice I have generally considered the market for cruise holidays as a whole, though noting that there is product differentiation within it.

Market research reveals that there is significant variety in cruise customers. There is a high proportion of repeat customers, for whom cruises may have no near substitutes, but it is possible that some other customers would regard package holidays as acceptable alternatives. It is unclear whether this proportion is sufficiently large as to constrain prices. Thus, while on a very broad view the market might be considered to be all overseas holidays, which cruises account for around 4%, the most appropriate frame of reference for judging the competitive effects of the merger is probably cruise holidays as a whole.

Geographic markets

Around 85% of all cruise sales for UK passengers are made through travel agents and thus the point of sale is within the UK. Cruise ships are usually dedicated to a particular nationality, with crew members, cuisine and entertainment designed for particular languages and cultural tastes. As a result, hardly any customers go on cruises marketed to other nationalities. The market is therefore considered likely to concern cruise holidays designed for and sold within the UK. In the EC case cited earlier, national markets were also identified because of differences in national consumer preferences and the national supply chain (see note 4).

Given the focus of sales, distinctive national characteristics and the fact that there does not appear to be significant cross-over between passengers of different nationalities, the relevant geographic market is taken to be the UK.

Horizontal issues

Pricing Behaviour

Cruise operators face very high fixed costs and relatively low variable costs and thus face significant commercial pressure to ensure full capacity utilisation. Retail prices are given in the annual publication of the tour brochure (see note 5). Discounting for early bookers is common, and is regarded by operators as part of their yield management system. First time cruise customers may be offered significant discounts to persuade them that a cruise is a "dream holiday" since a high percentage (68%) of all cruise passengers in 2000 were repeat business.

Mixed evidence has been received on the nature of pricing behaviour. P&O's strong position in the UK market might enable it to act as the price leader, with other companies following its prices. Price comparisons are very difficult, however, as no two products are identical and firms adopt different pricing policies during the year, with some seeking to extract higher premiums than others during the key holiday periods. The parties have provided evidence suggesting their prices show only a small premium when compared with other operators on a like-for-like basis.

An alternative approach to pricing issues has been a consideration of return on capital employed (ROCE). ROCE figures have indicated that the parties, together with Carnival, have achieved good or high returns in recent years. The parties claim that these are justified as an appropriate return on a high risk, capital intensive business involving discretionary consumer expenditure. A complainant has offered an alternative view that the 'high' return represents the ability to charge a brand premium.

Market Shares

Information on market shares (Table One) reveals the level of concentration in this market: the five firm concentration ratio is 65%, showing that the top five firms (all well known brands) account for most of the market. POPC is the clear market leader with 31% of passenger cruise days. This may be taken as a proxy for share of revenue if POPC's price per day is the same as that of competitors, but will be an underestimate to the extent that its prices are higher than its competitors. The merger would allow POPC to increase its already substantial share by a further 9% (of which First Choice accounts for 3%) and might thereby increase the potential for it to act as a price leader.

Table One: Market Shares

view table

Source: The parties and OFT calculation

Notes: Figures may not sum due to rounding. Royal is considered to have material influence over First Choice, hence its market share is included in the calculation of the parties' share calculation. The HHI calculation includes Fred Olsen, Star Cruises, Paradise Cruise Lines and Louis Cruise Lines.

40% is a significant market share, which requires careful consideration. Likewise the HHI calculation points to a significant increase in concentration in what is already a moderately concentrated market. Market shares might be higher if the market were defined by reference to a narrow 'premium' segment or if a distinct market were defined for Caribbean Cruises.

Vertical issues

Royal has a minority shareholding and a directorship in First Choice, which in our view gives it material influence over this travel agent, one of the big four in the UK. The merger would give POPC enhanced access to this distribution network. There is a possibility that the combined entity might be able to use its influence over First Choice to encourage First Choice to favour its offerings at the expense of others, weakening somewhat their access to customers.

There is a wider possibility that the enhanced bargaining power of the merged entity in relation to travel agents generally might lead to rivals being disadvantaged in the crucial travel agent distribution channel. I do not know how realistic this possibility is.

Barriers to entry

There are a number of possible barriers to entry. The cost of building a new ship, and the time required to make a return on the investment are significant but that does not prevent entry from well-capitalised firms. The parties estimated that a new modern 1,500 to 2,000 passenger ship can cost £150 million to £300 million to build and takes from two to three years from contract to delivery. An interested party has estimated a longer period, however, of up to five years. Lower cost entry may be possible through conversion of an existing passenger ship, or by chartering all or part of a cruise ship from another operator. It has not been possible to evaluate the extent of spare capacity represented by these options in recent years, although there appears to be significant capacity (arising from the failure of Renaissance) at the present time. It is conceivable that US operators could reconfigure existing ships to target the UK market. Anecdotal evidence suggests that the US market is more profitable, however, such that conversion might not be a likely strategy.

Market research suggests that consumers regard brands as important. POPC spent (see note 6) £million on UK brochure advertising alone whilst Royal spent around $million in 2001. A new entrant to the holiday business would face higher marketing costs to establish brand awareness amongst UK customers. Established players also have important incumbency benefits. POPC and Royal respectively claim (see note 6) of their customers are return customers. Each has a database of several million to target in mailshots. Thus reputational barriers to entry may be of some significance.

In recent years UK tour operators such as Airtours and Thomson have been able to use their brand name within the travel industry successfully to enter the cruise market. They entered the UK market in 1995 and 1996 respectively and are now number two and three in the UK. The major players made good profits in the period from 1995 to1999 and there is reason to believe this has encouraged rivals to commission new ships and expand their operations. The EC found that barriers to entry were not high, at least for well-established tour operators with an established distribution network and sufficient financial resources. Conversely, it would appear to be more difficult for companies without links to travel agents to enter the market in a significant way. The other operators such as Louis Cruises and Festival Cruises who have entered the market in recent years and built up their brand names have largely done so as niche players.

Cruise operators need access to ports – early planning allows slots to be booked and may allow for preferential treatment against new entrants. Weekend slots, which are already owned by incumbents, are significantly more attractive to cash-rich, time-poor customers. There is potential for such access to acquire de facto "grandfather rights". There would appear to be limited numbers of "slots" affecting the scope for new entry at Bermuda and at Glacier Bay, the latter of which is regarded as an important stopping point for Alaskan cruises. In the event of a schedule conflict, some ports grant priority to historical users on their historical day of the week where they wish to continue to do so. Exceptionally, a cruise operator will have made investments in port facilities or a longer term usage commitment, which entitles it to a degree of priority. These factors can make it harder for a new entrant to compete effectively. It has not been possible to quantify the significance of these to the cruise market overall, however, and it seems probable that competitors could normally offer cruises which bypass these local problems.

These possible barriers to entry may be individually surmountable. There has been mixed evidence on whether their combination might hinder or deter new entry, especially where companies lack distribution networks.

Buyer power

The majority of cruise sales are made through travel agents, 70-100% (see note 7) for POPC and Royal respectively. The big four UK travel agents are likely to have some degree of buyer power, although this would be offset to some extent insofar as the merged group's cruises would be must-carry products.

Other Issues

This paragraph has been excised at the request of the parties.

Third party views

There was a mixed response from third parties. One was very concerned. Others were relatively unconcerned. The views of third parties have been reflected in the assessment above.

Undertakings in lieu of reference

This merger raises a number of issues which it has not been possible to determine during a first stage investigation. In these circumstances it is not possible to identify any undertakings that could be recommended to address the competition issues raised.

Conclusion

There is a range of cruise holidays available from the industry, the parties, and indeed within an individual ship. As a result, it has not been possible to obtain definitive market shares on a revenue basis, although passenger cruise day figures appear to be a reasonable proxy. The merger would result in a combined entity with about 40% of the market (including First Choice, over which Royal appears to have material influence).

There are some possible barriers to entry. Each may be surmountable by a well-financed company, but their combination might discourage entry from companies which lack adequate distribution networks. Conclusions have therefore not been reached on whether the scope for entry could constrain any enhanced potential to exercise market power.

In sum, the OFT's preliminary examination has left sufficient concerns about the possible effect of the merger on competition for me to recommend that you should refer this transaction to the Competition Commission for further investigation.

NOTES

1. This figure has been excised at the request of the parties.

2. Case No IV/M.0334 - Costa Crociere/Chargeurs/Accor

3. Material has been excised in the following text at the request of the parties.

4. Reference as for note 2 above, para 19.

5. Material has been excised in the following text at the request of the parties.

6. Figures have been excised in this paragraph at the request of the parties.

7. A range has been used at the request of the parties.

Back to top